Relationships between indicators of acid-base chemistry and fish assemblages in streams of the Great Smoky Mountains National Park

The acidity of many streams in the Great Smoky Mountains National Park (GRSM) has increased significantly since pre-industrial (∼1850) times due to the effects of highly acidic atmospheric deposition in poorly buffered watersheds. Extensive stream-monitoring programs since 1993 have shown that fish and macroinvertebrate assemblages have been adversely affected in many streams across the GRSM. Matching chemistry and fishery information collected from 389 surveys performed at 52 stream sites over a 22-year period were assessed using logistic regression analysis to help inform the U.S. Environmental Protection Agency's assessment of the environmental impacts of emissions of oxides of nitrogen (NOx) and sulfur (SOx). Numerous logistic equations and associated curves were derived that defined the relations between acid neutralizing capacity (ANC) or pH and different levels of community richness, density, and biomass; and density and biomass of brook trout, rainbow trout, and small prey (minnow) populations in streams of the GRSM. The equations and curves describe the status of fish assemblages in the GRSM under contemporary emission levels and deposition loads of nitrogen (N) and sulfur (S) and provide a means to estimate how newly proposed (and various alternative) target deposition loads, which strongly influence stream ANC, might affect key ecological indicators. Several examples using ANC, community richness, and brook trout density are presented to illustrate the steps needed to predict how future changes in stream chemistry (resulting from different target deposition loads of N and S) will affect the probabilities of observing specific levels of selected biological indicators in GRSM streams. The implications of this study to the regulation of NOx and SOx emissions, water quality, and fisheries management in streams of the GRSM are discussed, but also qualified by the fact that specific examples provided need to be further explored before recommendations concerning their use as ecological indicators could be proposed.