Influence of the Matrix on Species Richness in Ugandan Forest Fragments

The power-law species area relation (SAR) has been used to predict that as the size of a habitat fragment decreases, the number of species supported by that fragment will fall. Despite the widespread usage of the SAR in conservation biology, the relation ship is confounded by a number of factors, including habitat heterogeneity, unequal sampling effort, and slow dynamics ('extinction debt'). If the SAR is to be used as a predictive tool, it needs to be demonstrated that there is a relationship between spa tial extent and species richness independent of history, habitat heterogeneity and sampling effort. We will use the SAR to explore diversity patterns in species-rich Ugandan forest fragments, standardizing for habitat heterogeneity, sampling effort, and history. These forests are not entirely isolated from one another: dispersal between fragments links the individual forests into a metacommunity. Dispersal depends on the properties of the intervening area, the matrix, and we predict that where the matrix is conducive for dispersal, species richness will be higher than in regions where the matrix is not conducive for dispersal. There are three types of matrix around the moist broadleaf forest fragments in Uganda: subsistence agroforestry; plantation agr iculture; and, for forest fragments on islands in Lake Victoria, water. We predict that the structurally-diverse agroforestry matrix will be more conducive for dispersal than the structurally-simple plantation agriculture or water, and that this differenc e will be detectable in the species richness of different forests. If the SAR analyses show that properties of the matrix are important for biodiversity of fragments, then conservation management policy will need to consider both the protection of fores ts, and management of the matrix.