Current global patterns of biodiversity result from processes that operate over both space and time and thus require an integrated macroecological and macroevolutionary perspective1,2,3,4. Molecular time trees have advanced our understanding of the tempo and mode of diversification5,6,7 and have identified remarkable adaptive radiations across the tree of life8,9,10. However, incomplete joint phylogenetic and geographic sampling has limited broad-scale inference. Thus, the relative prevalence of rapid radiations and the importance of their geographic settings in shaping global biodiversity patterns remain unclear. Here we present, analyse and map the first complete dated phylogeny of all 9,993 extant species of birds, a widely studied group showing many unique adaptations. We find that birds have undergone a strong increase in diversification rate from about 50 million years ago to the near present. This acceleration is due to a number of significant rate increases, both within songbirds and within other young and mostly temperate radiations including the waterfowl, gulls and woodpeckers. Importantly, species characterized with very high past diversification rates are interspersed throughout the avian tree and across geographic space. Geographically, the major differences in diversification rates are hemispheric rather than latitudinal, with bird assemblages in Asia, North America and southern South America containing a disproportionate number of species from recent rapid radiations. The contribution of rapidly radiating lineages to both temporal diversification dynamics and spatial distributions of species diversity illustrates the benefits of an inclusive geographical and taxonomical perspective. Overall, whereas constituent clades may exhibit slowdowns10,11, the adaptive zone into which modern birds have diversified since the Cretaceous may still offer opportunities for diversification.
We thank D. Redding for critical input in the early stages of this project; A. Mimoto, F. Ronqvist and M. Teslenko for help modifying MrBayes; I. Martyn for coding; R. Bowie, J. McGuire, A. Cooper, K. Burns and M. Sorenson among others, for unpublished phylogenetic material or information; M. Benton, T. Ezard, T. Price, M. Donoghue, J. Beaulieu, J. Belmaker, P. M. Hull, D. Field, N. Longrich, V. Saranathan, M. Steel, H. Morlon, J. Brown, A. Phillimore, R. Fitzjohn, R. Etienne, W. Stein and especially T. Stadler for data, important input and/or discussion; G. Smith, C. Schank, D. Thiele, T. M. Lee, F. La Sorte, C. Edwards, K. Ashton and J. Hazelhurst for help with spatial and phylogenetic data collection and management; C. Schank for help preparing the tree visualizations. This work was carried out using the BlueFern Supercomputing Facilities (http://www.bluefern.canterbury.ac.nz), University of Canterbury, the Advanced Computing Research Centre, University of Bristol (http://www.bris.ac.uk/acrc/) and the Interdisciplinary Research in Mathematics and Computer Sciences Centre, Simon Fraser University (http://www.irmacs.sfu.ca). This work was partly supported by NSF grants DBI 0960550 and DEB 1026764 and NASA Biodiversity Grant NNX11AP72G (W.J.); the Natural Environment Research Council (Postdoctoral Fellowship grant number NE/G012938/1 and the NERC Centre for Population Biology) (G.H.T.); and NSERC Canada, the Wissenschaftskolleg zu Berlin, the Yale Institute for Biospheric Sciences and Simon Fraser University (A.O.M.). Most importantly, we thank the many avian systematists and phylogeneticists who have contributed their data to public databases and so made our study possible.
This zipped file contains the Supplementary Data files - see Supplementary Information file (pg 32) for details.