Geographic origin of the UCSD juncos
Phylogenetic and co-ancestry analyses strongly supported the formpinosus as the sister group of the UCSD juncos, and not the population at the nearby Laguna Mountains as reported in previous studies. Phylogenetic, co-ancestry or population structure analyses failed to detect any significant admixture or clustering of birds breeding or wintering at UCSD with the thurberi individuals from Laguna Mountains. One UCSD breeder did present an intermediate assignment probability to the Laguna Mountains population, while one of the likely residents detected among the UCSD wintering birds showed some admixture with pinosus , which is consistent with the migration rates estimated in fastSIMCOAL2 analyses. These two birds also occupy intermediate positions between the pinosus -northern Oregon juncos and the UCSD clusters in the PCA, suggesting these two individuals are likely the result of recent introgression.
The population genomics analyses presented here thus support a recent colonization of urban San Diego by a few individuals from thepinosus population up the coast, followed by limited gene flow from other Oregon junco forms over the last ca. 40 years. Although gene flow between divergent populations could have acted as a homogenizing force, eroding population differentiation (Slatkin 1987; Wright 1931), there is theoretical and empirical evidence showing that under conditions of rapid population growth following a founder event, divergence of allele frequencies established during colonization may be resistant to decay by gene exchange (Boileau et al. 1992; Lombalet al. 2020). Moreover, moderate admixture can also contribute to colonization success by minimizing deleterious effects of reduced intra-population diversity or by introducing favorable alleles and enhancing evolutionary potential (Keller & Taylor 2010). That introgression between taxa contributes to adaptation and sorting of genes has become increasingly recognized as a major contributor to population divergence and speciation (Abbott et al. 2013; Lamichhaney et al. 2015; Wegener et al. 2019). Based on our genomic analyses and on the migration rates reported here and in Yeh and Price (2004), as well as in the phenotypic studies in Yeh (2004), it seems that several over-wintering birds on the UCSD campus came from populations other than pinosus , and some of these may have remained to breed on campus. It thus appears likely that there has been introgression into the UCSD population from forms other thanpinosus , and its relevance remains to be determined.
The form pinosus , from which the UCSD population originated, is a non-migratory form found along the coastal ranges of central California and occurs in a range of Mediterranean habitats, displaying a tolerance for low zonal conditions not found in thurberi (Miller 1941). Individuals from the pinosus population would generally be expected to produce higher fitness progeny at UCSD than thurberiindividuals. The amount of white on the tail feathers of pinosusis reduced relative to thurberi (Miller 1941), and more similar to UCSD individuals. Indeed, Price et al. (2008) inferred strong selection on juvenile juncos carrying excess white in their tail, which they attributed to both direct selection on white, and correlated responses to greater dispersal propensity, both of which would be most prominent in offspring from thurberi immigrants, and lead to lower rates of introgression. Nevertheless, some alleles may successfully introgress from other subspecies, and the large pool from which immigrants appear to be drawn from increases chances of adaptive alleles such as the ones we have inferred from entering into the UCSD population.