Journal of Hydraulic Engineering

One of the fundamental units in fluvial systems is a bifurcation, where an existing channel splits into two. Diversions are a special class of asymmetric bifurcation, in which one of the channels after bifurcation continues along the original channel. Over the years, engineered diversions have been used for irrigation and navigational purposes. Interest in diversions has also been fueled owing to their similarity with naturally occurring asymmetric bifurcations. Thus, dynamics of the flow and distribution of sediment between the two bifurcating channels at a diversion have been active areas of interest to river engineers for almost 100 years, when a series of experiments showed that the ratio of bed load entering the two channels after a diversion did not mirror the ratio of the water flowing into the two channels. This nonlinear phenomenon is often referred to as the Bulle Effect. Subsequently, other experimental studies corroborated these findings and expanded the parameter space upon which the phenomenon depends. Interest in diversions has been rekindled recently because of their possible use to divert sediment (and water) for mitigating land loss at river deltas. This paper charts the history of studies on river diversions, bringing together studies on the Bulle Effect spanning almost a century. The objective of is to evaluate the state of art in the field and draw some holistic conclusions in order to shed light on the fundamental mechanism driving the nonlinear phenomenon of the Bulle Effect and its implications for the design of sediment and water diversions for reclamation of land in sediment-starved deltas.

In several engineering applications, it is important to model flows that might transition between free surface and pressurized conditions. An explicit smoothed particle hydrodynamics (SPH) model for incompressible fluid is used to simulate flow in conduits during transitions between free surface and pressurized flow. The model was successfully tested against experimental data and numerical and analytical models in cases available in the literature. To appropriately produce laboratory experiments, procedures to properly define upstream and downstream boundary conditions were developed. These permit the description of conditions given in terms of either flow rate or water depth changing in time. Comparisons with commonly used one-dimensional (1D) models based on the Saint-Venant equations showed the ability of SPH to describe in details fluctuations of water depth, pressure distributions, and velocity fields. In the case of a propagation of a surge wave downstream, the rate at which the conduit pressurized was very close to measurements as well as existing 1D models. However, the dynamics driving the surge wave according to the SPH model appear to be different from these existing models.

This paper presents an experimental research for studying the relative and interactive effects of three physicochemical fluid characteristics (viscosity, pH, and ionic strength) on the incipient motion of a granular particle under laminar flow condition. Critical flow velocity for particle’s incipient motion, an important factor for evaluating flow-induced soil erosion, was used to quantify the relative erosive capacity of the test fluids. Response surface methodology (RSM), a statistical design of experiments, was used to design and implement 20 test fluids with various levels of the three fluid characteristics. Incipient motion of a highly spherical glass bead positioned atop a specially designed and manufactured support pocket within an enclosed flow cell was observed using a microscope video camera. The critical flow velocity was quantified for seven repeat trials for each of the 20 test fluids. Regression analyses were conducted on the results to generate a statistical model to describe the relative effects of the three factors and their interactions. The main findings include: (1) viscosity, pH, and their two-way interaction were determined to be the most influencing factors on critical velocity; (2) the effects of the three fluid characteristics on critical velocity follow the following decreasing order: viscosity, pH, and ionic strength; (3) viscosity and pH were both shown to have an inverse relationship with critical velocity; and (4) at higher pH (such as 10.5), the viscosity has greater influence on critical velocity than at a lower pH (such as 3.5).

A new shaft spillway composed of an inlet structure that includes inlet swirling-flow-generating piers, an annular weir, a vertical shaft, and an outlet structure is presented. Major factors influencing the inlet’s discharge coefficient, including the number of piers, pier weir angle, weir head, and pier height are discussed. An empirical discharge coefficient equation using a relative dimensionless weir head is presented in order to provide design guidance. Model experiments and numerical simulations are conducted regarding the flow behavior of the proposed inlet and vortex drop structure performance. Hydraulic characteristics such as the flow pattern, air core distribution, annular hydraulic jump position, pressure profiles, and water profiles of the outlet tunnel are obtained and agree well with the measured experimental data. Results show that the flow around the inlet is divided into a free-flow swirling region near the piers and submerged-flow region at the piers. The flow rate grows nearly linearly with the relative dimensionless weir head in the submerged region, and the discharge coefficient is approximately constant at 0.223. Experiments and simulations reveal the rotational flow movement mechanism of the self-regulating underwater swirl piers and the energy dissipation of the new vortex drop shaft spillway. Analytic calculations for the resultant velocity and water course thickness of the shaft sections correlate well with the results of a numerical simulation.

