• California spotted owl;
  • discrete-choice model;
  • foraging;
  • managed forests;
  • radiotelemetry;
  • resource selection function;
  • Strix occidentalis occidentalis.

Abstract: We linked radiotelemetry data from California spotted owls (Strix occidentalis occidentalis) with forest inventory data from mixed coniferous forests managed primarily via partial timber harvest practices. We estimated a discrete-choice resource selection function (RSF) based upon 21 choice sets from the forest inventories and nocturnal telemetry locations of radiotagged adult spotted owls occupying 17 home ranges. Nocturnal foraging was strongly associated with forests close to nests and small streams. The combined basal areas of Douglas-fir (Pseudotsuga menziesii), white fir (Abies concolor), and red fir (Abies magnifica) and basal area of hardwoods ≥20 cm diameter at breast height were positively and unimodally correlated to foraging habitat selection by owls, whereas the relative probability of selection decreased with increasing basal area of ponderosa pine (Pinus ponderosa). Opportunistically collected diurnal data indicated that owls roosted in forest stands that contained greater tree densities than those used for foraging. Topographic position, habitat heterogeneity, tree species composition, and forest density also influenced foraging site selection. Because our results indicated that forests can be too dense as well as too open, the study suggests that judiciously applied silvicultural prescriptions may maintain or improve owl foraging habitat. By linking the RSF with forest-inventory data and forest-growth models specific to the region of our study, forest managers can forecast potential consequences of silvicultural options on spotted owl foraging habitat at the level of individuals or that of a population.