Genetic tools can be useful in resolving the spatial ecology of marine organisms capable of extensive dispersal and migration. However, in marine turtles, sharing of common haplotypes defined by short mitochondrial control region sequences can limit inferences of population structure and connectivity. Previous research demonstrated that the juvenile green turtle (Chelonia mydas) aggregation foraging along the southern Texas coast is dominated by haplotypes CM-A1 and CM-A3, which are shared among nesting populations in Florida, Cuba, and Mexico. To test the hypothesis that Texas turtles may represent proximal western Gulf of Mexico (GoM) nesting populations, we analyzed novel rookery samples (n = 31) from Rancho Nuevo, Tamaulipas, Mexico (RNMX). We compared these with mitogenomes from Florida-nesting CM-A1.1 females to screen for population-informative variation. This process identified a single variable position that was subsequently screened in all available Mexican-nesting and Florida-nesting CM-A1.1 turtles (n = 6 and 229, respectively). We conducted a mixed stock analysis incorporating this new mitochondrial variation to estimate the relative contributions to southern Texas juveniles collected between 1998 and 2002 (n = 167, 99 of these CM-A1.1).
Results/Conclusions
The RNMX samples yielded haplotypes CM-A1.1 and CM-A3.1 in frequencies that were not significantly different from those of the central eastern Florida stock. However, mitogenomic position 12,958 proved diagnostic between RNMX and Florida CM-A1.1 lineages, indicating that Tamaulipas rookeries warrant recognition as a unique genetic stock. Of the 99 CM-A1.1 juveniles, 97 carried the conserved RNMX variant, excluding the Florida populations as major contributors to the Texas foraging aggregation. Mixed stock analysis suggested a western GoM origin for approximately 70% of the Texas foraging aggregation, and Quintana Roo was the only nesting population outside the GoM with measurable contributions (~ 8%). These findings are consistent with oceanic dispersal followed by natal homing by neritic juveniles, a process that has been proposed to explain juvenile green turtle dispersal connectivity elsewhere in the Greater Caribbean region. This study underscores the importance of satisfying the assumptions of mixed stock analyses to enable robust inferences of connectivity. Beyond collecting baseline data for mitochondrial position 12,958 for Mexican and Cuban CM-A1.1 females, further mitogenomic screening of shared haplotypes representing each nesting population is warranted to better resolve stock structure regionally. This is particularly true for ubiquitous CM-A3. The Texas coast serves as important developmental habitat for the proposed western GoM stock, reiterating the importance of continued international cooperation to ensure recovery of this population.