Variation in a Great River index of biotic integrity over a 20-year period

The Index of Biotic Integrity (IBI) has proven to be an important assessmen t tool for evaluating the resource quality of aquatic ecosystems. We used a modified IBI for Great Rivers based on collections from the Wabash River i n west-central Indiana. We evaluated natural, inherent variation in natural phenomena, gear effects and land use modification over a 20-year period. S ixty-three stations distributed through 260 km of river were sampled 2-3 ti mes per year to evaluate fish community structure and function. Natural cha nges in variation included temporal variation, hydrologic cycle and recruit ment. Our results suggest that removing gizzard shad, whose population leve ls fluctuate widely, from percentage metrics enhanced assessment and expose d subtle affects. IBI scores consistently declined in a downstream directio n. IBI profiles based on numbers were almost always higher than those using biomass, but patterns corresponded well for both metric systems. Combining biomass catches by seining and electrofishing did not substantially change the observed patterns compared to numeric electrofishing catches. The larg e number of small species taken by seining overwhelmed the subtle patterns observed at least impacted stations, but greater departures in quality were evident at impaired downstream locations. Floods decimated fish communitie s throughout the river. Fish moved out of stressed areas during severe drou ghts and into areas near clean tributaries. Certain structured attributes o f Great River fish communities showed statistically significant changes bas ed on ecoregion scales. Total number of species, number of sensitive specie s, number of centrarchid species and percentage of simple lithophils declin ed between the Corn Belt Plain and Interior River Lowland ecoregions, but w ater quality factors might also be influential. These changes in metric att ributes lowered IBI scores by 6-10 cumulative points.