Population structure of loggerhead sea turtles (Caretta caretta) nesting in the southeastern United States inferred from mitochondrial DNA sequences and microsatellite loci
Shamblin, Brian Michael
MetadataShow full item record
The loggerhead sea turtle Caretta caretta is a globally threatened species. The southeastern United States of America (USA) nesting assemblage is one of the two primary loggerhead rookeries in the world and the largest in the Atlantic. Determining the stock structure within this nesting aggregation is important for defining management units and providing baseline data for mixed stock analyses of foraging aggregations and stranded turtles. I conducted molecular analyses utilizing mitochondrial DNA (mtDNA) sequence and newly developed microsatellite primers on samples collected from nine sample sites from North Carolina to the Florida panhandle to characterize the population structure of rookeries comprising the southeastern USA nesting assemblage. Pairwise comparisons of mtDNA haplotype frequencies among eight sample sites utilizing genetic distance-based and frequency-based models suggested significant structure among most sites tested in Florida. Pairwise F-statistics and R-statistics revealed no significant nuclear allele frequency differences among any of the nine sample sites. This pattern of complex population structure may be attributable to migrationmediated gene flow among nesting beaches that are discrete by virtue of female nest site fidelity. However, pairwise tests of genotypic frequency divergence did suggest shallow but significant differences between Georgia and Florida sites and South Carolina and Florida sites, consistent with the segregation of northern subpopulation and Florida loggerheads on foraging grounds supported by carapace epibiont and tagging studies. Analysis of molecular variance (AMOVA) and pairwise comparisons suggested that the currently designated south Florida management unit may be comprised of multiple management units. More extensive geographical sampling is required to make more robust inferences about the integrity and boundaries of management units.