Journal of Hydraulic Engineering

Erosion tests were run on soft cohesive sediment beds freshly deposited from a concentrated slurry, in a large rotating annular flume. A total of seven tests were carried out with the bed age varying between 1.8 and 18.9 days. A surface type erosion was observed for the applied shear stress range (0.1 to 1 Pa). Following the increase in the water flow, the erosion rate increases quickly to a peak value and then drops off to about zero. The eroded depth was less than 8 mm for all beds, but the resistance to erosion at a given depth and for a given bed structure increased with increasing bed age. This phenomenon was attributed to thixotropy. The erosional strength of a given sediment layer is found to be one order of magnitude lower than its undrained and drained mechanical shear strengths measured with innovative testing methods. Hypotheses are given to explain that difference. However, at constant density and soil-water chemistry, the erosional and mechanical strengths seem to be affected by the same factors, i.e., bed age, structure, and temperature.

The analysis of a laboratory and field data set for sand-bed channels defines the characteristics of upper-regime plane bed with sediment transport. The occurrence of upper-regime plane bed relates to the laminar sublayer thickness δ and depends on shear velocity u*, median grain size d₅₀, and grain shear Reynolds number R*. Two different boundary conditions are recognized: (1) Transition to hydraulically smooth when R* < 11.6; and (2) transition to hydraulically rough when R* > 11.6. In the first case, upper-regime plane bed is obtained when ${\mathit{u}}_{*}^2$≅gδ, where g is the gravitational acceleration. In the second case, upper-regime plane bed is observed when d₅₀≅ 2δ. Temperature effects are possible when R* approaches 11.6, which explains the observations on the Missouri River. Resistance to flow for plane bed with sediment transport increases with the Shields parameter.

A numerical model that uniquely determines the geometry and boundary stress distribution of a stable channel with mobile bed was discussed in the preceding companion paper. The difference between this model and previous stable channel models is that it consistently accounts for the phenomenon of momentum diffusion. In this paper, the model is tested against available experimental and field data. Results indicate that the model adequately predicts the bank shape and significant dimensions of stable channels. To make the findings of the model more amenable to the design engineer, equations and plots based on the model results are developed. These can be used to determine a channel's shape and significant dimensions, giving the practitioner an alternative to running the numerical model. As with the theory behind this model, the design equations and plots apply to straight, symmetrical channels with uniform cross sections and coarse, noncohesive boundaries. Sediment is assumed to be coarse enough that suspension is precluded and bank vegetation and secondary current effects are neglected.

Sand and gravel transport were sampled in flow rates from just above to several times the critical shear stress of the bed material at two sites on Goodwin Creek, Mississippi. Transport rates representative of the channel cross section were calculated from these samples. The standard deviations of the transport rates were approximately equal to the mean for transport rates grouped by bed shear stress. Normalized probability density functions defined for the transport rates were most closely approximated by a gamma form of the probability function. The probable accuracy in the prediction of the time-averaged transport rate for a given number of measured transport rates can be calculated from this information.

A new computer model is presented to predict the spatial and temporal distribution of residual constituent in a pipe network under slowly varying unsteady flow conditions. Unlike the other available models, which use steady-state or extended-period simulation of steady flow conditions, thus neglecting inertial effects, the presented model is truly dynamic, using a lumped-system approach to compute unsteady flow conditions. This model also includes dispersion and constituent decay in pipes. Slowly varying flow conditions are computed by numerically integrating the governing equations by an implicit finite-difference scheme subject to the appropriate boundary conditions. The transport equation is solved to compute the propagation of a constituent with a first-order decay rate. To avoid numerical diffusion, the advection and dispersion are solved in two steps: The Warming-Kutler-Lomax explicit scheme is used to solve pure advection while an explicit scheme is used to calculate dispersion and decay. Complete mixing is assumed at the pipe junctions. The model is applied to two typical pipe networks to simulate the transport and decay of chlorine, and the results are compared with another model which uses the standard extended-period simulation technique. The results are found to be in good agreement at the beginning of the simulation. However, the chlorine concentrations at different nodes in the network differ when the flow becomes more unsteady and when reverse flows occur. The model may be used to analyze the propagation and decay of any substance with a first-order reaction rate.

