Date of Award

Fall 2024

Project Type

Dissertation

Program or Major

Ocean Engineering

Degree Name

Doctor of Philosophy

First Advisor

Diane L Foster

Second Advisor

Nathan J M Laxague

Third Advisor

Thomas C Lippmann

Abstract

In nearshore environments, small-scale bedforms, or ripples, are important contributors to net morphologic change by means of bedload and suspended load sediment transport. In this dissertation, new characterizations of boundary layer and sediment dynamics are made using high-resolution observations of mobile, rippled beds subjected to oscillatory flow for the purpose of evaluating existing parameterizations of bedload transport. The spatiotemporal distribution of shear stress at and within a mobile, rippled bed is obtained via the derivation and application of a new Momentum Integral Method (“MIM”) expression for bed shear stress to data from a two-phase Large Eddy Simulation and Discrete Particle Model (LES-DPM). The MIM expression makes no assumptions about the shape of the boundary layer, and is shown to be comparable to within 2% of direct estimates of local stress made with the LES-DPM simulation. New findings reveal that the shear stress changes sign between the top and bottom of the mobile layer of grains and that the magnitude is greatly reduced. The depth-averaged stress within the mobile layer, θML, is approximately 50% of the stress at the top of the mobile layer. Characterizations of mobile layer thickness and bedload flux using θML are shown to be more accurate when compared to the critical Shields parameter than the stress taken at the top of the mobile layer alone. This is further supported by particle image velocimetry (PIV) observations collected in a large-scale oscillatory water-sediment tunnel, whereby observed motion is in good agreement with expected motion based on instances where shear stress on the surface of the bedform exceeds twice the traditional critical Shields parameter.

Peak bed shear stress and bedload transport are shown to lead the free-stream velocity by up to 45◦, which is in contrast to common quadratic stress law parameterizations that assume bed shear stress is in phase with free-stream flow. If the phase separation between the free-stream and near-bed flow is known, however, this dissertation illustrates that estimates of bedload flux using the existing semi-empirical method of Meyer-Peter and Müller (1948) (“M-P&M”) are fairly accurate if either shear stress estimates are available over the entire ripple wavelength, or if a single point-estimate of shear stress is available at the ripple crest. M-P&M estimates of flux using single-point estimates of stress from elsewhere on the ripple wavelength, however, are shown to grow less accurate with increasing distance from the crest.

Finally, this dissertation investigates the role of vortex dynamics in bedload transport by examining near-bed flow and boundary layer characteristics across 21 independent experiments in a large-scale oscillatory water-sediment tunnel. The formation of a lee slope vortex is shown to cause a second peak in bed shear stress and mobile layer thickness to occur during each half-oscillation period in addition to the primary peak driven by fluid acceleration. As the relative magnitude of free-stream kinetic energy increases, the vortex becomes stronger and more concentrated, causing an increase in spatiotemporal extent of observed bedform motion. The responsiveness of the bedform to vortex-induced modifications in bed shear stress ultimately suggests that vortices play an important part in bedload transport in addition to suspended load transport over orbital-scale vortex ripples.

Share

COinS