Declining willow ptarmigan populations: The role of habitat structure and community dynamics

Abstract

The recent range contractions and population declines of many grouse species worldwide have been attributed to loss and fragmentation of their habitats, although the empirical evidence for the actual drivers is often weak. In case of the willow ptarmigan Lagopus lagopus in Fennoscandia, ungulate overbrowsing of willows has been hypothesized to exert such negative habitat-related impacts. Moreover, a steep population decline of willow ptarmigan in southern Fennoscandia has recently been attributed to community interactions linking the fate of the willow ptarmigan to a change in keystone tundra rodent populations. Community and habitat factors may also interact in their impact on willow ptarmigan abundance. Here we assess whether willow thicket structural characteristics sensitive to ungulate browsing impacted willow ptarmigan habitat occupancy and whether such impacts depended on small rodent population dynamics. We employed an extensive survey approach spatially encompassing three riparian tundra regions and covering the phases of increase, peak and crash of the small rodent cycle. Willow ptarmigan habitat occupancy increased with the areal extent of willow thickets, whereas it decreased with increasing degree of thicket fragmentation (i.e. habitat shredding). Both of these effects were consistent with ungulate over-browsing impacting willow ptarmigan abundance negatively. Over the 4-year study period, willow ptarmigan habitat occupancy declined steeply independently of spatial variation in willow thicket areal extent and fragmentation. Moreover, the expected increase in ptarmigan populations during the increase/peak phase of the rodent cycle was not observed. Thus although our study provides support for the hypothesis that intense ungulate browsing negatively impacts willow ptarmigan, our study also suggests that causes of the current steep decline of ptarmigan populations in northern Fennoscandia should be sought in factors other than habitat fragmentation and changed rodent population dynamics.