Journal of Hydraulic Engineering

In nature and industry there are many situations in which dense sediment-laden currents are overlain by fluid of density intermediate between that of the bulk suspension and of the interstitial fluid. An experimental investigation of such horizontally spreading turbidity currents at a constant inflow rate has been conducted. If the bulk density of such currents is initially larger than the ambient density, an underflowing gravitational turbidity current results. However, once the sediment in the suspension settles out, the current releases buoyant interstitial fluid through the interface between the current and the overlying ambient fluid. This settling of the sediment and the detrainment of buoyant interstitial fluid decreases the spreading velocity of the turbidity current until the current stops and lifts off the bed. The rising interstitial fluid accumulates at the free surface to form a new buoyant surface overflow. In this study, simple spreading relationships based on experimental observations are presented and measurements of the spreading rate and the location of the liftoff distances of turbidity currents made in a horizontal flume are in good agreement with the predictions from these relationships.

Yang's unit stream power formula is modified for the computation of total bed-material load in a sediment-laden river with a high concentration of fine suspended materials. Theoretical and laboratory studies indicate that modification can be made on the values of particle fall velocity, flow viscosity, and relative specific weight without any change of the coefficients used in Yang's original formula. Comparisons between computed results from the modified formula and field indicate that the modified formula can be used to compute total bed-material load in the Yellow River with accuracy.

This paper deals with the mechanical characteristics of saltation of bed-load transport in flowing water. Experiments have been carried out by means of high-speed photography and advanced data-processing techniques. Taking saltation as the mode of bed-load transport, systematic analysis has been made with regard to the forces acting on grains and the variation of their magnitudes along the saltation trajectories to obtain the kinematic and dynamic properties of these grains. Moveover, the relations between magnitudes of saltation parameters and flow intensity Θ are attained from the measured data. Some parameters in Bagnold's theory of bed-load transport are also discussed.

This paper deals with the stochastic characteristics of the saltation of bed-load transport. The trajectories of particles with different specific gravity are measured by high-speed photographic technique, and the experiments on bed-load transport processes also have been conducted. It is concluded that the probability density distributions of dimensionless saltation height and length follow the Γ-distributions and the dimensionless salation velocity follows the Gaussian distribution. Moreover, the probability density distributions of the concentration in the bed-load layer and bed-load transport rate also are formulated, and they follow the Γ-distributions approximately. Measured data in natural rivers and experimental flume agree with these distributions very well.

A mathematical model is formulated to simulate dam-break transient debris flows, in which the debris and water mixture is assumed to be isotropic and continuous. In terms of its viscoelastoplasticity, a three-dimensional (3D) rheologic model for Bingham body is used to describe the relationship between the stress tensor and the rate of strain. Only the deformation caused by the deviatoric stress tensor is taken into account. Under the prior considerations, the governing equations for two-dimensional (2D) depth-averaged transient debris flows are obtained by integrating the equation of continuity and equations of motion over depth for 3D flows. To deal with the transient and high-speed features of dam-break debris flows, the derived governing equations are discretized and solved using the method of characteristics. The model is applied to simulate the dam-break debris flow of April 30, 1986, for the Jinshan debris reservoir in Anhui, China. The debris and water mixtures reach as far as 900 m downstream from the initial tailings dam site, with a depth of 1.7 m near the tailings dam and 1.0 m at 900 m downstream. These model results are in good agreement with the available observations.

The transient effect of battery of injection wells on seawater intrusion into coastal confined aquifers is analyzed using a quasi-three-dimensional areal finite-element model derived based on the sharp interface approach. Linear triangular elements are used to discretize the domain. The validity of the model is tested using the theoretical solutions and experimental results. The model is then used to carry out detailed parametric studies on the efficacy of battery of injection wells in controlling seawater intrusion. Hypothetical cases over a wide practical range are considered for this purpose. In the study, the motion of the interface is tracked for different combinations of the well spacing, injection rate, and the duration. Relationships are established between the key parameters in nondimensional forms, which have wider practical application. The injection wells are found to be effective in controlling the intrusion over a longer period of operation, running into several years. Reduction of intrusion up to 60–90% is achievable under favorable conditions. A specific example is solved to illustrate the usefulness of the results.

A new method based on the depression cone volume concept is presented for the identification of hydraulic parameters of leaky aquifers with additional elastic storage release from the unpumped aquifer. The solution deals with the flow system by first considering the continuity equation based on the unpumped and pumped aquifers and constant pump discharges. The elastic storage coefficients are taken into account individually by relating them to the depression cone volumes. In addition, the unpumped aquifer flow is assumed to be transmitted to the pumped aquifer through the aquitard according to a pseudo steady-state flow law. The resulting first-order ordinary differential equation in terms of the depression cone volume is solved, leading to time-variant unpumped and pumped aquifer discharge expressions. The pumped aquifer is then considered as a confined aquifer with variable discharge to the well. The results are presented as type curves, which reduce to the Theis and Hantush solutions if the storage coefficient of the unpumped aquifer is neglected. The type-straight-line method is developed for hydraulic parameter estimations based on late time-drawdown data. The method is general and includes the De Glee steady-state and Jacob straight-line solutions as special cases. A dimensionless time-drawdown plot is proposed to determine whether the aquifer is leaky or not. Necessary steps in the application of the methodology are presented with actual field data.

Computation of suspended sediment load in open channels relies on the integration of the product of the suspended sediment concentrations and the fluid velocity profile. Sediment in the flow, however, travels at a lower velocity than the fluid. Errors due to the relative velocity between sediment particles and fluid are evaluated based on experimental data. The procedure for calculating the errors in Einstein's method does not involve calculating the actual suspended load; instead, it relies on determining the relative error between the loads by using both the fluid velocity profile and the sediment velocity profile. Although there is an abundance of data on the velocity profiles in open channel flows, with and without sediment in the flow, data on sediment velocity profiles is scarce. The analysis of the available data indicates that Einstein's method for computing the suspended sediment load overestimates the load by about 40, in some cases. Further data collection and subsequent analysis is needed to more accurately formulate the difference in sediment loads based on Einstein's method.