Journal of Hydraulic Engineering

Forum papers are thought-provoking opinion pieces or essays founded in fact, sometimes containing speculation, on a civil engineering topic of general interest and relevance to the readership of the journal. The views expressed in this Forum article do not necessarily reflect the views of ASCE or the Editorial Board of the journal.

New models for the prediction of laminar and turbulent flow of non-Newtonian fluids in open channels of different cross-sectional shapes are proposed. These models are compared with three previously published models for laminar flow and five previously published models for turbulent flow using a recently extended experimental database for non-Newtonian flow in open channels of different cross-sectional shapes. Flow of three different fluids (aqueous carboxymethylcellulose solutions and kaolin and bentonite suspensions) was investigated in open channels of four different cross sections (rectangular, semicircular, trapezoidal, and triangular) at slopes varying from 1 to 5°. The new laminar model gave a closer fit to the data than those from the previously published models. However, the presence of the yield stress still presents a problem that makes flow prediction in laminar flow for such fluids not very accurate. The new model for turbulent flow gave the best fit to the flow data compared with the five previously published models, which fell within a 30% error margin. A particular advantage of both new models is that they are applicable for all the different fluids tested and the four open channel cross-sectional shapes.

Levee failure due to overtopping during floods is a common phenomena. An experimental study was conducted in the hydraulics laboratory, University of South Carolina, to develop an understanding of the effects of soil properties on the breaching process of earthen levees. A total of eight experiments were conducted. A new methodology to measure the subaqueous breach evolution has been developed and validated. Measurements of breach width evolution and headcut migration are shown to be related to soil properties. Inclusion of noncohesive, silt size crushed silica in the levee material (45% by weight) significantly reduced the peak discharge and breach evolution time compared to a levee constructed with sand alone. Increase in kaolin clay in the levee material led to an increase in unconfined compressive strength which in turn increased resistance to levee failure. Nondimensional equations relating the breach width evolution and headcut migration to the soil properties have been developed by performing a multivariate analysis of the experimental data.

The method of characteristics (MOC) with the limitation of Courant’s stability condition is widely used in simulation of unsteady flow in a pipeline. However, the relatively complex method of implicit (MOI) provides the advantages of unconditional convergence and mutual independence between time and space mesh parameters. This study combines the MOC and MOI to simulate pipeline unsteady flow and hydropower transient processes. The boundary conditions for the coupled method are introduced and validated by simulating the water hammer in uniform and variable area duct, and the water-level fluctuation in a surge tank. Subsequently, the coupled methods are applied to study the transient processes in two hydropower stations: one is to determine the water-level fluctuation in the surge tank and investigate the effect of water inertia in the connecting pipe on the water hammer pressure, and the other is to determine the solution for the water hammer in a variable-area draft in a pump turbine system by comparing the results with actual measurements. The results show that the coupled method is effective in simulating water hammer in pipelines and transient processes of hydropower system.

Double-averaging methodology is very convenient for investigating spatially heterogeneous flows such as boundary-layer flows over permeable walls. However, spatial averaging volumes suitable for boundary-layer flows over rough walls are very thin in the wall-normal direction whereas those for porous-media flows usually have similar length in all three directions. This scale mismatch can be addressed by allowing the averaging volume to vary in space so that its size can be adjusted to the physical characteristics of particular flow regions. This paper presents a new spatial averaging theorem derived for a spatially variable averaging volume that may contain a stationary solid phase and that may also extend beyond the boundary of the problem domain. The theorem provides the expression for the difference between the average of a spatial derivative and the derivative of the spatial average (of a general flow quantity), here named the commutation correction ($CC$) term. The $CC$ term contains three parts accounting for (1) the presence of the solid phase in the averaging volume, (2) the averaging volume extending beyond the flow-domain boundary, and (3) the spatial variation of the averaging volume. The first two parts have been acknowledged in the literature; the third part is introduced in this paper and named the volume variation ($VV$) term. The averaging theorem is used to derive large-scale continuity and momentum equations, which contain the new $VV$ term. The equations are applied to steady, uniform, microscopically two-dimensional flow over a permeable wall. The averaging volume for the boundary-layer flow above the wall is a thin wall-parallel layer; on crossing the wall surface, the volume grows until its height reaches the size required for the homogeneous porous layer. The averaging procedure and the magnitude of the $VV$ term are illustrated by an example adopted from the literature that involves numerical simulation of two-dimensional open-channel flow over a bundle of circular cylinders.

This study shows that a jet-flipping phenomenon may occur during the scouring process downstream of a sluice gate. The flipping consists of a digging and a filling phase. The former scoured the bed, and the latter refilled the scour hole. The process is cyclical, the phases changed intermittently, and no equilibrium scour condition was reached. The scour depth during digging reached a high of 60% more than the filling phase. Moreover, the scour is more severe with jet flipping compared to no-jet-flip conditions. Generally, the maximum scour depth for the digging and filling phases increases with the increase in the tailwater depth and densimetric Froude number but decreases as the apron length increases. The jet-flipping frequencies increase with the increase in the densimetric Froude number and apron length but decrease as the tailwater elevation increases. The Coandă effect is the most significant factor that triggers the jet-flipping phenomenon. A review of relevant data from previous studies showed that the scouring duration tested for many cases was not long enough to observe the jet-flipping phenomenon. Prediction equations are proposed for the maximum scour depth at the end of the digging and filling phases and for the frequencies of jet flipping. A threshold diagram to predict the occurrence of jet flipping is also presented and could be applied in design to avoid the undesirable phenomenon in local scour below a sluice gate.

This study attempts to use a dam breach model to reproduce the well-monitored outflow hydrograph obtained during the dam breaching process of the Tangjiashan barrier lake, which was formed by a landslide triggered by the Wenchuan earthquake on May 12, 2008 in China. The key parameters that affect the model results, such as soil erosion and breach lateral enlargement, are reviewed by using field measurements followed by extensive sensitivity studies. The present paper advocates a hyperbolic model for soil erosion rate and a circular slip surface approach for breach lateral enlargement, which contribute to more reliable model results. The governing equations are solved using a numerical method that allows straightforward calculations coded in an Excel 2007 spreadsheet. This provides an easy, transparent, and robust tool that could enable practicing engineers to perform dam breach analyses with a comprehensive understanding of the uncertainties involved. This back analysis confirms that the peak outflow can be predicted with reasonable accuracy if the input values of the key model parameters are within well-understood ranges.

The idea of partitioning a flow has been widely used by hydraulic researchers and engineers as a mathematical treatment to simplify complex flows. Although the flow partitioning argument is intensive, no specially designed experiments have been conducted to verify them or the models, or to determine whether the existence of division lines can be inferred from mean velocity profiles. The aim of this paper is to examine the existence of division lines in smooth channel flow by analyzing the mean velocity distribution in a flume with a smooth flat or curved bed. The newly discovered division lines from experiments were compared with the existing models. It is found that some proposed models proposed have good agreement with the derived division lines obtained from the mean velocity distributions.