Inferring the timing of long-distance dispersal between Rail metapopulations using genetic and isotopic assignments.

The stochastic and infrequent nature of long-distance dispersal often makes it difficult to detect. We quantified the frequency, distance, and timing of long-distance dispersal in a nonmigratory, secretive wetland bird, the California Black Rail (Laterallus jamaicensis coturniculus), between an inland and a coastal metapopulation separated by greater than 100 km. Using 15 microsatellites in conjunction with stable carbon, nitrogen, and sulfur isotopes, we classified Rails as residents of their capture population, recent migrants that dispersed to their capture population less than one year before capture, established migrants that dispersed to their capture population more than one year before capture, and seasonal migrants that dispersed away from their capture population to forage, but returned the next season. Most Rails (195 of 204, or 95.6%) were classified as residents, but we detected two established migrants that had moved >100 km more than a year before capture. Seven Rails appeared to be seasonal migrants, but comparisons of feather isotope values with isotope values from wetland soils indicated that the isotope values in the feathers of these Rails likely resulted from natural environmental variation (e.g., source element effects) rather than long-distance dispersal of individuals. Thus, these seven Rails were most likely misassigned by isotopic population assignments due to small-scale variation in the isoscape. Using genetic data in conjunction with isotopic data allowed us to not only infer the timing of long-distance dispersal events, but to successfully track long-distance movements of nonmigratory Rails between metapopulations even when environmental variation of isotopes occurred across small spatial scales.