Evaluating Non-Indigenous Species Eradication Options in a Bayesian Network Derived Adaptive Management Framework

Many coastal regions are facing problems with the introduction and spread of non-indigenous species (NIS). Common efforts addressing this issue include eradicating these species, which can occur at different stages of the NIS invasions, such as elimination of these species before being introduced to the habitat, or removal of the species after settlement. Eradication methods can either target multiple species (such as with ballast water treatments) or single species eradication, with chemical and/or mechanical treatment options. Little information is available as to the consequences of some of these eradication practices in terms of ecological and toxicological impacts to the surrounding habitat. A regional risk assessment using a Bayesian Network Model, is being conducted in Padilla Bay, Washington, a National Estuarine Research Reserve. The objectives of this study are to 1) determine the vectors of introduction that are associated with higher risks of NIS invasions and 2) analyze various management options that will reduce the risk of NIS introductions, while being least disruptive to the marine community. The Bayesian Network Model is advantageous because it allows us to analyze various adaptive management options for controlling NIS, comparing and contrasting methods such as chemical and mechanical eradication, as well as various treatments of ballast water before it is released into coastal waters. The results from this study will allow us to evaluate the likelihood of NIS risk reduction outcomes from each management option in respect to the endpoints in the surrounding habitats. The framework from the risk assessment and adaptive management will be adaptable for other regions interested in the eradication of NIS organisms.