Latitudinal gradients in diversity: real patterns and random models

Mid-domain models have been argued to provide a default explanation for the best known spatial pattern in biodiversity, namely the latitudinal gradient in species richness. These models assume no environmental gradients, but merely a random latitudinal association between the size and placement of the geographic ranges of species. A mid-domain peak in richness is generated because when the latitudinal extents of species in a given taxonomic group are bounded to north and south, perhaps by a physical constraint such as a continental edge or perhaps by a climatic constraint such as a critical temperature or precipitation threshold, then the number of ways in which ranges can be distributed changes systematically between the bounds. In addition, such models make predictions about latitudinal variation in the latitudinal extents of the distributions of species, and in beta diversity (the spatial turnover in species identities). Here we test how well five mid-domain models predict observed latitudinal patterns of species richness, latitudinal extent and beta diversity in two groups of birds, parrots and woodpeckers, across the New World. Whilst both groups exhibit clear gradients in richness and beta diversity and the general trend in species richness is acceptably predicted (but not accurately, unless substantial empirical information is assumed), the fit of these models is uniformly poor for beta diversity and latitudinal range extent. This suggests either that, at least for these data, as presently formulated mid-domain models are too simplistic, or that in practice the mid-domain effect is not significant in determining geographical variation in diversity.