A new method is proposed to solve the two-dimensional shallow-water equations and derive the hydraulic properties in the wave tip region of the dam-break flow on a dry frictional horizontal channel of a rectangular cross section. Based on the assumptions that the resistance coefficient remains constant and the velocity at the wave tip region is spatially uniform, a relation between the wave tip velocity and time is established. Theoretically, the present results apply to the entire dam-break duration because the calculated wave tip velocity never reduces to zero. With bottom resistance, the water depth at the transitional position increases, whereas the wave tip velocity decreases and the wave front position is significantly retarded as time passes. Taking into account that the wave tip velocity is about half of the positive wave front velocity without bottom resistance, the transitional position is identified to move first forward and then backward into the reservoir. Accordingly, the mass and momentum of the wave tip region increase monotonically with time. In addition, the temporal water surface profile in the wave tip region is presented.

Focused waves are often used in physical and numerical studies as a representative condition for extreme waves or as a means to generate very steep and breaking waves at a prescribed location in space and time. They have also been combined with depth-varying currents in investigations of incipient wave breaking, wave breaking–induced energy dissipation, and wave–current induced loads on marine structures. A focused wave is created when all the components in a transient wave group come into phase. In the past, linear wave theory and iterative methodologies coupled with the linear Doppler-shifted dispersion relationship have been suggested to account for the presence of a current and achieve the required phase and amplitude focusing. In the majority of cases, linear or constant steepness spectra are used, which, compared to measured or theoretical spectra like the Joint North Sea Wave Project (JONSWAP), Gaussian, and Pierson-Moskowitz (PM) can be termed unrealistic. The effectiveness of these methodologies also decreases as the nonlinearity increases; therefore, in most studies, either weakly nonlinear conditions are used or the focus location is determined empirically. Here, an iterative methodology is suggested that can focus waves of any height at a predetermined temporal and spatial location even for wave groups propagating on a strong following or adverse current. An experimental apparatus developed to generate relatively stable sheared velocity profiles is also described. The depth-varying profile of the resulting currents diverges from that of classical wind-driven currents and comes closer to profiles measured in field sites important for the deployment of, for instance, tidal and wind energy converters. The methodology is successfully applied to wave groups traveling on still water, following, and adverse currents, and the results presented refer to linear, weakly nonlinear, and strongly nonlinear–focused waves generated for a range of realistic target spectra. The capability to generate wave groups with the same amplitude spectrum at a fixed location for a variety of flow conditions—still water, following, and adverse sheared currents—is also illustrated.

To simulate transient cavitating pipe flow, the discrete gas cavity model (DGCM) is combined with first-order and second-order finite-volume method (FVM) Godunov-type schemes. The earlier discrete vapor cavity model (DVCM) and DGCM based on the method of characteristics (MOC) are known to produce unrealistic pressure spikes. The new FVM-DGCM extends the previously developed FVM-DVCM through the introduction of a very small amount of free gas at the middle of each computation cell. Importantly, a pressure adjustment procedure is proposed to establish the relation between the cavity and the halves of the reach. Predictions of FVM-DGCM are compared with those of FVM-DVCM and MOC-DGCM and with experimental data. Results show that the proposed model reproduces the experimental pressure histories considerably better than the other two models. In particular, it produces fewer spikes, but—as in the old models—the first pressure peak due to cavity collapse is predicted much better than the subsequent peaks. The second-order FVM-DGCM is found to be accurate and robust, even for Courant numbers significantly less than 1.

