SPATIAL HETEROGENEITY EXPLAINS THE SCALE DEPENDENCE OF THE NATIVE–EXOTIC DIVERSITY RELATIONSHIP

While small-scale studies show that more diverse native communities are less invasible by exotics, studies at large spatial scales often find positive correlations between native and exotic diversity. This large-scale pattern is thought to arise because landscapes with favorable conditions for native species also have favorable conditions for exotic species. From theory, we proposed an alternative hypothesis: the positive relationship at large scales is driven by spatial heterogeneity in species composition, which is driven by spatial heterogeneity in the environment. Landscapes with more spatial heterogeneity in the environment can sustain more native and more exotic species, leading to a positive correlation of native and exotic diversity at large scales. In a nested data set for grassland plants, we detected negative relationships between native and exotic diversity at small spatial scales and positive relationships at large spatial scales. Supporting our hypothesis, the positive relationships between native and exotic diversity at large scales were driven by positive relationships between native and exotic beta diversity. Further, both native and exotic diversity were positively correlated with spatial heterogeneity in abiotic conditions (variance of soil depth, soil nitrogen, and aspect) but were uncorrelated with average abiotic conditions, supporting the spatial-heterogeneity hypothesis but not the favorable-conditions hypothesis.