Journal of Hydraulic Engineering

The design of appurtenant structures for dams, such as spillways, is a complex task because of the turbulent high-velocity aerated nature of the flowing water. In previous years, physical models would have been the only practical tool available to gain insight into the three-dimensional and time-dependent nature of spillway flows. Now, hydropower dam designers have multiple engineering tools to assess the hydraulic behavior in spillway structures. The high complexity level of these flows challenges the capabilities of computational fluid dynamics methods. Within this scope, the smoothed particle hydrodynamics method appears as a quite appropriate solution to solve flows dealing with large deformations, complex geometries, and movable boundaries. It has been used in a wide range of applications, including dam spillways. This paper provides an overview of how this method has been applied to investigate flow features in real spillways within the past decade. Its current state of knowledge with relevance to dam engineering practice is addressed and current limitations are highlighted in sight of future research could resolve them.

Embankment dams are common worldwide, and the majority of past dam failures are related to such dams. It is imperative to improve the prediction accuracy of embankment dam failure to better estimate the potential impacts on downstream areas and create effective emergency response plans. A substantial database of 162 case studies was established based on previous studies. Three groups of regression equations relating to peak outflow discharge $Q_p$, mean breach width $B_m$ and breach height $H_b$ were developed as a function of two hydraulic conditions (volume of water above breach bottom $V_w$, and depth of water above breach bottom $H_w$). These proposed prediction formulas have good prediction accuracy and statistical significance. Comparative analyses are presented to further validate the accuracy of the prediction formulas developed in this study. The results of relative weights analysis indicate that both $V_w$ and $H_w$ exert a significant impact on $Q_p$, and the effect of $V_w$ on $B_m$ is significantly higher than that of $H_w$ on $B_m$, while $H_w$ is considered a prominent factor of $H_b$. In addition, two main factors leading to embankment dam failure, overtopping and piping, were proven to cause a significant difference in the three breaching parameters ($Q_p$, $B_m$, and $H_b$).

The conditional double-averaged streamwise velocity, spatially averaged (SA) turbulence and dispersive quantities in flows over a water-worked gravel bed (WGB), and a screeded gravel bed (SGB) were studied. The effects of the temporal and advective bursting events on the WGB and SGB flows were analyzed by measuring the instantaneous flow field using a particle image velocimetry system. The flow conditions (flow rate and flow depth) were maintained the same in both the beds. The SA turbulence and dispersive quantities studied were the SA Reynolds shear and normal stresses, SA turbulent kinetic energy fluxes, dispersive shear and normal stresses, and dispersive kinetic energy fluxes. In experimental beds, owing to the water work, the WGB exhibited a spatially organized roughness structure as compared with a randomly organized roughness structure in the SGB, resulting in a higher WGB roughness than the SGB one. Analysis of the temporal bursting events revealed that the conditional SA turbulence quantities corresponding to the temporal sweeps were dominant within the roughness layer, whereas those corresponding to the temporal ejections govern above the roughness layer. Analysis of the advective bursting events indicated that the dispersive quantities involving advective sweeps were the key mechanism within the roughness layer, whereas those involving advective ejections were the prevailing mechanism in the main flow layer. The comparison of the results in the WGB and SGB flows elucidated that owing to the higher WGB bed roughness than the SGB one, the conditional SA turbulence and dispersive quantities in the WGB were greater than those in the SGB. Within the roughness layer, the bursting events in the WGB were more persistent and frequent than those in the SGB.

Certain closed conduits may undergo processes of rapid filling, in which the flow regime transits from open channel into pressurized flows. Air bubbles may be entrained through such pressurization interfaces, and this entrainment can have effects on the flow dynamics. Whereas air entrainment in static hydraulic jumps has been studied in closed conduits, such entrainment has not been investigated in moving pipe-filling bores. Experimental tests involving the creation of air-entraining pipe-filling bores were conducted, and key characteristics of the process were recorded. Theoretical propositions based on conservation laws are compared with results of experimental data to develop an expression to estimate air entrainment in the moving bores. The nondimensional dragged air flow rate in the two-phase flow generated by the jump is quantified based on a relatively small number of nondimensional parameters: supercritical Froude number, and the ratio between the bore velocity and the supercritical velocity, being grouped by the closed conduit slope. Agreement between the proposed expression and experimental data in terms of $r^2$, the coeffcient of multiple determination in multiple regression, is good, and may indicate an alternative way to compute entrained air pocket volume during rapid filling processes.

