Species-area relationships and extinction forecasts

Abstract

The species-area relationship (SAR) predicts that smaller areas contain fewer species. This is the basis of the SAR method that has been used to forecast large numbers of species committed to extinction every year due to deforestation. The method has a number of issues that must be handled with care to avoid error. These include the functional form of the SAR, the choice of equation parameters, the sampling procedure used, extinction debt, and forest regeneration. Concerns about the accuracy of the SAR technique often cite errors not much larger than the natural scatter of the SAR itself. Such errors do not undermine the credibility of forecasts predicting large numbers of extinctions, although they may be a serious obstacle in other SAR applications. Very large errors can arise from misinterpretation of extinction debt, inappropriate functional form, and ignoring forest regeneration. Major challenges remain to understand better the relationship between sampling protocol and the functional form of SARs and the dynamics of relaxation, especially in continental areas, and to widen the testing of extinction forecasts.