Aim The evolution of avian speciation patterns across much of Eurasia is under-explored. Excepting phylogeographic patterns of single species, or speciation involving the Himalayas, there has been no attempt to understand the evolution of avian distributional patterns across the rest of the continent. Within many genera there is a pattern of (presumed) sister species occurring in adjacent areas (western, eastern or southern Eurasia), yet this pattern cannot be explained by existing biogeographic barriers. My aim was to examine the possible role of climate-driven vicariance events in generating avian distributions.
Methods I constructed a molecular phylogeny of Phoenicurus redstarts, and assembled phylogenetic data from published studies of seven other Eurasian bird genera. On each phylogeny, I assessed the distributional patterns of species and clades relative to refugial areas in western, eastern and southern Eurasia. I also estimated the timing of lineage divergences via a molecular clock, to determine whether distributional patterns can be explained by well-defined periods of climate change in Eurasia that are recorded from dated sediments in the Chinese Loess Plateau.
Results Species relationships in a well-supported phylogeny of Phoenicurus show a pattern of distributions consistent with repeated speciation in major refugial areas, where one lineage is isolated in a single area of Eurasia relative to its sister lineage. This same pattern is evident in Eurasian Turdus thrushes, and six additional avian genera distributed across Eurasia. Molecular clock dating indicates that divergences within each genus are the result of multiple rounds of speciation in refugia through time, during major climate-driven episodes of vicariance.
Main conclusions Analyses revealed substantial evidence supporting a repeated, non-random pattern of speciation within and across eight songbird lineages since the Late Miocene. The pattern of speciation supports a model of isolation in refugia during major episodes of vicariance, specifically periods of either intensified desertification of Central Asia or Eurasian glacial cycles. The densely sampled clades used here preclude inter-continental dispersal as an alternative explanation for distributions. The signature of climate-driven vicariance across epochs is, given the absence of extant biogeographic barriers, a suitable hypothesis to explain major lineage divergences in widely distributed Eurasian songbird lineages.