Scaling population density and spatial pattern for terrestrial, mammalian carnivores

A large part of ecological theory has been developed with the assumption that intra- and inter-specific patterns of density and spatial distribution can be consistently and reliably compared, and that these patterns have represented populations across nonstudied landscapes. These assumptions are erroneous. We found that log(10) population density estimates consistently decreased linearly with log(10) spatial extent of study areas for species of terrestrial Carnivora. The size of the study area accounted for most of the variation in population estimates, and study areas increased with the female body mass of the study species. But study sites consistently had higher densities than can be expected for nonstudy sites, regardless of the size of the study area, because study sites are typically chosen based on a priori knowledge of high density. Inter-specific comparisons of density and distribution might provide more insight into community organization after intra-specific density estimates have been scaled by the study areas, and related to the nonstudied landscapes within each species' geographic range.