This paper presents the application of a nonoscillatory second order relaxation scheme to simulation of flow in open channel networks. The approach is based on the Saint Venant equations written in conservative form. Junction flow conditions in the channel network are solved explicitly using the continuity principle at each junction and the characteristic equations. When compared with other models and algorithms, which are based on the Saint Venant equation in either nonconservative or conservative form, the proposed algorithm is of higher accuracy and is applicable to cases where propagation of a shock wave or discontinuity is involved. In this paper we discuss the application of the proposed algorithm to both single channel and channel network problems, both with and without shock waves. Numerical results obtained are presented comparatively with analytical solutions or results obtained from other numerical solutions wherever applicable.

This paper introduces a new design concept for energy dissipation below flow regulators. Instead of using a single-leaf sluice gate, a double-leaf gate design is suggested. The two leaves are arranged in such a way that a jet issues at an angle toward the water surface through the opening between the two leaves. The bottom leaf remains fixed while the top leaf (control leaf) is moved up and down to regulate the flow. By using this arrangement, the high velocity supercritical stream is deflected away from the bed to the water surface. In addition to minimizing erosion, a lifting device of smaller capacity will be adequate for operating the control leaf. The existing regulators can be modified by adding an upward sloping sill just below the gate. The flow patterns, velocity decay, momentum decay, and bed-shear stress for different submergence and offset ratios are studied experimentally. The main characteristics of the deflected jet and the surface jet are also investigated and compared with those of the free and submerged jumps and surface jets. The new dissipator performs well in comparison with free and submerged jumps.

The reduction of leakage is an important objective of the United Kingdom water industry. The inclusion of pressure-dependent leakage terms in network analysis allows the application of formal optimization techniques to identify the most effective means of reducing water losses in distribution systems. The development of an optimization method to minimize leakage in water distribution systems through the most effective settings of flow reduction valves is described. The method shows a significant advantage compared with previously published techniques in terms of robustness and computational efficiency. A particular feature of the approach is the use of an objective function that allows minor violations in the targeted pressure requirements. This allows a much greater improvement in the exceedance of minor pressure requirements than would otherwise be achieved.

Pipeline transport of solids containerized in capsules has several attractive features. As compared to slurry transport the cargo is not wetted or contaminated by the carrier fluid; no mechanism is required to separate the transported material from the fluid other than the use of capsules; and foremost, it requires less power for maintaining the flow. Although considerable research work has appeared in the last three decades, design equations have not been attempted as yet. In this paper the design of a capsule transporting pipeline has been formulated as a geometric programming problem. The solution of the problem yielded the design variables in the form of explicit equations.

Artificial instantaneous discharge of buoyant fluid in ambient waters is usually carried out at the time of maximum cross-flow so that the resulting dilution is optimal. The determination of the characteristics of this problem is important for environmental impact study but has seldom been studied. To investigate the transport, mixing, and dilution of an isolated cloud of buoyant fluid in cross-flow ambient, a three-dimensional buoyancy extended k-ε model is employed. The computed results show that two flow regimes exist. First is the initial dumping regime in which the effect of crosscurrent and the associated turbulence on the longitudinal width and the average dilution is insignificant as compared to the effect of buoyancy on those quantities. The flow and mixing characteristics in this regime are close to those in the case of stagnant ambient. Second is the ultimate regime in which the effect of buoyancy on the longitudinal width and the average dilution is small as compared to the effect of cross-flow on those quantities. In the ultimate regime, the crosscurrent produces ambient turbulence that increases the longitudinal width of the cloud (and hence increases the dilution), but only slightly affects the vertical frontal position of the cloud. Laboratory experiments are also conducted for the verification of the model and both results are in reasonable agreement.