A comparison of approaches for estimating relative impacts of nonnative fishes.

Lack of standard methods for quantifying impact has hindered risk assessments of high-impact invaders. To understand methodological strengths and weaknesses, we compared five approaches (in parentheses) for quantifying impact of nonnative fishes: reviewing documented impacts in a large-scale database (review); surveying fish biologists regarding three categories of impact (socioeconomic, ecological, abundance); and estimating frequency of occurrence from existing collection records (collection). In addition, we compared game and nongame biologists' ratings of game and nongame species. Although mean species ratings were generally correlated among approaches, we documented important discrepancies. The review approach required little effort but often inaccurately estimated impact in our study region (Mid-Atlantic United States). Game fishes received lower ratings from the socioeconomic approach, which yielded the greatest consistency among respondents. The ecological approach exhibited lower respondent bias but was sensitive to pre-existing perceptions of high-impact invaders. The abundance approach provided the least-biased assessment of region-specific impact but did not account for differences in per-capita effects among species. The collection approach required the most effort and did not provide reliable estimates of impact. Multiple approaches to assessing a species' impact are instructive, but impact ratings must be interpreted in the context of methodological strengths and weaknesses and key management issues. A combination of our ecological and abundance approaches may be most appropriate for assessing ecological impact, whereas our socioeconomic approach is more useful for understanding social dimensions. These approaches are readily transferrable to other regions and taxa; if refined, they can help standardize the assessment of impacts of nonnative species.