The purpose of this study is to analyze the effect that inserting an elastic tube into a pressure pipeline has on the water hammer phenomenon. This research draws upon theoretical analysis, experimental testing, and numerical simulations. Assuming perfect elastic behavior of the system, the formula for the constant pressure wave velocity in a pipeline with an inserted tube was derived. Experimental tests were carried out, aimed at reducing the pressure increase in the pipeline due to inserting a silicone rubber tube in it. A significant reduction of the pressure increase has been achieved. Theoretical values of the pressure wave velocity were significantly lower than measured. Numerical calculations were performed, the purpose of which was to simulate the course of pressure changes in the pipeline with inserted tube. An approximate model of unsteady flow was used, which relates elastic behavior of water, pipeline, and tube materials to the continuity equation via the variable pressure wave velocity. By taking into account the variable celerity of the pressure wave and diffusive term, it was possible to obtain an acceptable compliance between the experimental data and the results of the numerical calculations.

Stormwater geysers are commonly observed as explosive releases of air-water mixtures from municipal systems, which can cause property damages and human safety concerns. In Edmonton, Alberta, Canada, a number of geyser events were reported in recent years from a 27-m-deep vertical shaft above a junction chamber connecting a 3.5-m-diameter incoming pipe to an outgoing pipe with a 4-m drop. To understand the conditions for the geyser formation and select suitable mitigation methods, an experimental study was conducted using a simplified conceptual $1\text{:}20$ scaled model. Three series of experiments were conducted with a sudden increase of the inflow rate and different initial flow conditions: Series A with an initial downstream open-channel flow, and Series B and C, both with an initial downstream full-pipe flow, but an air pocket was deliberately entrapped in the upstream pressurized pipe in Series C. A geyser was not observed in Series A mainly due to the available large outflow capacity in the downstream pipe. With the downstream pipe initially full, a geyser event was observed in Series B as pressure surged in the chamber caused by the increased flow rate. Severe geyser events were observed in Series C, with the first phase triggered by the transient pressure wave and the second phase triggered by air released from the air pocket. The relationship between the measured maximum pressure and the geyser height was established. An analytical model is proposed for predicting the magnitude and period of the pressure oscillation induced by the sudden inflow rate increase. The amount of water splashed out of the riser was also measured.

Using similitude relationships, laboratory-scale models have been used for decades to predict and confirm prototype hydraulic structure performance. For free-flow weir discharge conditions, gravity and inertia typically represent the dominant forces, and the Froude number scales the head-discharge performance between model and prototype. Under low-head flow conditions, other forces (e.g., viscous and surface-tension forces) can become relevant, resulting in differences between the model and prototype performance. In this study, the head-discharge relationships for 15 different nonlinear weirs (labyrinth and piano key) with prototype-to-model length ratios of 2, 3, 6, and 12 (based upon Froude modeling) were evaluated, with the largest weirs ($\sim 1\mathrm{m}$ tall) serving as prototypes. This study found differences in the head-discharge performance between prototype and model that exceeded what could be explained solely by model and measurement effects, confirming the presence of scale effects. The range of small upstream heads influenced by scale effects, in general, increased with decreasing model size. The minimum dimensionless head above which scale effects were negligible increased with decreasing model size. As such, model size and geometric scale appear relevant because no single minimum upstream head limit was found that characterized a scale effects limit for all nonlinear weirs tested herein.

