Predation risk impacts spatial patterns of small mammal seed caching

Background/Question/Methods

Predators influence both the abundance and behavior of their prey, which can have cascading effects on plant community structure and composition. Canid predator dynamics in northern Wisconsin provide a natural experiment to understand the effects of predation on small mammal communities, since recolonizing wolves competitively exclude coyotes but not foxes (small mammal specialists). It remains unclear if apex predator recolonization affects spatial patterns of small mammal seed caching, which could subsequently affect seed dispersal. We investigated two hypotheses: 1) wolf recolonization changes the abundance and identity of potential seed dispersers (numerical response), and 2) wolf recolonization alters spatial patterns of small mammal seed caching (behavioral response). Additionally, we investigated whether small mammal behavioral responses are moderated by availability of refuge and whether the response to refuge is a function of wolves. We tested the numerical response by live-trapping small mammals at nine sites that vary in wolf and shrub layer abundance. We tested the behavioral response by providing 200 Quercus rubra seeds to small mammals at these sites. Seeds were labeled with 14mm nails, allowing seed remnants to be recovered with a metal detector. We measured the distance between point of recovery and the initial seed source (“removal distance”).

Results/Conclusions

We found no evidence that wolf recolonization suppresses small mammal abundance or alters small mammal community composition. However, we did find evidence of changes in small mammal seed caching behavior. We recovered approximately 5% of seeds (n=98), most of which were consumed (n=93).We observed a marginally significant negative relationship between wolf abundance and median removal distance. Shrub abundance had no effect on median removal distance, but areas with more shrubs had less variable removal distances (lower coefficient of variation). The interaction between wolf abundance and shrub cover and no effect on dispersal distance.  We conclude that apex predators may have a subtle effect on small mammal seed caching behavior by restricting spatial dispersion of seed caches. But, apex predators do not alter the potential pool of seed-caching small mammals. Wolf recolonization may therefore indirectly harm some plant species cached by small mammals, since surviving seeds remain closer to conspecifics where they are more likely to experience density-dependent mortality.