Generalist predator, cyclic voles and cavity nests: testing the alternative prey hypothesis.

The alternative prey hypothesis (APH) states that when the density of the main prey declines, generalist predators switch to alternative prey and vice versa, meaning that predation pressure on the alternative prey should be negatively correlated with the density of the main prey. We tested the APH in a system comprising one generalist predator (pine marten, Martes martes), cyclic main prey (microtine voles, Microtus agrestis and Myodes glareolus) and alternative prey (cavity nests of common goldeneye, Bucephala clangula); pine marten is an important predator of both voles and common goldeneye nests. Specifically, we studied whether annual predation rate of real common goldeneye nests and experimental nests is negatively associated with fluctuation in the density of voles in four study areas in southern Finland in 2000-2011. Both vole density and nest predation rate varied considerably between years in all study areas. However, we did not find support for the hypothesis that vole dynamics indirectly affects predation rate of cavity nests in the way predicted by the APH. On the contrary, the probability of predation increased with vole spring abundance for both real and experimental nests. Furthermore, a crash in vole abundance from previous autumn to spring did not increase the probability of predation of real nests, although it increased that of experimental nests. We suggest that learned predation by pine marten individuals, coupled with efficient search image for cavities, overrides possible indirect positive effects of high vole density on the alternative prey in our study system.