UVM Theses and Dissertations
Format:
Print
Author:
Lefever, Danica W.
Dept./Program:
Civil and Environmental Engineering
Year:
2006
Degree:
MS
Abstract:
Streambank retreat (SBR) is a complex process that is both temporally and spatially variable. Many attempts have been made to evaluate the extent of SBR through the use of streambank stability models and, in the State of Vermont, Rapid Geomorphic Assessments (RGAs). It is important to be able to identify stream reaches that are most susceptible to SBR in order to mitigate sediment and nutrient inputs to the stream from eroded streambank material. Bank retreat was measured using repeated cross-section surveys over three years for 8 sites and 15 cross-sections along the LaPlatte River and Lewis Creek in the Champlain Valley of Vermont, USA. The measured SBR rates were compared to results of two bank stability models, the Bank Stability Model (BSM) and SLOPEIW, as well as the resulting RGA scores based on the Vermont Agency of Natural Resources' protocols.
The SBR rates ranged from -0.15 to 0.43 m/yr, and geomorphic conditions ranged from fair (0.6) to reference condition (0.9; based on a scale from 0.0 to 1.0). RGA scores were not well correlated with measured rates of SBR. In addition, only the Factor of Safety (Fs) from the BSM model correlated moderately well with measured rates of SBR (correlation coefficient = 0.51; p = 0.05). The RGA scores did, however, correlate with the modeled Fs values, suggesting that the methods are comparable, but neither predicts rates of bank retreat well. Additional tools are needed to identify streambanks in need of mitigation measures to limit sediment and nutrient inputs.
The SBR rates ranged from -0.15 to 0.43 m/yr, and geomorphic conditions ranged from fair (0.6) to reference condition (0.9; based on a scale from 0.0 to 1.0). RGA scores were not well correlated with measured rates of SBR. In addition, only the Factor of Safety (Fs) from the BSM model correlated moderately well with measured rates of SBR (correlation coefficient = 0.51; p = 0.05). The RGA scores did, however, correlate with the modeled Fs values, suggesting that the methods are comparable, but neither predicts rates of bank retreat well. Additional tools are needed to identify streambanks in need of mitigation measures to limit sediment and nutrient inputs.