This paper documents a prototype-scale investigation of cavitation damage to a newly resurfaced spillway being used operationally during phased construction work to replace all piers and radial gates. Initial concrete damage on the right abutment spillway steps occurred in December 2015 as a result of over 300 h of continuous spilling from a new spillway gate. No cavitation damage had been observed on the original spillway constructed in the 1930s. Factors that contributed to the cavitation damage for this spillway are reviewed, including the duration of continuous spill and the increased cavitation potential of both a smooth concrete surface compared to one that is uniformly rough and sharp-edged steps compared to steps with rounded edges. Continued spilling through 2016 exacerbated concrete damage to the spillway steps leading to an experimental interim measure of constructing a tetrahedron concrete wedge to reduce the magnitude of flow separation and mitigate material loss in the location of greatest damage. Although damage to the steps continued to progress with ongoing spilling, the concrete wedge was successful in reducing the extent of damage at that location. A numerical model was calibrated by comparing simulated pressures to observed concrete damage allowing a long-term cavitation mitigation option to be designed with increased minimum pressures of approximately 50% and greater at the spillway steps compared to the 2015 as-built geometry.

Despite the numerous studies on scouring processes, the prediction of scour-hole dimensions downstream of hydraulic structures remains challenging because of the complexity of the phenomenon and its dynamic sensitivity to structure and sediment properties. This study experimentally focuses on scour-hole development downstream of a sloped grade-control structure (GCS) in alluvial channels. A large series of laboratory experiments were carried out in a rectangular channel with a noncohesive sediment bed. Based on the data from this study and data collected from previous studies, the effect of the downstream face slope of a GCS on scour morphology was analyzed. In this regard, it was found that the face slope has an effect only if it is smaller than the slope of the upstream equilibrium scour side obtained with a GCS of a vertical downstream face. Before reaching an equilibrium state, the scour process evolves into three distinct phases, a very rapid initial phase, an intermediate gradual phase, and an equilibrium phase. A general empirical expression for predicting temporal scour evolution is proposed and extended to different types of GCS. Moreover, a new scaling approach is proposed that leads to the derivation of new equations predicting equilibrium scour profiles with different entering jet-flow typologies. To make these equations operational, a series of estimating expressions for the characteristic lengths of equilibrium scour holes is also proposed.

High-return-period floods involve complex flows over large floodplains covered with different types of resistive elements, from bed roughness to large emerging obstacles. In order to characterize the flow resistance in such configurations, experiments are conducted in a flume with emergent aligned obstacles on smooth and rough beds. Under uniform flow conditions, the drag force is measured by a hydrodynamic force measurement device, thereby obtaining the relative contribution of the drag and bed friction forces to the total flow resistance. The results show a strong correlation between the reference surface ratio (${\lambda }_f$), i.e., the obstacle frontal area divided by the cell bed surface, and the relative contribution of the drag to the flow resistance, with negligible influence of bed roughness and shallowness. Equal contributions of bed roughness and form drag to flow resistance occur for ${\lambda }_f\approx 0.02$. After showing that the drag coefficient is roughly constant and the bed friction is consistent with Darcy-Weisbach’s head-loss coefficient, a Chezy-type equation is proposed to predict the mean uniform velocity $U$ as a function of the flow and land occupation parameters.

Ephemeral streams with mobile beds represent a challenging environment for sediment transport prediction. Bed topography from relatively infrequent high flow events may persist after hydrograph recession, resulting in sediment transport over bed topography that is not in equilibrium with flow conditions. Previous research has shown that an equilibrium sand bed formed at a high flow rate followed by a near-instantaneous reduction in discharge and depth produced a gradual reduction in sediment load that could be modeled with a two-term exponential equation in which the first term described the reduction in bed-form celerity and the second term described the reworking of the bed to a new equilibrium condition. In this paper, flume experiments with a range of flow rates that produced dune bed forms were conducted to establish a more general predictive relationship for sediment load following rapid reductions in discharge and depth from equilibrium bed and flow conditions. It was found that the period of adjustment, as larger bed forms were reworked into smaller ones, depended on the final reduced rate of sediment transport. The rate constant for the term of the exponential function, which sets the gradual rate of sediment load reduction after the initial rapid drop, was shown to also describe the gradual reduction in bed-form amplitude that occurred after the discharge and depth were reduced. The results of this work may be used to estimate declining sediment transport rates in a sand-bedded channel resulting from rapid reduction of discharge and depth.

