RESUMEN
AbstractSpecies-area relationships (SARs) describe how the number of species increases with the size of the area surveyed. They usually take the form of a power law on regional spatial scales. A metareview of empirical data has shown that the exponent of the power law is larger on average when the areas are sampled in a nested manner, compared with sampling of isolated areas such as islands or nature reserves of different sizes. Because this finding contradicts ecological reasoning, we performed computer simulations of three qualitatively different models that generate species distributions in space and time driven by the mechanisms of speciation, dispersal, and extinction. We find that in all cases and over a wide parameter range the SARs obtained by nested sampling have a smaller slope on the regional scale than those obtained by independent sampling. However, the slopes change considerably with spatial scale and in a different manner for the two sampling methods. By evaluating the spatial scales on which the two types of empirical data were taken, we can explain the discrepancy between simulation results and empirical data.