Biogeographic consequences of historic and contemporary climate change in boreal forest birds.

摘要

In this dissertation, I combine ecological niche models (ENMs), which can be extrapolated through time to predict historic and future changes in species distributions, with mitochondrial and nuclear genetic markers to study the biogeographic consequences of historic and contemporary climate change on boreal forest birds, and in particular Blackpoll Warbler.;Blackpoll Warbler's ENM was projected to Last Glacial Maximum (LGM) and suggested three suitable areas in North America. Based on this paleodistribution I tested ten alternative historic hypotheses spanning one, two, and three refugia and four divergence times, by comparing empirical data to 1000 genealogies and 500 sequence datasets simulated within each alternative model according to the coalescent process. Despite significant genetic structure among western, eastern, and Newfoundland populations, all multiple refugia hypotheses were rejected in favor of a single refuge in southeastern North America. Current genetic structure is likely the result of limited gene flow due to isolation by distance.;To reevaluate the historic concordance among codistributed boreal forest birds, I compared Blackpoll Warbler results to three additional species by building paleodistribution models and reanalyzing published datasets using coalescent methods. Blackpoll Warblers, Yellow-rumped Warblers, and Dark-eyed Juncos exhibit congruent demographic patterns with unstructured haplotype networks and single refugia. However, this may represent phylogeographic pseudocongruence, as paleodistribution models suggest these species were not codistributed at LGM. Swainson's Thrush differs from the others in that it persisted in two refugia, but was likely codistributed at LGM with Blackpoll Warblers in the east, and Yellow-rumped Warblers in the west, representing a phylogeographic pseudoincongruence.;Fifteen species of boreal forest birds are predicted to shift north and decrease or disappear from mountain populations by 2080 under two carbon emissions scenarios. To determine the genetic consequences of this shift in Blackpoll Warblers I compare genetic variation at nine microsatellite loci between mountain populations predicted to go extinct and northern populations predicted to persist. Alleles private to mountain populations represent a significant percent of diversity, but were all found at low frequency. Losing these alleles would not significantly affect heterozygosity and more than 99% of genetic structure is predicted to survive the loss of mountain populations.