WAVE-INDUCED SEDIMENT TRANSPORT: SIMULATION OF TURBULENT WAVE BOUNDARY LAYERS
Two manuscripts are presented which develop a numerical method for studying boundary layer flow and sediment transport on the seafloor. Inviscid flow, either through an analytic solution or a numerical wavetank (NWT) is used to force a previously developed three-dimensional Navier-Stokes model. The resulting hybrid model is able to simulate complex turbulent flows near the ocean bottom or around obstacles.
The first manuscript reports on developments of a perturbation approach to large-eddy simulations (LES) of wave-induced boundary layers. In the present formulation, the total velocity and pressure fields are expressed as the sum of irrotational and near-field viscous perturbation, where the irrotational field is known a priori. The LES equations are formulated and solved for the perturbation fields only, which are forced by the known incident fields. Results are presented for laminar oscillatory boundary layers, as well as for laminar steady streaming induced by small-amplitude waves, which show convergence to known analytic solutions. To demonstrate potential applications, forcing from a two-dimensional NWT is applied, showing the steady streaming that exists in a laminar boundary layer under large-amplitude water waves. Results are also shown for turbulent oscillatory boundary layers, which agree well with published experimental data.
The second manuscript presents LES results of sediment transport over vortex ripples. A conformal mapping is used to match the computational domain to an experimentally derived shape of vortex ripples formed in a large-scale oscillatory water tunnel. While the instantaneous velocity field and time-averaged velocity field agrees reasonably well with published experimental data, the time- and ripple-averaged velocity profile differs substantially. As well, the suspended sediment concentration above the ripple crest is substantially different than that observed experimentally. These effects, likely the result of insufficient resolved turbulent intensity in the LES, result in poor predictions of suspended sediment transport rates.