DNS study of turbulence effects on hyporheic mixing in aquatic sediment beds

Abstract:

Particle resolved direct numerical simulations (DNS) are performed to investigate the influence of varying permeability Reynolds number (Re_k, based on friction velocity, particle bed permeability and kinematic viscosity) on hyporheic mixing at sediment-water interface (SWI) in aquatic systems such as the Columbia river basin. Three different Re_k numbers (2.56, 6.6 and 13.0) corresponding to low to high permeability randomly-packed beds are studied. Statistics of mean flow and turbulence are compared to the data from experimental setup of Voermans et al. (J. Fluid Mech., vol. 824, 2017, pp. 413-437). It is found that at the SWI, stream-wise and vertical turbulence intensities, form-induced vertical and shear stress increase with Re_k, consistent with the experimental data. The double averaged (DA) TKE budget is analyzed to quantify the relative importance of different terms in energy transport mechanisms. Proper orthogonal decomposition (POD) is used to identify the primary modes of motion at SWI, and to correlate the large-scale turbulence above the sediment with small scale turbulence within the bed. The anisotropy distribution of Reynolds stress at SWI is studied to analyze the turbulence structures.