Understanding how interactions among microevolutionary forces generate genetic population structure of exploited species is vital to the implementation of management policies that facilitate population persistence. Philopatry displayed by many coastal shark species can impact gene flow and facilitate selection, and thus has direct implications for the spatial scales of management plans. Here, genetic structure of the blacktip shark (Carcharhinus limbatus) was examined using a mixed-marker approach based on mitochondrial control region sequences and 4,339 SNP-containing loci generated using ddRAD-Seq. Genetic variation was assessed among young-of-the-year sampled in 11 sites in waters of the United States in the western North Atlantic Ocean, including the Gulf of Mexico. Spatial and environmental analyses detected 68 nuclear loci putatively under selection, enabling separate assessments of neutral and adaptive genetic structure. Both mitochondrial and neutral SNP data indicated three genetically distinct units – the Atlantic, eastern Gulf, and western Gulf – that align with regional stocks and suggest regional philopatry by males and females. Heterogeneity at loci putatively under selection, associated with temperature and salinity, was observed among sites within Gulf units, suggesting local adaptation. Furthermore, five pairs of siblings were identified in the same site across timescales corresponding with female reproductive cycles. This indicates that females re-used a site for parturition, which has the potential to facilitate the sorting of adaptive variation among neighboring sites. The results demonstrate differential impacts of microevolutionary forces at varying spatial scales and highlight the importance of conserving essential habitats to maintain sources of adaptive variation that may buffer species against environmental change.

Fin clips were collected from blacktip sharks (Carcharhinus limbatus) within or near 11 estuaries used for parturition in waters of the United States in the western North Atlantic Ocean. A modified version of double digest restriction-site associated DNA sequencing (ddRAD-Seq) was used to prepare genomic libraries containing 488 young-of-the-year individuals plus technical replicates, which were sequenced using 11 lanes of an Illumina HiSeq 4000 (paired-end 150 bp reads). To map and improve the genotyping efficacy of HiSeq data, a separate library consisting of 27 individuals sampled across Atlantic and Gulf locations at multiple life-history stages was prepared using the same protocol but sequenced on an Illumina MiSeq lane (paired-end 300 bp reads).

All MiSeq and HiSeq reads were quality-trimmed using default parameters in dDocent. dDocent was also used to assemble MiSeq reads into a reduced genomic reference. dDocent was then used to map HiSeq reads to the MiSeq reference and genotype SNPs.

SNPs were filtered using VCFtools and R functions in a customized workflow. Subsequently, haplotypes were generated by collapsing SNPs on the same contig to produce a multi-allelic SNP-containing loci dataset. Technical replicates and non-randomly sampled kin were then removed.

In addition, a 915 bp portion of the mitochondrial control region (1070 bp total length) was amplified for a subset of individuals.

The MiSeq reference, `dDocent` config files, `genepop` file, and strata data can be opened and viewed in a text editor. The `.arp` and `.ars` files can also be opened and viewed in a text editor but should be executed in Arlequin.