The Bernoulli principle is one of the basic and most famous concepts in fluid mechanics, and the related equation is an extremely useful and effective tool in the solution of a wide range of practical engineering problems. However, the Bernoulli equation is not Galilean-invariant (i.e., it does not remain the same as a result of a change in the inertial frame of reference when the new frame moves with constant velocity relative to the original one). In this paper, a frame-independent expression of the Bernoulli equation is obtained from the Euler equation of inviscid motion for the case of steady flow, for the two classic versions valid along a streamline, and for irrotational flow. Compared to the conventional formulation of the equation, an additional term is present in the definition of the Bernoulli constant. An interpretation of the new equation is provided on the basis of the first law of thermodynamics. Some classic applications are presented, and the results found are in accordance with the consolidated knowledge of fluid mechanics.

A free-surface immersed-boundary lattice Boltzmann method for wave–structure interaction and hydrodynamic force estimation is introduced as a proposed model with example applications. The single-phase free-surface lattice Boltzmann method is coupled with a simple immersed-boundary modification in order to simulate demanding free surface wave problems and to evaluate the hydrodynamic forces on a breakwater. The immersed-boundary modification also enables a seepage flow simulation without any external model. First, the proposed model is applied to incident wave propagation in a shallow water zone to demonstrate the model performance on traveling successive waves and wave effects on groundwater on a porous slope. The wave–breakwater interactions and wave forces on a breakwater are then analyzed. The results agree with those of Goda’s formulae in all cases and with the third-order force expression from standing wave theory in nonoverflowing cases, confirming that the proposed model has a high potential for application to complex analysis of coastal engineering problems.

Design-based courses often have more latitude with respect to implementing interesting and unique teaching styles and curricula; they can also provide opportunities for broader-context problem solving, thereby improving student learning. Deductive teaching methods are used extensively throughout math, science, and engineering curricula [i.e., students are taught the “how” (fundamental principles), and the “why” is learned by applying those principles through solving problems]. When students successfully grasp why something is important, they often become more engaged in developing and understanding the fundamentals underlying the solution (inductive teaching style). With some creativity and planning, design-based classes can offer a positive alternative learning environment that utilizes both deductive and inductive learning processes. Four keys for enhancing student-learning experiences in a design-based course are discussed. The recommendations and techniques presented herein come from the authors’ first-hand experiences developing and teaching a hydraulic structures design course at Utah State University.

Cavitation erosion in hydraulic machinery, such as in turbines and pumps, often leads to significant reduction of the service life of the affected components, with serious consequences for their maintenance costs and operation efficiency. In this study, the potential contribution of acoustic emission (AE) measurements to the assessment of cavitation damage is evaluated from experiments in a cavitation tunnel. Stainless steel samples were exposed to cavitation and damage was characterized from pitting tests carried out on mirror-polished samples. The pits were measured using an optical profilometer and cavitation damage was characterized by pit diameter distribution. In parallel, AE time signal was measured directly from behind the samples. A dedicated signal-processing technique was developed in order to identify each burst in the AE signal and determine its amplitude. The AE amplitude distribution compares well with PVDF and pressure sensor measurements from literature. It is concluded that AE signal analysis can be used to monitor the formation of pits without visual examination of the damaged surface. This provides a basis for possible future applications of nonintrusive cavitation erosion monitoring in hydraulic machines, provided the findings remain true in a more complex environment.

The structure of a mobile bed in a laboratory channel composed of sand and gravel ($D_{50}=8\mathrm{mm}$) was characterized over a series of experiments with steady flows from a low flow, where only the finer fractions of the bed material were in motion to flows in which most bed material grain sizes were in motion. In each experiment, sediment transport rates were observed to initially be greater than the long-term mean rates, and fluctuations in transport rate decreased in period as bed shear stresses were increased. The bed surface median grain size increased with bed shear stress, while the sand fraction of the bed material organized into longitudinally extended corridors, which persisted as flow and transport rates were increased. The presence of the sand corridors was reflected by changes in the probability density function of the bed-surface elevation standard deviation evaluated at the grain-scale. The formation and organization of these corridors may have a strong influence on sand and gravel transport in channels with mixed sand and gravel bed material.

