Genomics outperforms genetics to manage mistakes in fisheries stocking of threatened species

Hatchery production and fisheries stocking is a widespread and high profile management practice because it allows recreational fisheries to continue in threatened species. Human-mediated transfer of fish across the geographic boundaries of intraspecies lineages or closely related species can cause introgression and occasionally outbreeding depression. Hybridization can be difficult to detect due to limited morphological differences among close lineages and the relatively low power of traditional genetic datasets. Here we showcase the use of genomic techniques to detect admixture of the economically important and threatened golden perch (Macquaria ambigua) in the Murray-Darling Basin, southeast Australia. We detected admixture through a genome-wide dataset of 6,862 single nucleotide polymorphisms (SNPs) across 174 Murray-Darling sourced fish and 15 fish from each of two neighbouring basins: the Lake Eyre and Fitzroy basins. Individuals with partial ancestry from both neighboring basins were detected using genomics throughout the Murray-Darling, suggesting the release of individuals and introgression into the Murray-Darling Basin. Importantly, a traditional microsatellite dataset was unreliable for identifying admixed individuals. The SNP-detected admixed individuals were also found in Murray-Darling impoundments, where fish are solely sourced from government-managed hatcheries, suggesting that some broodstock in hatcheries might have non-endemic ancestry. Stocking programs for golden perch release over one million fingerlings each year, and so could impact the genetic variation in the wild. We advocate for using genomics to check the ancestry of broodstock and for increasing collaboration between managers and academics—as done here—to better integrate the power of genomics into biodiversity management and conservation.