Turbulent Structure of Water and Clay Suspensions with Bed Load

The discusser wishes to congratulate the authors for their excellent paper. He would like to discuss some possible drag reduction effects. The discusser has reanalyzed the authors' data assuming quasi-uniform flows at the end of the channel. For the clearwater flow experiments (runs C-1 and C-2), the data suggest an equivalent roughness height k, of the flume (smooth plexiglass) between 0.15 mm and 0.19 mm. With these values, it is possible to compute the clear-water friction factors f for the clay-mud-flow experiments (M-1 and M-2), and to compare these values with the mudflow friction factor f,. It is interesting to note that the clay-mud flow (runs M-1 and M2) exhibits smaller flow resistance f, than the clear-water flow: i.e. f)f < 1 (Fig. 18). Other researchers observed similarly some drag reduction caused by the presence of suspended sediments (Table 4). Fig. 18 presents model and prototype data of friction factor reduction as a function of the mean volumetric sediment concentration C,,. Most data were obtained with suspended sediments without depositing material. The data of Buckley (1923) must be considered with great care as the changes in friction factor due to variation in bed configuration might be important. Despite earlier controversies, several researchers, including the authors, observed a logarithmic velocity distribution in the inner flow region and a viscous sublayer. Chanson and Qiao (1994) suggest that the presence of sediment particles in the flow layers next to the bottom increases the density and the viscosity of the flow, and induces a thickening of the sublayer and a reduction of bottom shear stress. By analogy with dilute polymer solutions, an increase of the viscosity in the flow layers next to the boundary might explain the observed drag reduction in suspended particle flows. It must be emphasized however that drag reduction in suspended sediment flows is observed only: (1) for starved-bed flows or rising-flood flows (i.e. with no sediment deposition); and (2) with microparticles. An increase of friction is indeed observed with large particle sizes. Rashidi et al. (1990) investigated particularly the effects of particle size, density, and concentration. Their results indicate that the particle density has little effect, but the particle size is an important parameter. Large particles (d = 1.1 mm) cause an increase in the number of turbulent bursts, an increase of Reynolds stresses and larger friction losses. But small particles (d = 0.088 mm) bring about a decrease in the number of wall ejections, Reynolds stresses and friction losses. And these effects are enhanced by the particle concentration. The authors stated, "The measured rate of bed-load transport was considerably greater in mud flow than it was in clearwater flow for the same energy gradient." The discusser wonders if this observation would in fact be caused by a possible drag reduction, i.e., mudflows exhibit smaller flow resistance than clear water. The resulting increase of flow velocity might increase the bed-load transport. The subject is open to discussion.