In the mixing layer at the interface between an open-channel flow and its adjacent lateral open-channel cavity, vortices are periodically shed and advected towards the downstream corner. As these vortices impinge the downstream corner of the cavity, they either reach the mainstream or enter the cavity. The presence of these coherent turbulent structures within the cavity plays a major role for momentum and mass transfers between the outer layer and the core region of the recirculation cell within the cavity. Using an experimental approach based on PIV measurements and a combination of proper orthogonal decomposition (POD) and coherent structure identification, we show that the vortices entering the cavity are advected by this gyre and can either travel all the way towards the upstream corner or end up in the core region of this gyre. Along their trajectory, they can split into smaller vortices and/or merge with others. Moreover, whereas the vortex shedding in the mixing layer is highly periodic, the entering of vortices in the cavity is intermittent. Indeed, near the downstream corner of the cavity, several successive vortices can be transferred from the mixing layer to the cavity or to the mainstream. In average, the vortices are distributed almost equally between the two downstream regions (mainstream or cavity) but we reported up to four successive vortices reaching the same region.

Predictive tools are widely used to study coastal and deltaic systems in support of basic research, planning efforts, engineering design, and the implementation of restoration or protection strategies. They have been extensively used to evaluate the effectiveness of natural and nature-based solutions (NNBS) to support ecosystem functions and services of coastal ecosystems and human communities experiencing increased risk from sea-level rise and severe storms. The potential benefits of NNBS are being increasingly recognized, particularly in remote areas or areas that are either technically or financially infeasible to be protected with levees or other difficult engineering alternatives. Local communities, however, are often excluded from proposing, screening, or evaluating NNBS as restoration and protection strategies. Communities are also not sufficiently involved in the development or application of the predictive tools. This research effort outlines an approach to developing knowledge-based predictive tools and a community engagement process to evaluate NNBS strategies proposed predominantly by local communities. Incorporating knowledge from local communities benefits and potentially improves the performance of predictive tools and their ability to capture visible trends and observations. To illustrate this concept, the authors present landscape models for coastal Louisiana that successfully reproduced the frequency of flooding of local roads, rate of shoreline erosion, salinity pattern changes, and presence/absence of key species (e.g., brown shrimp, oysters, and so forth). While these qualitative measures are not a substitute for well-established rigorous and quantitative model performance assessment approaches, they offer an effective approach to engage local communities and incorporate their knowledge in the development of the predictive models and the proposed protection and restoration strategies to be examined.

Short-span masonry arch bridges constitute a significant proportion of the existing bridge inventory in the UK, Ireland, Europe, and the northeastern United States. These historic structures are founded typically on relatively shallow footings of unknown depth and configuration or on timber piles weakened by age and environmental exposure. Flume experiments used two common forms of short-span, masonry arch bridge and showed how local scour may readily undermine such structures. The experiments demonstrated that traditional hydraulic adjustments do not significantly reduce scour depth: streamlining abutments with wing walls and using cutwaters on piers. The usefulness of stone armor placed around abutments and piers was proven, but such armoring may relocate scour downstream of the bridge. The findings emphasize the scour-countermeasure benefits of foundation strengthening, flow bypassing by means of relief culverts (where feasible), and the use of channel armoring and energy-dissipation structures downstream of the bridge.

The existing analytical solutions of dam-break flow do not consider simultaneously the effects of wet downstream bottom and bed slope on the dam-break wave propagation. In this study, a new analytical solution for the shallow water equations (SWE) is developed to remove this limitation to simulate the wave caused by an instantaneous dam break. The approach adopts the method of characteristics and has been applied to simulate the dam-break flows with different downstream water depths and slopes. The analytical solutions have been compared with predictions by the lattice Boltzmann method and the agreement is good. Although the proposed analytical solution treats an idealized case, it is nonetheless suitable for assessing the robustness and accuracy of numerical models based on the SWE without the frictional slope.