This paper investigates mixing of a saline water jet into ambient freshwater at rest in a large basin. Reynolds-averaged Navier–Stokes (RANS) and diffusion-convection equations of the saline water volume fraction are used to model the mixing and the propagation of the saline gravity current jet. The comparison of the numerical and experimental front positions shows good agreement for Reynolds and Richardson numbers varying in the ranges $\mathrm{2,222}<\mathsf{R}<\mathrm{3,889}$ and $0.003<\mathsf{Ri}<0.01$, respectively. The numerical simulations of the hydrodynamic fields show that the velocity maximum is located at 0.18 $z_{0.5}$, where $z_{0.5}$ is the height at which the mean velocity $\overline{u}$ is equal to half the maximum velocity ${\overline{u}}_{\max }$. The local gradient Richardson number ${\mathsf{Ri}}_g$ shows that maximum turbulent mixing occurs at $z\approx z_{0.5}$ in the first stage of the gravity current close to the inlet, whereas far from the inlet, the turbulent mixing collapses. The entrainment depends both on the front position and on the values of the isodensity threshold, whereas it is independent of bulk Reynolds and Froude numbers in the narrow ranges ($800<{\mathsf{R}}_{\mathrm{b}}<1200$; $0.3<{\mathsf{F}}_{\mathrm{b}}<0.5$).

Poor mixing in storage tanks can cause stagnant zones that could pose negative public health effects. To eliminate stagnant zones in a cylindrical water storage tank, this study investigated the feasibility of a novel internal piping configuration consisting of a sprinkler-type inlet piping that distributed the incoming flow evenly across the water surface and a corresponding upside-down sprinkler draining configuration at the tank bottom. The experiments were conducted on seven types of inlet and outlet configurations. Experiments were performed using acoustic Doppler velocimeter (ADV) and electrical conductivity (EC) measurements. Various stagnation metrics, like the time for the water to be well mixed, and water particle velocity and direction, were used for each configuration. In addition, a new metric is introduced to evaluate short-circuiting conditions in storage tanks. Inlet or outlet convergence time is defined as a simple indicator in which the lower the value, the higher the risk of short-circuiting. In the next step, experiments were performed with a temperature difference between the tank and the inflow because of density-driven flows. Results indicated that the novel piping resulted in parallel downward streamlines that eliminated most of the stagnation zones in the tank.

A physical model study was conducted to investigate the flow regimes in a plunging-type dropshaft under the conditions of no air supply and limited air supply. When the top of the dropshaft is sealed, three flow regimes are observed with the increase of the water flow rate: (1) free-flow regime, where the falling water partially occupies the cross-section and air can move into the dropshaft from the bottom with pressure atmospheric inside the dropshaft; (2) water-plug regime, where a water plug is formed at the bottom of the dropshaft, blocking the air flow and causing negative pressure; and (3) choked regime, where the water plug rises and submerges the inflow pipe and the dropshaft operates under the full pipe flow condition. The flow regime changes are studied as a function of the dimensionless water flow rate (using the dropshaft diameter and gravitational acceleration). A large drop height leads to an early formation of the water-plug regime. When the top of the dropshaft is partially open to the outside atmosphere by a valve, the water-plug regime appears slightly later than the cases with the dropshaft top sealed. The flow regime has a direct impact on the energy dissipation in the dropshaft.

Friction factor determination is important for modeling fluid flows in pipes. The Colebrook equation has been widely used for estimating the pipe friction factor in fully developed turbulent regime. Because of the implicit nature of Colebrook equation, various regression-based approximations, Lambert W-function-based solutions, series-based solutions, and analytical approximations are developed for explicitly determining the friction factor. This study focuses on approximate analytical solutions for the Colebrook equation with a minimum number of natural logarithms and noninteger powers (lower computational cost). For this, a new mathematically equivalent representation of the Colebrook equation is presented. This form consists of two nonlinear equations which are very well suited for developing the analytical solutions for the friction factor. The simple analytical solutions developed in this study with the maximum relative errors less than 0.85, 0.25, 0.054, and 0.0028% (solutions with different degrees of accuracy) are among the most accurate analytical approximations to the Colebrook equation. Simple form and superior efficiency of the proposed solutions make them preferable to currently available approximate solutions to the Colebrook equation.