Species distributions and the recognition of risk in restoration planning: A case study of salmonid fishes

One of the risks faced by habitat restoration practitioners is whether habitats included in restoration planning will be used by the target species or, conversely, whether habitats excluded from restoration planning would have benefitted the target species. With the goal of providing a quantitative decision-making approach that represented varying levels of risk tolerance, we used multiple probability decision thresholds (PDT) to predict the range of occurrence for three anadromous fishes (Oncorhynchus spp.) in a watershed in southwestern Washington, USA. For each species, we compared the predicted range of occurrence to the distribution used for restoration planning and quantified the amount of habitat blocked by anthropogenic barriers. Coho salmon (O. kisutch) had the broadest predicted range of occurrence (3,061.6–6,357.9 km; 0.75–0.25 PDT), followed by steelhead trout (O. mykiss; 1,828.8–2,836.8 km) and chum salmon (O. keta; 1373.9–1,629.1 km). For each species, the predicted range of occurrence was similar or greater than the distribution used for restoration planning, suggesting that the current plan may exclude habitats that would benefit each species. Coho salmon had the greatest percentage of habitat blocked by anthropogenic barriers, followed by steelhead trout and chum salmon, respectively. Modelling species distributions at multiple risk tolerance scenarios acknowledges uncertainty in restoration planning and allows practitioners to weigh the ecological benefits and budgetary constraints when considering locations for restoration. To effectively communicate restoration science to support practitioners in decision making, we developed a R Shiny application online user interface available at: https://shiny.wdfw-fish.us/ChehalisRiverBasinSalmonidRangeOfOccurence/.