Concentration-Duration-Frequency Curves for Stream Turbidity: Possibilities for Assessing Biological Impairment1

Authors

  • John S. Schwartz,

    1. Respectively, Assistant Professor, University of Tennessee, Department of Civil and Environmental Engineering, Knoxville, Tennessee 37996-2010.
    Search for more papers by this author
  • Melanie Dahle,

    1. Respectively, Assistant Professor, University of Tennessee, Department of Civil and Environmental Engineering, Knoxville, Tennessee 37996-2010.
    Search for more papers by this author
  • R. Bruce Robinson

    1. Professor Emeritus, University of Tennessee, Department of Civil and Environmental Engineering, Knoxville, Tennessee 37996-2010.
    Search for more papers by this author

  • 1

    Paper No. JAWRA-07-0080-P.R2 of the Journal of the American Water Resources Association (JAWRA). Discussions are open until December 1, 2008.

(E-Mail/Schwartz: jschwart@utk.edu)

Abstract

Abstract:  Siltation and subsequent biological impairment is a national problem prompting state regulatory agencies to develop sediment total maximum daily loads (TMDL) for many streams. To support TMDL targets for reduced sediment yield in disturbed watersheds, a critical need exists for stream assessments to identify threshold concentrations of suspended sediment that impact aquatic biota. Because of the episodic nature of stream sediment transport, thresholds should not only be a function of sediment concentration, but also of duration and dose frequency. Water quality sondes can collect voluminous amounts of turbidity data, a surrogate for suspended sediment, at intervals that can be used to characterize concentration, duration, and frequency of elevated turbidity events. To characterize turbidity sonde data in an ecologically relevant manner, a methodology for concentration-duration-frequency (CDF) curves was developed using turbidity doses that relate to different levels of biological impairment. To illustrate this methodology, turbidity CDF curves were generated for two sites on Little Pigeon River in the Great Smoky Mountains National Park, Tennessee, using over 30,000 sonde data measurements per site for a one-year period. Utilizing a Poisson arrival approach, turbidity spikes were analyzed stochastically by observing the frequency and duration of recorded events over a turbidity level that relates to a biological dose response. An exponential equation was used to fit duration and frequency of a specified turbidity level to generate concentric-shaped CDF curves, where at specific turbidities longer durations occurred less frequently and conversely shorter durations occurred more frequently. The significance of the equation fit to the data was accomplished with a Kolmogorov-Smirnov goodness-of-fit test. Our findings showed that the CDF curves derived by an exponential function performed reasonable well, with most curves significant at a 95% confidence level. These CDF curves were then used to demonstrate how they could be used to assess biological impairment, and identify future research needs for improved development of sediment TMDLs.

Ancillary