|
|||||
|
|
pH and Acid Anion Time Trends in Different Elevation Ranges in the Great Smoky Mountains National Park |
|
| Authors: | R. Bruce Robinson Thomas W. Barnett Glenn R. Harwell Stephen E. Moore Matt Kulp John S. Schwartz |
| Affiliation: | 1Armour T. Granger Professor of Civil and Environmental Engineering, 223 Perkins Hall, Univ. of Tennessee, Knoxville, TN 37996 (corresponding author). E-mail: rbr@utk.edu 2Graduate Research Assistant, Dept. of Civil and Environmental Engineering, 223 Perkins Hall, Univ. of Tennessee, Knoxville, TN 37996. E-mail: tbarnett1@utk.edu 3Graduate Research Assistant, Dept. of Civil and Environmental Engineering, 223 Perkins Hall, Univ. of Tennessee, Knoxville, TN 37996. E-mail: gharwell@usgs.gov 4Head Fishery Biologist, U.S. Dept. of Interior, National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN 37738. E-mail: steve_e_moore@nps.gov 5Fishery Biologist, U.S. Dept. of Interior, National Park Service, Great Smoky Mountains National Park, Gatlinburg, TN 37738. E-mail: matt_kulp@nps.gov 6Assistant Professor of Civil and Environmental Engineering, 223 Perkins Hall, Univ. of Tennessee, Knoxville, TN 37996. E-mail: jschwart@utk.edu |
| Abstract: | Quarterly base flow water quality data collected from October, 1993 to November, 2002 at 90 stream sites in the Great Smoky Mountains National Park were used in step-wise multiple linear regression models to analyze pH, acid neutralizing capacity (ANC), and sulfate and nitrate long-term time trends. The potential predictor variables included cumulative Julian day, seasonality, elevation, basin slope, stream order, precipitation, surrogate streamflows, geology, and acid depositional fluxes. Modeling revealed statistically significant decreasing trends in pH and sulfate with time at lower elevations, but generally no long-term time trends in stream nitrate or ANC. The best forecasting models were chosen based on maximizing the r2 of a holdout data set. If conditions remain the same and past trends continue, the forecasting models suggest that 30.0% of the sampling sites will reach pH values less than 6.0 in less than 10?years, 63.3% in less than 25?years, and 96.7% in less than 50?years. The pH forecasting models explain 65% of the variability in the holdout data. |
| Keywords: | Acid rain pH Regression analysis Time series analysis Water quality Monitoring Elevation Mountains |