Journal of Hydraulic Engineering

The Third International Symposium on Stratified Flows was convened on February 3–5, 1987, in Pasadena, California. One hundred and seventy‐six engineers and scientists representing 14 countries attended and presented papers on internal waves and their stability, blocking and intrusion, topographic effects, rotating stratified flows, turbulent stratified shear flows, mixed‐layer dynamics and convection, frontal phenomena and gravity currents, plumes, and thermals. New results were also presented on direct numerical solution of the Navier‐Stokes equations. This report provides a brief summary of the meeting presentations and comments upon the implications for the civil engineering profession.

In this paper, a slope stability analysis for steep banks is used in conjunction with a method to calculate lateral erosion distance, to predict bank stability response to lateral erosion or bed degradation. The failure plane angle, failure block width, and volume of failed material per unit channel length may be calculated for the critical case. These parameters define the bank geometry following failure and form the starting point for subsequent analyses. The calculation procedure is illustrated by a worked example. Following mass failure slump, debris accumulates at the bank toe. The debris is removed by lateral erosion prior to further oversteepening or degradation generating further mass failures. Any process‐based model for channel width adjustment must account for the combined effects of lateral erosion and mass instability in producing bank instability. The approach adopted here represents a marked improvement over earlier work, which does not account for changes in bank geometry due to lateral erosion prior to mass failure. The engineering applications are presented in a companion paper.

Bank retreat occurs by a combination of lateral erosion by the flow and mass failure under gravity. A new analysis of bank erosion and failure is developed, using a critical shear‐stress concept to account for lateral erosion and a slope stability criterion for mass failure. In this paper, we apply the analysis to two problems of bank retreat often encountered by practicing engineers dealing with alluvial channels. The first application is to the prediction of degradation downstream of a dam for the case in which bed lowering causes bank instability. We show that rapid bank retreat can occur once the threshold height for mass failure of the banks is reached. This supplies sediment to the flow, tends to limit the depth of degradation, and drives complex response downstream. The second application is to the modeling of flow in channel bends and the prediction of the equilibrium cross section. We show that scour depth at the outer bank may be limited by the critical bank height. If scouring causes the outer bank to fail, then the channel will migrate laterally rather than scour down vertically. The analysis presented here can be used to predict the equilibrium cross section and migration rate incorporating bank stability considerations. It could also be used to predict the likely increase in scour depth resulting from outer bank stabilization in a bendway.

The transport and fate of pollutants in groundwater is an important environmental concern. Adsorption processes, such as cation exchange, affect the movement of many contaminants. Accurate modeling of these processes is necessary for the design of efficient aquifer restoration methods. That accuracy depends on both the mathematical description of the transport process and the quantification of the effective adsorption parameters. The objectives of this study were: (1) to formulate a procedure for estimating effective aquifer adsorption parameters in situ, and (2) to test the procedure in controlled laboratory experiments. The parameter estimation procedure used a groundwater contaminant transport model combined with a nonlinear parameter estimation technique. Two‐ and three‐component cation exchange experiments were performed in a two‐dimensional injection‐production flow apparatus. The parameter estimation procedure computed effective exchange parameters that were consistent for all of the experiments.

This paper describes a technique for the solution of steady state flows in a multi‐quality water distribution system. The model presented here is capable of solving for the heads, flows, and salt or ion concentrations in looped networks. It uses the concept of mixing various qualities of water within the same network to obtain the desired qualities at the consumer outlets. This method offers an alternative to the expensive dual or multiple water distribution system. The model is based on the nodal equations and the user may specify the head, the flow, or the salinity at the consumer outlets. Unlike conventional network solvers, this model reformulates parts of the equations due to changes in flow direction. It is therefore capable of simulating solute flow in complex looped networks where the flow direction is not known prior to the solution of the governing set of nonlinear equations. A case study for a rural water distribution system is given to demonstrate the feasibility of application of such a model to practical cases.

The standard kinematic wave (KW) method used in many models of open channel flow routing of runoff hydrographs in watershed models is examined as to the significance of the computational errors due to numerical‐diffusion and the selection of computational effort. It is shown that a wide range of modeling results are possible from a KW model depending on the choice of computational reach length and time‐step size used in the KW approximation. In comparison, the simple convex hydrologic routing method demonstrates only a small fraction of the variation in results demonstrated by the KW model. It is recommended that use of the KW method for channel routing in watershed models be reconsidered.

When the flow rate from a port discharging buoyant fluid into a denser environment is less than a critical value, a counter flow of dense fluid will become established through the port. Experiments were conducted at near prototype scales which confirmed that the critical rate of discharge could be expressed in terms of a densimetric Froude number based on the port diameter and the buoyancy of the discharge. The value of the critical Froude number was found to depend primarily on the port orientation and shape.