摘要：Three assessment methods (individual metrics, multimetric indices, and multivariate statistics) were compared for their ability to identify impairment in fish, macroinvertebrate, and periphyton stream communities. The multivariate statistical analysis showed the greatest ability to differentiate temporal and spatial differences. Some individual metrics were able to identify large spatial differences (e.g., stream order) and some smaller between site differences. The multimetric indices were least able to identi6 differences. Results were similar for all three communities. While the multivariate method provided the best ability to identify differences, a disadvantage is that some of these differences may not be due to impairment. We found the optimum approach to be a pairing of the multivariate analysis with specific individual metrics targeted to species of concern, providing a way of linking any significant differences with relevant impairments.
