Global patterns of phylogenetic beta-diversity components in angiosperms

Questions: Geographic gradients of beta-diversity help understanding the relationship between species and their environment. However, on a global scale, such patterns are only known for a few taxa, mainly terrestrial vertebrates, especially when considering the phylogenetic dimension. Here, we present the first global analysis of phylogenetic beta-diversity (PBD) for angiosperms. We aim to disentangle the relative contribution of PBD components (turnover- and nestedness-resultant differences) and the deviation of PBD given the taxonomic beta-diversity (TBD) along environmental and geographic gradients. Location: Global. Methods: We compiled range maps of 207,146 angiosperm species at 1 degrees cells and calculated PBD for assemblages formed by each focal cell and its neighboring cells in radii of 1.5 degrees and 2 degrees ("moving-window" approach). PBD was decomposed into turnover- and nestedness-resultant components, evaluating their relative importance as the proportion of nestedness-resultant PBD to the total PBD (PBDratio). To evaluate lineage exchanges, we calculated the deviation of PBD from TBD (PBDdev). We assessed the breakpoints of relationship between PBD and geographic (latitude and elevation) and environmental (temperature and precipitation) gradients using linear piecewise regressions. Results: The turnover-resultant component was predominant in shaping the global angiosperm PBD pattern. PBDratio was positively correlated with temperature, having a breakpoint around 14 degrees. Because PBDdev was mostly positive, TBD prevailed over PBD; PBDdev was correlated significantly with latitude and temperature gradients, being higher (i.e., low lineage replacement compared with species replacement) at latitudes above 50 degrees N, and in colder climates (below 2 degrees C). Conclusions: We provided the first global assessment of current geographic PBD patterns for angiosperms. Our results showed that such patterns are largely dictated by global environmental and geographic gradients, with lineage replacement being more important than lineage loss in virtually all areas, except at higher latitudes and on islands and peninsulas.