Near-bed, high-density suspensions (HDS), or fluid muds as they are sometimes called, have been observed to be dominant mode of cross-shore sediment transport on a number of continental margins (shown in white in the photograph below). These flows result from the generation of a concentrated, near-bed suspension of fine sediment. In response to the need for experimental data on these unusual flows, we constructed a facility to recreate them in the laboratory. The facility is a U-tube, where a piston forces oscillatory motions in a duct (schematic). The facility can produce wave orbital motions of nearly 1 m/s on a tilting platform (in excess of 10 degrees).
As a part of his MS thesis, Mike Lamb was able to produce steady-state HDS in the new facility using crushed silicate with a mean grain size of about 20 microns (i.e., silt). Preliminary experiments attempting to investigate the interaction of a dense bottom saline layer underneath waves failed to produce a steady state. It seems that only the difference between the dense salty layer and the sediment layer was the finite settling velocity of the particles. That is, the settling flux into the concentrated bottom boundary layer in the sediment case was sufficient to maintain the high concentrations. In fact, the settling was most likely enhanced because the HDS were enriched with coarser material. The sand concentrations were often five times greater in the HDS than the substrate. A grain-size profile of a typical experiment is shown in the figure below.
Another interesting finding was that the boundary layers we observed seemed to shrink or even disappear, when sediment concentrations became significant (the trend in the figure below). With a new acoustic Doppler velocimeter (a SonTek microADV) and a technique to separate the waves from the turbulence, the turbulent dissipation and production rates were then measured. Then from the turbulent kinetic energy budget, we were able to demonstrate that most of the turbulence was transported there from an extremely small region (< 0.3 mm) near the bed. Future work with more extensive instrumentation will be required to definitively test this hypothesis by measuring directly the transport of turbulent energy. In the meantime, we have submitted a manuscript on the flow structure of HDS to Journal of Geophysical Research.
In addition to the interesting results of our analysis of the water column produced, intriguing bedforms were also found. We did not expect to find bedforms in material that was dominantly silty because silt does not generally participate in saltation (a necessary precursor for bedform formation). However, like in some unpublished photographs from the Eel River margin, ripples were often found in our silty material. Some of these deposits were locally enriched in the sand that was concentrated in the HDS (darker material in the photograph below), but for the most part, the substrate consisted predominantly of the fine-grained material. The spacing of the ripples allowed us to identify them as anorbital ripples. Anorbital ripples are extremely difficult to produce in the laboratory owing to the dissimilarity in scaling of the wave orbital diameter and grain size for shallow (< 5 m) surface gravity waves. Because our facility is entirely closed, we are able to produce waves that had orbital diameters much greater than the depth of the tank. The geological ramifications of these preliminary experiments were summarized in a manuscript that has been submitted to Journal of Sedimentary Research.
New experiments are planned for the fall of 2004 to better characterize HDS. In particular, we are attempting to measure the gravitational flux of sediment when the bed is sloping. Our measurements will be used to test the Wright et al. (2000) model, which can predict gravitational flux of sediment given wave conditions and slope. We also intend to formulate new analytical tools that predict the bed shear stress in the presence of heightened sediment concentrations.