Estimating the relative fitness of escaped farmed salmon offspring in the wild and modelling the consequences of invasion for wild populations

Throughout their native range, wild Atlantic salmon populations are threatened by hybridization and introgression with escapees from net‐pen salmon aquaculture. Although domestic–wild hybrid offspring have shown reduced fitness in laboratory and field experiments, consequential impacts on population abundance and genetic integrity remain difficult to predict in the field, in part because the strength of selection against domestic offspring is often unknown and context‐dependent. Here, we follow a single large escape event of farmed Atlantic salmon in southern Newfoundland and monitor changes in the in‐river proportions of hybrids and feral individuals over time using genetically based hybrid identification. Over a three‐year period following the escape, the overall proportion of wild parr increased consistently (total wild proportion of 71.6%, 75.1% and 87.5% each year, respectively), with subsequent declines in feral (genetically pure farmed individuals originating from escaped, farmed adults) and hybrid parr. We quantify the strength of selection against parr of aquaculture ancestry and explore the genetic and demographic consequences for populations in the region. Within‐cohort changes in the relative proportions of feral and F1 parr suggest reduced relative survival compared to wild individuals over the first (0.15 and 0.81 for feral and F1, respectively) and second years of life (0.26, 0.83). These relative survivorship estimates were used to inform an individual‐based salmon eco‐genetic model to project changes in adult abundance and overall allele frequency across three invasion scenarios ranging from short‐term to long‐term invasion and three relative survival scenarios. Modelling results indicate that total population abundance and time to recovery were greatly affected by relative survivorship and predict significant declines in wild population abundance under continued large escape events and calculated survivorship. Overall, this work demonstrates the importance of estimating the strength of selection against domestic offspring in the wild to predict the long‐term impact of farmed salmon escape events on wild populations.