Evaluating an invasive species policy: ballast water exchange in the Great Lakes

Improvements in environmental policy require an accurate diagnosis of the shortcomings of existing policy. We develop a model for assessing the efficacy of policy instruments aimed at reducing the introduction of nonindigenous species. The model identifies and accounts for several features of the nonindigenous species introduction-detection process that complicate interpretations of monitoring data. Specifically, the model includes explicit attention to the pathway of introduction, a probabilistic description of species detection, and the possibility of attenuation of species introductions over time. We apply this theoretical model to the case of mid-ocean ballast water exchange, which was implemented by the United States in 1990 for the North American Great Lakes. Contrary to other authors who take the recent increase in discoveries of nonindigeneous species (NIS) in the Great Lakes as evidence that ballast water exchange is ineffective, we find that the observed detection record could just as plausibly be explained by a lag of a few years between introduction and detection, even if ballast water exchange was 100% effective. Model results suggest that, under current monitoring regimes, several more years of data would be required to make a conclusive evaluation of ballast water exchange. Better estimation of the lag time between introduction and detection, and a shortening of that lag time with better monitoring, would allow more precise and timely evaluation of the efficacy of ballast water exchange and other policy instruments.