Using population genetics and demographic reconstruction to predict outcomes of genetic rescue for an endangered songbird

Genetic rescue can be a successful way to restore species genetic diversity, but it can also lead to outbreeding depression (decreases in hybrid fitness) if attempted in incompatible populations. Thus, population genetic profiles and demographic history are needed to evaluate the feasibility of translocation. We used population genetic analyses and Approximate Bayesian Computation (ABC) to assess genetic rescue as an option for two populations of the yellow-shouldered blackbird (Agelaius xanthomus), an endangered Puerto Rico endemic. The candidate recipient population, a managed population in Pitahaya (southwestern Puerto Rico), had been characterized previously for its mating system and population genetics. Here, we used nine microsatellite loci to measure the genetic diversity of a candidate source population, a subspecies (A. x. monensis) on Mona Island, 66 km west of Puerto Rico. We compared genetic diversity and inferred historical and contemporary gene flow between the two populations. We found clear population structure and no migration between populations, as well as evidence that the Mona population descended from the Pitahaya population approximately 95 generations ago. Despite the historical gene flow, the degree of contemporary genetic and environmental divergence means the Mona population may not be suitable for immediate use as a source population. We recommend (a) further investigating whether the observed population structure is due to adaptive or neutral forces, (b) testing the Mona population for behavioral plasticity in different environments, and (c) evaluating other source populations in addition to the Mona population for genetic rescue.