Login
Register
Print Friendly
Risk assessment and screening of potentially invasive fishes
Research Fishery Biologist; U.S. Geological Survey; Upper Midwest Environmental Sciences Center; 2630 Fanta Reed Road; La Crosse, Wisconsin 54603 USA
Email: ckolar@usgs.gov
Invasive species are a leading threat to global biodiversity, are economically costly, and cause ecological damage. Management of invasive species has historically focused on mitigation and control after invaders become established, after eradication is no longer a viable management option, and well after prevention efforts could have precluded their entry. In order to better protect their ecosystems and to reduce monetary damage, various countries around the world have enacted legislation or have legislation pending requiring the screening of species to identify potentially invasive species prior to their importation. Screening tools used for such purposes need to be scientifically based, transparent, relatively easy-to-use, and unbiased (i.e., give the same categorization regardless of the person using the screening tool) in order to be defensible and fair.
Screening tools to predict potentially invasive species are based on the assumption that invasive species are somehow predictably different from species that tend not to invade. This assumption had not often been quantitatively assessed until recently. Focused studies examining one ecosystem, taxon, and invasion stage (e.g., establishment or invasiveness) have shown that patterns of species characteristics are indeed emerging. Several studies, for instance, have identified species characteristics common to successfully establishing birds and to weedy plants in particular ecosystems. Once identified, the characteristics common to invasive species can be used to develop statistical models to identify high risk species for a given ecosystem. These models could be used to screen species being considered for importation or to develop species-specific management strategies concentrating efforts on preventing the entry of high risk species by other pathways. I will use research from my dissertation, directed by David M. Lodge at the University of Notre Dame, Notre Dame, Indiana, USA, on fish invasions in the Laurentian Great Lakes, as an example of these models may be developed and implemented. This general approach, however, could be applicable to a variety of ecosystems and taxa.
David Lodge and I developed statistical models based on species life history characteristics to differentiate fishes introduced successfully (n=25 species) from those that failed to become established in the Great Lakes (n=21 species). We considered 26 variables (14 life history characteristics, 5 environmental tolerances, 6 aspects of invasion history, and degree of human use) and found that four (relative growth rate, range of inhabitable temperatures, range of salinity tolerance, and history of invasion) accurately discriminated successful from unsuccessful fishes with 87-96% accuracy using discriminant analysis and categorical and regression tree analysis. Using our statistical models and life history information from Eurasian fishes, we next identified 18 fishes from the Ponto-Caspian basin, a donor region for species in the Great Lakes, at a high risk of becoming established in the Great Lakes if introduced. We also used similar models to discriminate fishes that have spread quickly through the Great Lakes ecosystem from those that have spread more slowly, and discriminated fishes perceived as a nuisance from those that are not. Specific management recommendations can be developed after the identity, distribution, and biology of potential invaders are understood.
Legislation is pending in the United States Congress which would require the screening of plant and animal species imported into the country that have not been in trade since 1990. Another method of determining the potential distribution of aquatic invasive species, General Algorithm for Rule-set Predictions (GARP) will be presented. New management efforts are needed to prevent the realization of these predictions.
Room 1 Monday 12.50 pm
