Session E1: Geophysical: Ocean I

Quantifying Suspended Sediment Diffusion Through Direct \textit{in situ} Measurements of Turbulent Schmidt Number

In this study we investigate how the diffusion of suspended sediment differs from the diffusion of fluid momentum using both laboratory and \textit{in situ} field measurements. The most common model for turbulent diffusion considers eddy diffusivity ($D_{T})$ to be proportional to the eddy viscosity (\textit{$\nu $}$_{T})$ scaled by the turbulent Schmidt number (\textit{$\sigma $}$_{T})$. But accurate selection of \textit{$\sigma $}$_{T}$ values is challenging because sediment, by virtue of its inertia, is transported differently than either momentum or passive scalars. We directly measure \textit{$\sigma $}$_{T}$ over a variety of flow cases using a novel Volumetric Particle Imager (VoPI) developed for this purpose. VoPI is a field-deployable quantitative imaging device that can obtain three-component particle velocity records in a volume. By computing velocity variances and integral timescales from measured Lagrangian velocity records, we compute $D_{T}$ (for particles) and \textit{$\nu $}$_{T}$ (for tracers) directly using Taylor's (1921) formulation. We present the construction and calibration of the device as well as validation of its measurements. We also report the connections between the measured \textit{$\sigma $}$_{T}$ values and the flow conditions in which they occur and suggest predictive methods for when direct measurements are unavailable.   

High-Schmidt-number mass transport mechanisms from a turbulent flow to absorbing sediments

We have investigated the mechanisms involved in dissolved oxygen (DO) transfer from a turbulent flow to an underlying organic sediment bed, populated with DO-absorbing bacteria, relying on the coupling between the bio-geochemistry of the sediment layer and large-eddy simulation for the transport on the water side [Scalo et al., \emph{J. Geophys. Res.}, \emph{117}(C6), 2012]. Time correlations at the sediment-water interface (SWI) show that the diffusive sublayer acts as a de-noising filter with respect to the overlying turbulence; the mass flux is not affected by low-amplitude background fluctuations in the wall-shear stress but, rather, by energetic and coherent near-wall transport events, in agreement with the surface renewal theory. The spatial and temporal distribution of the mass flux is therefore modulated by rapidly evolving near-wall high-speed streaks (associated with intermittent peaks in the wall-shear stress) transporting patches of (rich-in-oxygen) fluid to the edge of the diffusive sublayer, leaving slowly-regenerating elongated patches of positive DO concentration fluctuation and mass flux at the SWI. The sediment surface retains the signature of the overlying turbulent transport over long time scales, allowed by the slow bacterial absorption.   

Sedimentation of porous and solid particles in stratified fluids

Marine aggregates, particles composed of organic and inorganic material in the ocean, are fundamental to marine carbon cycling both in their importance to bacterial remineralization and carbon flux from the surface ocean. Understanding the function of marine aggregates in carbon biogeochemistry requires knowledge of their small scale settling dynamics in different physical environments. We have conducted experiments to study the settling behavior of single solid and porous spheres and natural marine aggregates through sharp vertical density stratification in ambient fluids. Additionally, we have investigated the behavior of particle clouds. In all of these situations, particles demonstrate decreased settling velocity at the density transition which could be brought about by entrainment of less dense fluid from above and/or diffusion-limited retention. By comparing experimental results to models including entrainment and diffusion, we have identified the mechanisms underlying this delayed settling phenomenon. Discussion of the models will be presented.   

Numerical simulation on fine sediment transport in steady and oscillatory boundary layer -- The role of rheology

Turbulence-resolving 3D numerical simulations of fine sediment transport in both steady and oscillatory boundary layers are carried out to study the interplay between turbulence modulation and rheological stress. A high-accurate scheme is developed to resolve all the scales of carrier flow turbulence. Fourier expansions are adopted in both streamwise and spanwise directions. To incorporate both the hindered settling effect and rheology models, a sixth-order compact finite difference scheme is implemented in vertical direction to keep the spectral-like accuracy. A recent numerical study (Ozdemir et al. 2010, J. Fluid Mech.) on fine sediment transport in the wave boundary layer reveals the evolution of transport regimes from a well-mixed sediment distribution, to the formation of lutocline and a complete laminarization of wave boundary layer due to increasing sediment availability and settling velocity. We are motivated to further study the effect of rheology in determining the transition of flow regimes and hydrodynamic dissipation. By including the rheology, simulation results shows that the increased effective viscosity tends to increase the thickness of viscous sub-layer and further reduce the thickness of turbulent regime, which is limited by the lutocline.   

Mechanisms of complete turbulence suppression in turbidity currents

The sustained propagation of turbidity current depends on a tight interplay between suspended sediments and turbulence. This work explores the phenomenon of complete turbulence suppression in a dilute turbidity current due to stratification of suspended sediments. Direct numerical simulations of turbidity currents are carried out to understand the dynamics of complete turbulence suppression. We observe that stratification of sediments leads to damping and spatial redistribution of hairpin and quasi-streamwise turbulent structures in the flow. These turbulent structures are known to be responsible for sustaining turbulence in the flow. We propose that beyond a critical stratification limit, the existing vortical structures in the flow are damped to an extent where they loose their ability to auto-generate subsequent turbulent structures, which ultimately leads to complete loss of turbulence. We also identify three parameters: Reynolds number ($Re_\tau$), Richardson number ($Ri_\tau$) and sediment settling velocity ($V_z$) to control the flow dynamics. Therefore a criteria for complete turbulence suppression can be defined as a critical value for $Ri_\tau V_z$. Based on simulations, experiments and field data, the critical value appears to have logarithmic dependence on $Re_\tau$.