A numerical investigation of turbulent coherent structures and sediment transport under waves in shallow coastal zone

Date
2018
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Understanding sediment transport driven by surface waves is key to a resilient coastal community and its sustainable development. To understand nearshore wave, turbulence and wave-driven sediment transport processes, this study focuses on utilizing a turbulence-resolving simulation approach and a multiphase flow methodology. ☐ The investigations of turbulent coherent structures (TCS) were conducted using a 3D large eddy simulation (LES) in the swash zone and shallow nearshore region. The LES model was first applied to simulate swash flow for a single dam-break flow over a slope in order to match the field-scale Reynolds number in a smaller laboratory flume. Two distinct turbulence characteristics were identified for each stage, i.e., quasi-2D turbulence during uprush due to swash bore and near-bed generated turbulence during backwash due to bottom boundary layer. Moreover, it was found that the near-bed finger patterns during backwash are due to boundary layer instabilities injected vertically from the bed. The 3D LES model with more sophisticated dynamic/unstructured mesh was further used to investigate the long wave-induced TCSs (or so-called whirlpools), and their impact to coastal areas, which are often observed in the nearshore during and right after tsunami events. The long wave-induced TCSs are also of quasi-2D turbulence due to the limited water depth and are quasi-homogeneous in azimuthal direction with the origin following the center of the vortex. However, unlike the horizontal flow structure, the growth and decay of long wave-induced TCSs are vertically inhomogeneous and intermittent particularly due to the interaction with the bathymetry and solid boundaries (e.g., a breakwater). In both LES studies, it was found that a simple balance between turbulence production and turbulent dissipation rate is violated when TCSs prevail. ☐ To study sediment transport driven by realistic surface waves, a free surface resolving two-phase sediment transport model, called SedWaveFoam was developed. SedWaveFoam is able to concurrently resolve the free surface wave field, bottom boundary layer, and sediment transport processes in a single numerical modeling framework using OpenFOAM. At present, the full profiles of flow and sediment transport driven by waves are solved using Reynolds-averaged Eulerian two-phase flow equations with closures of inter-granular stresses and a k-ε turbulence model. The SedWaveFoam was first validated with a sheet flow dataset under monochromatic nonbreaking nonlinear shallow water waves measured in a large wave flume. The mechanism enhancing net onshore sediment transport under the free surface waves is revealed as a nonlinear wave-stirring effect. The major assumptions used in the conventional parameterization for sediment transport driven by progressive wave streaming were evaluated and new intra-wave and wave-averaged parameterizations were proposed. The SedWaveFoam was further applied to understand sediment transport driven by near-breaking waves with large velocity skewness and acceleration skewness. Comparing SedWaveFoam results with 1DV model results without free surface effects, it was found that large acceleration skewness significantly enhances net onshore sediment transport together with progressive wave streaming. This onshore sediment transport is associated with momentary bed failure and near-bed shear instabilities due to large horizontal pressure gradient that cannot be resolved by 1DV models. Conventional parameterizations for progressive wave streaming need to be revised for waves with large acceleration skewness. ☐ For the future work, the evolution of breaking waves in the inner-surf zone (e.g., landward of a surf zone sandbar) is needed to reveal complex turbulence characteristics and its impact on sediment transport. A preliminary model inter-comparison suggests that LES produces stable turbulent kinetic energy (TKE) distribution more consistent with measured data while the k-ε turbulence model produces a large spreading rate, which results in an unrealistic offshore advection of TKE from the inner-surf zone.
Description
Keywords
Citation