Colonization patterns and genetic structure of peripheral populations of the trumpeter finch (Bucanetes githagineus) from Northwest Africa, the Canary Islands and the Iberian Peninsula

This paper has three aims: (1) to reconstruct the colonization history of two peripheral populations of the trumpeter finch (Bucanetes githagineus) presumably originating from the same source, one the result of an ancient expansion process and the other recently established and still expanding; (2) to estimate the importance of key events, such as past and current gene flow and bottlenecks, in both expansion processes and their contribution to the present population structure and genetic diversity; and (3) to find out whether two peripheral populations that established at widely differing times also differ in terms of genetic diversity. Northwest Africa (assumed source population), Canary Islands (long-established peripheral) and south-eastern Iberian Peninsula (recently established peripheral). Bayesian analysis of population structure, individual assignment tests, F-statistics, maximum likelihood migration estimates, genetic diversity indices and bottleneck tests were calculated with microsatellite data from 194 trumpeter finches from five breeding and two seasonal non-breeding sites. Our data support the existence of two subpopulations (Canary Island and Ibero-African) as the most likely population structure. Seasonal sites in the Iberian Peninsula had the highest percentage of birds assigned to other, mainly Iberian, sites. Pairwise F-ST values showed that the Canary Island localities were very similar to each other, but differed from the rest. Gene flow estimates within subpopulations were only slightly higher in the Canary Island population than in the Ibero-African one. Gene diversity indices were similar at all localities. Canary Island sites show evidence of bottlenecks, whereas the Ibero-African sites do not. Our data show that, at present, birds from the Canary Islands are genetically differentiated from those in North Africa and continental Spain. We could not unequivocally confirm the African origin of Canary populations because the contrary is also plausible. The Iberian Peninsula seems to have repeatedly received individuals from North Africa, which would have led to the relatively high genetic diversity found in these recently established localities and prevented bottlenecks. Movements of individuals towards sites outside their current range during the non-breeding season are likely to precede the establishment of new breeding sites at the periphery of the distribution range.