Optimally Foraging Deer Mice in Prairie Mosaics: A Test of Habitat Theory and Absence of Landscape Effects

The foraging behavior of individuals, as well as the patterns of abundance that it helps create, is likely to depend on habitat and its associated landscape context. I tested this idea by measuring deer-mouse abundance and quitting-harvest rates (giving-up densities of resources in artificial foraging patches) along transects crossing badland boundaries between small and very large patches of prairie habitat in western Canada. Giving-up densities declined with increased population density among replicated census transects, they were independent of density at the foraging scale spanning habitats within transects. Giving-up densities were higher in prairie than in adjacent badland, deer mouse densities were opposite. Population density in the badland declined toward the border with prairie, presumably because individual foraging ranges near the boundary include an increasing proportion of the less rewarding prairie habitat. The pattern in density is consistent with the assumptions that habitat selection depends on the optimal behavior of individual foragers, and that it tends to equalize expected fitness between habitats. Predation risks, assessed from foraging in sheltered vs unsheltered patches, appeared higher in prairie than in badland. Landscape had no influence on predation risk, on patterns of foraging behavior, or on population density across habitat boundaries. Whereas landscape no doubt influences regional population size, and thereby influences probabilities of extinction and colonization, it may play an insignificant role in determining patterns of abundance across boundaries at the scale where optimally foraging individuals attempt to maximize their fitness.