Mobbing behavior of songbirds in response to calls of an ambush-predator, the Northern Pygmy-Owl

Abstract

Mobbing is a widespread but risky behavior where prey actually approach and may even attack their predators. Small songbirds gather in multi-species flocks to direct mobbing at birds of prey such as the Northern Pygmy-Owl (Glaucidium gnoma), a small diurnal species of owl in mountainous western North America. We hypothesized that intensity of mobbing behavior would vary according to risk to songbirds and that risk varied seasonally and across elevations. In particular, we expected mobbing intensity to change seasonally as a function of two primary sources of risk: seasonal dietary changes of pygmy-owls and distribution of pygmy-owls as a function of elevation. Pygmy-owls are known to shift their diet from mostly mammals in early spring to birds in early summer, so we hypothesized risk of predation would vary in parallel. Likewise, risk was expected to vary as the distribution of pygmy-owls was thought to vary seasonally across elevations. To elicit mobbing in a controlled manner, we conducted owl call playback experiments in the Oregon Coast Range mountains, United States. Overall, we found mobbing to be rare, observing it in 8.1% of 663 experiments. The frequency of mobbing did not appear to vary across elevations. Instead, mobbing varied mostly by season. The frequency of mobbing was highest during fall at lower elevations. Mobbing was rarest (1.3% of trials) during winter. Mobbing occurred during seasons when pygmy-owls tend to consume more birds, consistent with the hypothesis that seasonal variation in risk influences mobbing. In our study, mobbing happened when risk was higher. However, the increased risk may be offset by the influence of mobbing group size. Mobbing was more likely to occur at sites where the number of songbirds, especially chickadees (Poecile sp.) and nuthatches (Sitta sp.), was larger. Yet, even when songbird numbers were high, mobbing was rare during winter, suggesting an interaction between risk and energetic costs of mobbing. Additional experiments manipulating food resource availability may be able to reveal the strength and influence of the interaction on mobbing decisions.