Predicting local space use by Peromyscus leucopus

Background/Question/Methods   The white-footed mouse (Peromyscus leucopus), a very common small mammal in northeastern deciduous forests, is involved in a diverse array of ecological interactions. It preys on the invasive gypsy moth (Lymantria dispar) and eggs of ground nesting songbirds, is an important prey item for various terrestrial and avian predators, is a highly competent reservoir for the lyme disease bacterium and host for the black-legged tick. Spatial distributions of P. leucopus may significantly affect their ecological interactions, so investigating factors driving this heterogeneity is crucial. We hypothesized that P. leucopus space use would be determined by three major categories of variables: food availability, cover to avoid predation, and spatial structure due to mouse interactions (territory defense and dispersal). We measured fine scale (<30 m by 30 m) mouse activity from May to September at the Cary Institute of Ecosystem Studies in Millbrook, NY, using tracks on track plates to quantify space use. Sampling took place on six 5.3 ha sites consisting of 12-16 regularly spaced 30 m by 30 m “spots” containing four sampling points arranged in a 15 m by 15 m square, centered in the spot. Three track plates were placed at each point, pairs of plots were sampled over two-week intervals, and track plates were checked three times per week for a total of over 32,000 trackplate-nights. We sampled invertebrates, seeds, and blueberries (the major components of diet in the area) to quantify forage biomass available for mice. We also estimated understory cover (woody debris, runways, bare rock, etc).

Results/Conclusions   Proportions of plates tracked showed considerable variation from point to point. Substantial variation was seen in forage biomass and aspects of cover as well, suggesting the possibility for mice to exploit heterogeneous microhabitat characteristics. However, generalized linear models assessing the predictive value of diet, cover, and social structure for determining local space use by P. leucopus indicate that the most important predictor is conspecific activity in the surrounding sample points, with aspects of both diet and cover being relatively unimportant (P>0.05). This suggests that mouse interactions are the primary cause of spatial heterogeneity in small-scale space use, and that, at our sites, predicting local space use by P. leucopus using habitat characteristics may not be reasonable.