Ability of Current Phylogenetic Clustering to Detect Speciation History

Abstract

Phylogenetic diversity aims to quantify the evolutionary relatedness among the species comprising a community, using the phylogenetic tree as the metric of the evolutionary relationships. Could these measures unveil the evolutionary history of an area? For example, in a speciation hotspot (biodiversity cradle), we intuitively expect that the species in the community will be more phylogenetically clustered than randomly expected. Here, using a theoretical simulation model, we estimate the ability of phylogenetic metrics of current diversity to detect speciation history. We found that, in the absence of dispersal, if the incipient species do not coexist in the region of speciation (as expected under allopatric speciation), there was no clear phylogenetic clustering and phylogenetic diversity failed to detect speciation history. But if the incipient species coexisted (sympatric speciation), metrics such as standardized effect size of Faith’s Phylogenetic Diversity (PD) and of Mean Nearest Taxon Distance (MNTD) were able to identify areas of high speciation, while Mean Pairwise Distance (MPD) was a poor indicator. PD systematically outperformed MNTD. Dispersal was a game-changer. It allowed species to expand their range, colonize areas, and led to the coexistence of the incipient species originating from a common ancestor. If speciation gradient was spatially contiguous, dispersal strengthened the associations between phylogenetic clustering and speciation history. In the case of spatially random speciation, dispersal blurred the signal with phylogenetic clustering occurring in areas of low or no speciation. Our results imply that phylogenetic clustering is an indicator of speciation history only under certain conditions.