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.

The selective withdrawal of two-layer stratified flows with a point sink was investigated, considering the effect of the boundary. An analytical expression was developed for predicting the incipient Froude number as a function of the distance from the sink to the boundary (scaled by the distance from the sink and the interface). The effect of the boundary becomes significant when the relative distance is less than 2. The incipient Froude number reduces to 1.27 from 2.54 as the point sink approaches to the reservoir bottom. An analytical expression for computing the discharge ratio when both the layers withdraw was formulated and the velocity within each layer can be predicted. The analytical results were compared with the computational fluid dynamics (CFD) results. The CFD solver can successfully model velocities and the interface position in a selective withdrawal scenario for two-layer stratified flows.

In an open channel, the transformation from a supercritical flow into a subcritical flow is a rapidly varied flow with large turbulent fluctuations, intense air entrainment, and substantial energy dissipation called a hydraulic jump. New experiments were conducted to quantify its fluctuating characteristics in terms of free-surface and two-phase flow properties for a wide range of Froude numbers ($3.8<{\mathsf{F}}_1<8.5$) at relatively large Reynolds numbers ($2.1\times {10}^4<\mathsf{R}<1.6\times {10}^5$). The time-averaged free-surface profile presented a self-similar profile. The longitudinal movements of the jump were observed, showing both fast and very slow fluctuations for all Froude numbers. The air–water flow measurements quantified the intense aeration of the roller. Overall the present findings demonstrated the strong interactions between the jump roller turbulence and free-surface fluctuations.

Many of the empirical formulas used for the prediction of the expected scour depth at piers are excessively conservative, providing substantial overestimations. On the other hand, the recently proposed neural networks methods generally issue accurate predictions but also high percentages of underpredictions, due to the use of a symmetric error function for their parameterization. A novel error function is proposed in this paper for optimizing neural networks, giving more weight to underestimation than to overestimation discrepancies, in order to obtain safer design predictions. The performances of the proposed model on independent field records are compared with those of a conventionally trained neural network and with those of a set of widely used formulas. The asymmetric error function (that might be applied to parameterize any other model or equation, as a proficient alternative to least-square errors or envelope curves) allows researchers to obtain predictions closer to the measurements than those issued by traditional formulas, substantially reducing the extent of unnecessary overdesign and at the same time the percentage of severe underestimations is comparable with those of the safest formulas.

Experimental results on the piping erosion process in an earthen embankment emplaced in a laboratory flume are reported in this paper along with the details of the experimental set-up and procedures. A mixture of sand, silt, and clay with different compaction rates is used for constructing the embankment. An image processing technique is successfully applied to track the erosion process from both side-looking and bottom-up views. The paper discusses changes in the depth, area, and volume of erosion during the piping phenomenon in a model embankment using a visual technique. The results show that increasing compaction of the construction layers significantly increases the time required for erosion but has little effect on the final-average depth of erosion. The ratio of the average depth of erosion to the average bottom width of piping is close to one at different time intervals. Exponential equations to estimate the depth of erosion, side area of the piping zone, and volume of eroded material are presented. Two approaches are investigated to estimate the volume: the approach based on the processing of images from both the side–view and bottom–view gives more accurate results than that based on processing the side–view.

The majority of the bridges in the U.S. National Bridge Inventory (NBI) are built over waterways. Many will experience problems with scour, bank erosion, and channel instability during their design life. A number of studies in the recent past have attempted to quantify the uncertainty in predicting these erosional processes, particularly in pier scour, and developing probabilistic estimates of scour as a means of incorporating uncertainty. However, none of these studies have examined the overall uncertainty in local pier or abutment scour in combination with contraction scour. This study quantifies the model uncertainty in commonly used scour equations as well as parameter uncertainty. The overall reliability of the scour equations is then assessed for the individual components of scour and combined scour. The results lead to a set of scour design factors that are based on the reliability of the design estimate.

Air-entrainment and detrainment processes in tailrace channels downstream of Pelton turbines were investigated in a physical laboratory model of the new hydropower plant Veytaux 2, Switzerland. These processes are mainly influenced by the emergent jet velocity of the Pelton turbine and the resulting air entrainment to the water cushion due to high-speed jet impact. The air demand in the turbine housing, and the air concentration and velocity distributions in the tailrace channel are analyzed. The project-specific results are generalized by additional investigations involving a simplified model geometry. A comparison with previous model and prototype data is presented. The results indicate that the deaeration length relative to the tailwater flow depth depends mainly on the jet Froude number made up by the emergent jet velocity relative to the tailwater propagation velocity, plus the water discharge, the tailwater cross-sectional area, and the presence or absence of a flow straightener at the tailrace-channel entrance.

Open-water disposal of sediment is an important component in many coastal engineering projects. Numerous studies have focused on small-scale dynamics and claimed their results can be scaled up by cloud number scaling. However, this scaling method is largely empirical and unexamined. The present paper confirms that the cloud number scaling provides a rational way to extrapolate small-scale lab experiments or numerical simulations to the field operations.

The turbulence anisotropy in flow over noncohesive sediment beds at an entrainment threshold, as compared to that over immobile beds, is studied. The streamwise component of anisotropy tensor in near-bed flow for threshold beds yields more anisotropic stream as compared to that for immobile beds, whereas all other components of anisotropy tensor in near-bed flow for threshold beds yield less anisotropic stream. The anisotropic invariant maps illustrate the anisotropy in near-bed flow layer tending to be a two-component isotropy for both immobile and threshold beds. The data plots for immobile beds with an increase in vertical distance have an affinity to a one-component isotropy, whereas those for mobile beds show a tendency to become a three-dimensional isotropy near the midportion of the flow depth. Invariant function plots show a better two-dimensional isotropy in the inner layer of flow and a quasi-three-dimensional isotropy in the outer layer of flow over threshold beds than those over immobile beds.