Applying an Index of Biotic Integrity Based on Stream-Fish Communities: Considerations in Sampling and Interpretation

Physical and chemical monitoring of water bodies is relatively common. However, water resource assessment suffers from lack of integrative and reliable measures of the biotic condition of aquatic systems. This situation persists despite the fundamentally biological nature of water resource degradation. We examined three aspects of applying an index of biotic integrity (IBI) that uses attributes of fish communities to assess stream degradation: (1) relative contributions of individual metrics to the final IBI assessment; (2) effects of sampling effort on the IBI; and (3) effects of including young-of-year fish data on IBI computations. Relative contributions of individual metrics to IBI assessments varied substantially among data sets from Illinois, Ohio, and West Virginia, and variation in contributions reflected differences in ranges of metric scores and types of degradation being assessed. No metric was consistently best or worst at detecting degradation, and metrics did not appear redundant with respect to each other. The IBI scores increased with increasing length of stream sampled, largely due to increases in the number of species accumulated in fish collections. The IBI scores from short reaches were more variable than scores from long reaches; adequate sample reaches should contain several pool-riffle sequences. Exclusion of young-of-year fish from IBI computations typically reduced IBI scores by 2-10 units; the magnitude of reduction reflected the proportion of young of year in the samples. Species-richness metrics were affected more by exclusion of young of year than other metrics. The IBI appears to be broadly useful to water resource managers and, if used in conjunction with other measures of water and habitat quality, should provide a comprehensive approach for assessing resource quality.