Identifier

etd-10042015-152131

Degree

Master of Science (MS)

Department

Civil and Environmental Engineering

Document Type

Thesis

Abstract

Stream wise 1-dimensional numerical modelling couples well with long term, large domain physical modelling because of its ability to perform simulations quickly. The downsides are limitations in replicating some complex hydrodynamics and sediment transport behavior. In this study, a 1-dimensional model of a section of the Expanded Small Scale Physical Model (ESSPM) is developed using the USACE HEC-RAS software with a goal of investigating the ability of a 1-dimensional model to accurately replicate hydraulics at ESSPM time scales. Additional simulations are conducted to examine the impact of varying distortion scales, non-frictional energy losses and synthetic sediment on hydraulic behavior. The ultimate goal of this work is assess the potential advantages and limitations of 1-D numerical modelling in capturing the hydraulics in small scale physical models of channelized riverine systems. big picture has seen significant changes (or may see) in terms of sediment diversions, flood management, etc. The original bathymetry and topography utilized is at the ESSPM scale; i.e., 1:6000 horizontal and 1:400 vertical. First, the model was calibrated and validated to a series of steady and unsteady physical model experiments performed in the “guinea pig” model. Results indicate HEC-RAS is capable reproducing water surface profiles at ESSPM scale. The numerical model was then scaled to prototype size and a distortion of 7.5 (D7.5 = 1:3000H; 1:400V) and 1 (D1 = 1:400H; 1:400V) to assess the influence that distortion and scaling have on hydraulic behavior. Comparisons of water surfaces between measured values and distorted numerical models indicate roughness scaling is necessary, especially at D7.5 and D1. Models with scaled roughness show that HEC-RAS can accurately reproduce the water surface profiles for a range of distortion scales. Furthermore, velocity comparisons between 1) measured data 2) the 1-D numerical model and 3) a similar 3-D numerical model suggest accurate longitudinal ESSPM velocity predictions can be achieved with HEC-RAS, which is more favorable to cumbersome 3-D model. The lack of spatial difference in velocity suggests model types of a higher order should be used to capture more detail in velocity and sediment patterns. However, the model suggests that certain hydraulic behaviors may be accurately reproduced, providing a useful tool for predicting the big picture outcomes of changes in Mississippi River management.

Date

2015

Document Availability at the Time of Submission

Release the entire work immediately for access worldwide.

Committee Chair

Willson, Clinton

DOI

10.31390/gradschool_theses.3786

Share

COinS