Date of Award

January 2023

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Civil Engineering

First Advisor

Yeo H. Lim

Abstract

Analysis of streamflow levels is critical to flood protection and management. Computer modeling software such as the U.S. Army Corps of Engineers Hydraulic Engineering Center River Analysis System (HEC-RAS) is used to calculate water surface elevations for statistical or historical flows. A key element of modeling river hydraulics is the development of one-dimensional (1-D) cross sections or two-dimensional (2-D) terrain from elevation data. Airborne LiDAR (Light Detection and Ranging) is now commonly used to collect high-density digital elevation model (DEM) data. Models developed from near-infrared LiDAR DEMs alone however, do not have bathymetry since the LiDAR pulses are adversely affected by water and are not representative of the channel bottom. If the LiDAR data is collected during a low flow condition, then the amount of missing bathymetry is smaller, but missing bathymetry would conversely increase with higher flow conditions at LiDAR acquisition. The effect of near-infrared LiDAR on water also results in triangular surface artifacts that can affect hydraulic model results. The objective of this study was to use an available LiDAR dataset for a portion of the Tongue River in the area of Cavalier, North Dakota and create a synthetic cross section to replace the missing bathymetry below the LiDAR water surface. In this study, LiDAR data was downloaded from the State of North Dakota Department of Water Resources. A steady flow representing the flow at the time of LiDAR acquisition and applied above the LiDAR surface was used to estimate the flow area of the missing bathymetry. Prior to flow calculation, the triangular artifacts were manually removed by hydro-flattening the LiDAR bed slope and cross sections. A trapezoidal shape was used to represent the synthetic bathymetry. HEC-RAS models were run with the hydrograph for a historical 2013 flood flow at the USGS Akra Gage on the Tongue River. The rating curves were plotted for the USGS Gage at Akra, an existing calibrated HEC-RAS model with surveyed and interpolated bathymetry, the LiDAR model and the trapezoidal synthetic bathymetry model. The root mean square error (RMSE) of the model results from the USGS gage were calculated. The LiDAR RMSE of 1.9349 feet was proportionately higher as expected. The RMSE of the existing calibrated model and the trapezoidal synthetic bathymetry model at 0.4556 feet and 0.6647 feet respectively were significantly close to each other and close to the USGS Gage. This indicated that the trapezoidal synthetic bathymetry results were a good fit to both the existing calibrated model and the USGS gage. More study is needed to simplify or automate the process of generating synthetic bathymetry.

Download

Share

COinS