Understanding the effects of fragmentation on species diversity is an enduring problem in ecology and conservation biology, epitomized by the SLOSS and fragmentation per se debates. Contributing to these controversies is the lack of a statistical expectation for the effects of habitat sub-division on the number of species a given area of habitat will contain. Species-area relationships derived from sampling theory describe the scaling of species richness with area under varying patterns of aggregation and species abundance distributions but have previously been unable to predict the number of species shared among patches, limiting their application in the habitat sub-division debates. Here, we use sampling theory to derive an expectation for the number of species shared in multiple samples (zeta diversity), allowing the expected effects of sub-division on diversity to be determined analytically. Under random placement of individuals, the model is parameter-free. For non-random spatial distributions, it can be modelled at the cost of a single fitting constant, which scales with sampling area according to a power function.
Results/Conclusions
We validate model predictive performance in sub-divided landscapes simulated by sampling from empirical (stem-mapped forest plots) and simulated data under random and non-random placement of individuals. Models accurately predicted the expected number of shared species, gamma diversity (combined number of species sampled) and the number of species confined to a single patch (i.e., single patch endemics). Only under random placement of species does sub-division have no effect on species richness. Under any amount of intraspecific aggregation, the number of species in a single contiguous patch is less than that found in any sub-division of the area. With this expectation for species richness in sub-divided habitat established, we suggest there should be no surprise in finding this in empirical data. Thus, the models partially explain long-standing debate over the effects of sub-division on species diversity, while providing a means to explore the separate effects of habitat loss and sub-division on diversity.