A hidden species-area curve

The species-area curve, which describes the relationship between the number of observed species in a geographical region and the area of the region, plays a central role in biogeography. Beyond its scientific importance, the species-area curve is commonly used to assess the loss of species due to habitat loss. When the species-area curve is estimated from spatial samples, the existence of species with low or highly spatially variable abundance exaggerates the true rate at which species accumulate with area. Here, a hidden species-area curve is defined that accounts for this sampling effect and its estimation by maximum likelihood is outlined. Both the species-area curve and the hidden species-area curve are conditioned on the observed species list; thus the analysis does not depend on the total number of observed and unobserved species, that is, species richness. The method is tested by sub-sampling some tropical forest data and found to work well. It is also applied to a classic data set from the deep sea. For these data, accounting for this sampling effect has a large impact.