Date of Award

January 2018

Document Type


Degree Name

Master of Science (MS)


Civil Engineering

First Advisor

Yeo H. Lim


The Tongue River in Cavalier has experienced severe erosion and bank failures in recent years. Two homes have been evacuated and demolished, and another lost ten ft of their yard overnight when a tree slumped into the river. The city lies downstream of Renwick Dam and nine other dams upstream of it which have greatly reduced the average flows through town. The resulting lowered water surface is a probable source of the streambank instability. A series of rock weirs was proposed to be installed in town to raise the minimum water surface elevation and potentially provide benefit to all problem sites identified in town. The existing conditions and several potential weir locations were analyzed using the Hydrologic Engineering Center River Analysis System (HEC-RAS) sediment transport analysis. The Park Street weir location was chosen as optimal for maximizing benefit at all problem sites, minimizing scour, and addressing the most urgent needs.

The Red River Riparian Project team originally identified twenty problem sites in town that needed attention. Since then, several of the locations have installed projects to protect the streambanks. Three potential weir locations were identified for this study: Woodland Terrace, Park Street, and Division Ave. The Woodland Terrace site is immediately downstream of the last original problem site. However, that site and the next upstream of it have had projects constructed. The Park Street weir is immediately downstream of the last problem site that has had no remediation. The Division Ave location is closer to the area where more severe erosion was occurring and houses needed to be removed.

The existing conditions unsteady flow model geometry was created using LiDAR, survey, and other available data. Hydrology was developed with a U.S. Army Corps of Engineers (USACE) Hydrologic Engineering Center Hydrologic Modeling System (HEC-HMS) model of the Pembina River watershed and U.S. Geological Survey (USGS) gage data. Then the unsteady flow model was calibrated to 2013 flow and elevation data. The sediment transport model used the gradation data collected at the time of survey, USGS gage data, and several assumed parameters. The results do not precisely model actual erosion and deposition depths due to these assumptions and some software limitations. This study compares potential project impacts from each weir location. The parameters stay the same across all models and a preferred option can be selected by a relative comparison. The results can be verified with more detailed data and modeling in the future.