Search for a command to run...
Glacial cycles operating across Beringia have repeatedly exposed large swathes of the Bering Land Bridge, intermittently isolating and reuniting North American and Eurasian taxa. In high-latitude birds, these cycles are hypothesized to have been important in driving divergence and speciation. These repeated events have resulted in multiple trans-Beringian avian sister populations of varying degrees of taxonomic depth distributed across modern Beringia. We asked how these cyclic pulses have affected the temporal distribution and number of overall divergence events across Beringia. We sequenced full mitogenomes at high depth from 39 lineage pairs of varying levels of divergence, totaling 432 individuals of seven orders, 14 families, and 49 species from both Eurasia and North America. We then used a hierarchical approximate Bayesian comparative (hABC) approach to estimate the number and distribution of divergence events between the population pairs, using subsampled datasets. Net nucleotide divergence (<i>D<sub>A</sub></i> ) and Jukes-Cantor distance (JC-distance) were also calculated for each pairwise comparison to estimate divergence dates between taxa, using calibrated rates appropriate for shallow avian divergence events. Average divergence times were 200,000 ya for population-level taxa (<i>n</i> = 16), 720,000 ya for subspecies (<i>n</i> = 12), and 1 Mya for species (<i>n</i> = 11), although we consider these dating estimates conservative because of a lack of appropriate calibration for data of this quality. We found eighteen taxon pairs to be significantly differentiated (<i>p</i> < 0.05) by <i>F<sub>ST</sub></i> or substantially differentiated by haplotype clade, bounding the number of potential overall divergence events from 1 to 18, and two subsets of the full mitogenomic dataset analyzed in MTML-msBayes strongly supported simultaneous divergence of all Beringian lineages. However, this finding of simultaneous divergence is biologically unusual given the substantial variation in divergence dates among taxa and might indicate a relatively continuous spread of vicariance events, which is difficult to distinguish from a single, simultaneous vicariance event.