Evaluating non-indigenous species management in a bayesian networks derived framework, Padilla Bay, WA

Many coastal regions are encountering issues with the introduction and spread of non-indigenous species (NIS). There are many vectors that can transport NIS to coastal areas and estuaries. In this study, I conducted a regional risk assessment using a Bayesian networks relative risk model (BN-RRM) to analyze multiple vectors of NIS introduction to Padilla Bay, Washington, a National Estuarine Research Reserve. Bayesian networks models are advantageous because they are parameterized with quantitative data and knowledge, uncertainty can be incorporated into these models, and the calculated risk is described as a distribution of risk for the various endpoints of interest. The objectives of the study were to 1) determine if the BN-RRM could be used to calculate risk from NIS introductions; 2) determine which regions and endpoints were at greatest risk from NIS introductions and impacts; and 3) examine a management option and calculate the reduction of risk to the endpoints if it were to be implemented. Efforts to manage NIS colonization include eradication of the species. This can occur at different stages of NIS invasions, such as the elimination of these species before being introduced to the habitat, or removal of the species after settlement. A management option was easily incorporated into the model to observe the risk to the endpoints if the treatment were to be implemented. This risk could then be compared to the initial risk estimates. The results from this risk assessment indicate the southern portion of Padilla Bay, Regions 3 and 4 had the greatest risk associated with them and the changes in community composition, Dungeness crab, and eelgrass were the endpoints with the most risk due to NIS introductions. The Currents node, which controls the exposure of NIS to the bay, was the parameter that had the greatest influence on risk to the endpoints. The ballast water management treatment displayed one percent reduction in risk in this Padilla Bay case study. These models provide an adaptable template for decision makers interested in managing NIS and aquatic environments in other coastal regions and large bodies of water.