Direct Numerical Simulation of Oscillatory Flow Over a Wavy, Rough, and Permeable Bottom

Abstract

The results of a direct numerical simulation of oscillatory flow over a wavy bottom composed of different layers of spherical particles are described. The amplitude of wavy bottom is much smaller in scale than typical bed forms such as sand ripples. The spherical particles are packed in such a way to reproduce a bottom profile observed during an experiment conducted in a laboratory flow tunnel with well‐sorted coarse sand. The amplitude and period of the external forcing flow as well as the size of the particles are set equal to the experimental values and the computed velocity field is compared with the measured velocity profiles. The direct numerical simulation allows for the evaluation of quantities, which are difficult to measure in a laboratory experiment (e.g., vorticity, seepage flow velocity, and hydrodynamic force acting on sediment particles). In particular, attention is focused on the coherent vortex structures generated by the vorticity shed by both the spherical particles and the bottom waviness. Results show that the wavy bottom triggers transition to turbulence. Moreover, the forces acting on the spherical particles are computed to investigate the mechanisms through which they are possibly mobilized by the oscillatory flow. It was found that forces capable of mobilizing surface particles are strongly correlated with the particle position above the mean bed elevation and the passage of coherent vortices above them.

Interface-resolved direct numerical simulations of sediment transport in a turbulent oscillatory boundary layer

“…figure 4), it is reasonable to suppose that the sediment flow rate is closely related to the bed shear stress. Figure 6 Three contributions to the value of τ b appearing in figures 6 and 7 can be identified (Uhlmann, 2008;Mazzuoli et al, 2018): (i) the contribution due to the viscous stress (τ visc ), (ii) the contribution due to the turbulent stress (τ turb ), (iii) the contribution due to the flow-particle interactions (τ part ). The procedure used to compute the different contributions is described in more details by Mazzuoli et al (2018) and .…”

“…The oscillations of the force per unit area are defined "small" when compared with the oscillations which are observed when the value of τ b is averaged over a much smaller horizontal surface. The value τb of σ 12 averaged over a portion of the instantaneous bottom surface which is 4 δ * long, 2 δ * wide and centered around the point ( Three contributions to the value of τ b appearing in figures 6 and 7 can be identified (Uhlmann, 2008;Mazzuoli et al, 2018): (i) the contribution due to the viscous stress (τ visc ), (ii) the contribution due to the turbulent stress (τ turb ), (iii) the contribution due to the flow-particle interactions (τ part ). The procedure used to compute the different contributions is described in more details by Mazzuoli et al (2018) and .…”

“…The value τb of σ 12 averaged over a portion of the instantaneous bottom surface which is 4 δ * long, 2 δ * wide and centered around the point ( Three contributions to the value of τ b appearing in figures 6 and 7 can be identified (Uhlmann, 2008;Mazzuoli et al, 2018): (i) the contribution due to the viscous stress (τ visc ), (ii) the contribution due to the turbulent stress (τ turb ), (iii) the contribution due to the flow-particle interactions (τ part ). The procedure used to compute the different contributions is described in more details by Mazzuoli et al (2018) and . The time development of τ b , for the run 2 (fix) qualitatively agrees with that measured by Jensen et al (1989) (see figure 1) even though the Reynolds number of the laboratory experiment is somewhat different from that of the numerical simulation but, more importantly, the bottom of the experimental apparatus was smooth.…”

Interaction of Very Large Scale Motion of Coherent Structures with Sediment Particle Exposure

“…Hydrodynamic forces, in particular drag and lift, acting on a sediment particle in turbulent flows have been extensively studied through experiments [18][19][20][21][22][23][24][25][26][27] and numerical simulations [28][29][30][31][32][33][34]. Recent investigations applied vortex-detection methods in order to identify the coherent structures that are responsible for the generation of hydrodynamic forces.…”

Sediment transport under oscillatory flows

“…Recently, Mazzuoli et al (2016Mazzuoli et al ( , 2018Mazzuoli et al ( , 2019Mazzuoli et al ( , 2020 have made direct numerical simulations (DNS) of the oscillatory flow generated by surface waves close to a bottom made up of spherical particles mimicking sand grains (both fixed (Mazzuoli et al, 2018) and mobile (Mazzuoli et al, 2016(Mazzuoli et al, , 2019(Mazzuoli et al, , 2020 particles). In particular, the approach employed by Mazzuoli et al (2016Mazzuoli et al ( , 2019Mazzuoli et al ( , 2020 evaluates the flow around the moving particles, the dynamics of which is explicitly evaluated by means of the numerical simulation of the Newton-Euler equations describing the translation and rotation of a rigid body.…”