Doctoral Dissertations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy


Civil Engineering

Major Professor

John S. Schwartz

Committee Members

Glenn A. Tootle, Daniel C. Yoder, Devon M. Burr


Estimates of bedload transport rates developed from existing transport models are notoriously inaccurate(Wilcock 2001). The gravel bed models addressed in this study include the Meyer-Peter and Muller; Parker, Klingeman, and McLean; and Wilcock two-fraction models. The question of whether or not these models predict bedload transport rates in a Southern Appalachian Ridge and Valley stream is complicated by the fact that these models have only been previously assessed in terms of their agreement with bedload transport rates measured in the Western regions of the U.S. Further, due to the strongly non-linear form of bedload transport models discrete errors and cumulative uncertainty in input parameters can result in excessive error and uncertainty in results.

The research presented in this dissertation approaches these issues through introduction of a new bedload transport data set collected on Little Turkey Creek in Farragut, Tennessee using a continuously monitoring bedload collection station with estimated collection efficiencies of nearly 100%. Use of 20-liter pail pit samplers is addressed for estimating bedload particle size distributions and transport model calibration. Finally, the issue of error and uncertainty in model input parameters is addressed through evaluation of the results of discrete error and cumulative uncertainty within the region of observed variation in bedload transport observations.

The results of this research suggest similarity between bedload transport characteristics in Southern Appalachian Ridge and Valley streams and those of streams in the Western region of the U.S. It was found that 20-liter pail pit traps are suitable for collection of bedload transport particle size distribution data and only marginally well suited for model calibration. It was illustrated that selected bedload transport models are most sensitive to errors in estimates of Manning’s n and slope. Further, it was found that uniform uncertainty of more than 20% in model input parameters produces results that are at the outer edge of the observed variation in bedload transport rates. The body of work presented in this dissertation is intended to provide stream restoration design professionals with additional background to inform bedload transport estimates on streams in the Southern Appalachian Ridge and Valley Region.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."