Disparity between range map‐ and survey‐based analyses of species richness: patterns, processes and implications

Abstract Species richness patterns are characterized either by overlaying species range maps or by compiling geographically extensive survey data for multiple local communities. Although, these two approaches are clearly related, they need not produce identical richness patterns because species do not occur everywhere in their geographical range. Using North American breeding birds, we present the first continent‐wide comparison of survey and range map data. On average, bird species were detected on 40.5% of the surveys within their range. As a result of this range porosity, the geographical richness patterns differed markedly, with the greatest disparity in arid regions and at higher elevations. Environmental productivity was a stronger predictor of survey richness, while elevational heterogeneity was more important in determining range map richness. In addition, range map richness exhibited greater spatial autocorrelation and lower estimates of spatial turnover in species composition. Our results highlight the fact that range map richness represents species coexistence at a much coarser scale than survey data, and demonstrate that the conclusions drawn from species richness studies may depend on the data type used for analyses.