Stream Water Quality Modeling in the Great Smoky Mountains National Park.

The purpose of this study was to examine water quality in the acid-impacted Great Smoky Mountains National Park (GRSM). Water samples have been collected roughly quarterly at ninety sampling sites throughout the Park. These samples were analyzed for pH, acid neutralizing capacity (ANC), conductivity, major cations, and major anions. This thesis utilizes data from samples collected from October, 1993 to November, 2002.

The trout fisheries of the GRSM are considered some of the best in the eastern United States. However, fisheries biologists at the GRSM believe that some of the streams that once supported trout populations twenty or thirty years ago, no longer do. This thesis outlines and quantifies surface water quality conditions that might be harmful to trout populations through a literature review. This thesis identifies 71 sites (79% of total sampling sites) that currently have a median pH of greater than 6.0, above which, is unlikely to be harmful to trout species unless a high runoff of acid, Al-rich water creates a mixing zone where Al(OH)₃ precipitates. The precipitate can accumulate on the gills and impede normal diffusion of O₂, CO₂, and nutrients. There are 17 sites (18%) that have median pH values in the 5.0 to 6.0 range. This range of pH values is likely to be harmful to trout species when aluminum concentrations exceed about 0.2 mg/l. The lower end of this range is probably harmful to the eggs and fry of trout and also to non-acclimated trout especially when calcium, sodium, and chloride concentrations are low. The mechanisms adversely affecting trout in this range are ionoregulatory dysfunction and respiratory stress. Only two sampling sites have median pH values in the 4.5 to 5.0 range. This pH range is likely harmful to eggs, fry and adult trout, particularly in the soft water conditions prevalent in the GRSM. Ionoregulatory dysfunction, respiratory stress, and circulatory stress are the mechanisms that affect trout in this pH range. Currently, there are no sampling sites with median pH values less than 4.5, although pH values could be lowered by more than one pH unit during high-flow episodic events depending on the ANC in the stream. About 38% of the sampling sites have median ANC values in the 50 to 200 microequivalents per liter which the EPA classifies as sensitive. About 54 % of the sites have ANC in the 0 to 50 microequivalents per liter range which the EPA classifies as extremely sensitive. Four percent of the sites are classified as acidic (ANC < 0). Only three percent of the sites have ANC above 200 microequivalents. These low ANC conditions raise concerns that GRSM waters are susceptible to very low pH in storm events and continued acid deposition.

This thesis uses water quality data to investigate elevational trends. Significant (p

Stepwise multiple linear regression was used to model pH, ANC, nitrate and sulfate. This thesis incorporates basin characteristics, time, acid deposition data, USGS stream flow data as surrogate hydrologic data, and precipitation data, e.g., inches of rain on preceding days, to determine whether these variables are associated with water quality. Each of these variables were found to be statistically significant (p

This thesis also analyzes temporal trends in pH. Modeling reveals statistically significant decreasing trends in pH with time. If conditions remain the same and past trends continue, models suggest that 30.0 % of the sampling sites will reach pH values less than 6.0 in less than 10 years, 63.3 % of the sites will reach pH values less than 6.0 in less than 25 years, and 96.7 % of the sites will reach pH values less than 6.0 in less than 50 years. The models used to predict future pH values explain around 70 % of the variability in the data.