A Benthic Index of Biotic Integrity (B‐IBI) for Rivers of the Tennessee Valley

Abstract

Invertebrate data from rivers in the Tennessee Valley were used to: (1) evaluate the utility of 18 characteristics of invertebrate assemblages (attributes) to assess the biological condition of streams and (2) develop a comprehensive benthic invertebrate index that reflects important aspects stream biology and responds to the effects of human society in detectable ways. We used data from the Tennessee Valley Authority's (TVA) fixed—station monitoring program established in 1986 to evaluate water resources in the Tennessee Valley. The streams covered sites with various types and levels of human disturbance. TVA biologists sampled riffles and pools using quantitative methods. To evaluate attributes we determined (1) the ability of each attribute to distinguish among sites (analyses of variance), (2) the associations among attributes (correlation coefficients), (3) the concordance of assessments of sites using each attribute with two independent site assessments (water quality and analysis of fish assemblages based on the index of biotic integrity), and (4) the variability of each attribute across habitats (riffles and pools) and ecoregions. Thirteen attributes (total taxa richness and taxa richnesses of intolerant snails and mussels, mayflies, caddisflies, and stoneflies; relative abundances of Corbicula, oligochaetes, omnivores, filterers, grazers, and predators; dominance; total abundance) were valuable in discriminating sites, exhibited concordance with other methods of assessment, and were relatively uncorrelated among themselves. Five attributes (sediment—surface taxa richness; relative abundances of shredders, gatherers, detritivores, and chironomids) were not acceptable for inclusion in the index because they either were not in concordance with other methods of assessment or were strongly correlated with other attributes. Based on the results of this evaluation, 13 attributes were included in a benthic index of biotic integrity (B—IBI). The B—IBI was evaluated using a more extensive fixed—station data set and was tested using an independent data set where samples were taken above and below the release of an industrial effluent to the North Fork Holston River. The fixed—station sites were correctly ranked, although rankings were not always consistent for pool and riffle habitats. B—IBI also distinguished the industrial impact. The B—IBI has potential as an assessment tool for streams, although more testing and evaluation of both the attributes and the index are warranted.