Journal of Hydraulic Engineering

Gravity currents play an important role in many aspects of natural processes and engineering practice. This paper presents a systematic lock-exchange experimental study to investigate the influence of linearly stratified environments and ramp slope on the hydrodynamics of gravity currents. A high-speed camcorder and particle image velocimetry (PIV) are applied to analyze the macro- and microstructures of gravity currents. It is observed that the front velocity of the gravity currents increases first then decreases for both stratified and unstratified environments. However, the density contrast that drives the currents decreases more quickly in stratified environments as the currents descend the ramp, causing more intense deceleration of the front velocity, which indicates the damping of ambient stratification on gravity currents. Based on the thermal theory, a new set of analytical formulas is then developed to determine the front location and velocity at the deceleration stage by considering the ambient density variation at depth. Further downstream, the currents would separate from the ramp and then horizontally intrude into the ambience where the currents are neutrally buoyant if the relative stratification parameter $S>1$. An improved equation taking into account of the ramp slope, ambient stratification, and inflow buoyancy flux is proposed to predict the separation depth where the horizontal intrusion happens. Furthermore, velocity profiles in the body part of gravity currents are well predicted by two equations with three fitted parameters obtained by using the present experimental data. The vorticity fields of the gravity currents are quantitatively investigated to show that the weaker stratified ambience can lead to a stronger vorticity field, in which the Kelvin-Helmholtz instabilities and turbulent billows at the interface significantly affect the entrainment and mixing. In comparison with the gravity currents in the unstratified ambiance, the turbulent mixing in the stratified environments is generally reduced. The presence of stratification can result in a more complicated motion and fluid structure of the gravity currents.

An experimental study was conducted on a large embankment weir with smooth and rough crests for a range of flow conditions $0.011\le Q\le 0.226{\mathrm{m}}^3/\mathrm{s}$. Close agreement between the rough and smooth crest experiments was observed in terms of flow patterns, free-surface profiles, and pressure distributions. The crest roughness had effects on the velocity distributions, the boundary layer properties, and the shear stress. On the rough crest, the velocity distributions were shifted upwards, increasing the growth rate of the boundary layer. The calculation of the shear stresses with the law of the wall, outer flow region, and momentum integral methods showed consistent results of about $\tau /(\rho gH)\approx 0.0015–0.0018$ for the smooth crest and of $\tau /(\rho gH)\approx 0.0024$ for the rough crest. The crest roughness reduced the discharge performance, while upstream and downstream ramps did not have any effects. The present observations highlighted the relevance of roughness and scaling on the hydraulic performances of weirs.

A comprehensive knowledge of energy dissipation during the accelerations and decelerations in transient flow is essential to develop robust and accurate hydraulic transient solvers. Currently, most transient solvers are not accurate enough to describe the physical phenomenon, tending to underestimate observed energy dissipation. The current paper aims to contribute to a better understanding of transient flow by measuring instantaneous velocities using 2D particle image velocimetry (PIV) and discussing key features of the mean and of the turbulent flow. Measurements were carried out at the middle section of a coiled copper pipe for Reynolds and Dean numbers of 7,000 and 1,040, respectively. In steady-state flow, the time-averaged velocity profiles have shown a strong asymmetry, with maximum axial velocity in the outer bend region of the pipe, caused by the centrifugal force. During transient flow, the ensemble-averaged velocity profiles have clearly shown regions with strong gradients near the pipe wall and reverse flow. The direction of flow is firstly reversed in the inner bend region of the pipe. The unsteady wall shear stress is different for the outer and inner wall, and is lower than in the steady state, confirming that energy dissipation cannot be estimated based on steady-state formulae for the wall shear stress.

This paper presents a method using Monte Carlo simulations for assessing uncertainty of moving-boat acoustic Doppler current profiler (ADCP) discharge measurements using a software tool known as QUant, which was developed for this purpose. Analysis was performed on 10 data sets from four Water Survey of Canada gauging stations in order to evaluate the relative contribution of a range of error sources to the total estimated uncertainty. The factors that differed among data sets included the fraction of unmeasured discharge relative to the total discharge, flow nonuniformity, and operator decisions about instrument programming and measurement cross section. As anticipated, it was found that the estimated uncertainty is dominated by uncertainty of the discharge in the unmeasured areas, highlighting the importance of appropriate selection of the site, the instrument, and the user inputs required to estimate the unmeasured discharge. The main contributor to uncertainty was invalid data, but spatial inhomogeneity in water velocity and bottom-track velocity also contributed, as did variation in the edge velocity, uncertainty in the edge distances, edge coefficients, and the top and bottom extrapolation methods. To a lesser extent, spatial inhomogeneity in the bottom depth also contributed to the total uncertainty, as did uncertainty in the ADCP draft at shallow sites. The estimated uncertainties from QUant can be used to assess the adequacy of standard operating procedures. They also provide quantitative feedback to the ADCP operators about the quality of their measurements, indicating which parameters are contributing most to uncertainty, and perhaps even highlighting ways in which uncertainty can be reduced. Additionally, QUant can be used to account for self-dependent error sources such as heading errors, which are a function of heading. The results demonstrate the importance of a Monte Carlo method tool such as QUant for quantifying random and bias errors when evaluating the uncertainty of moving-boat ADCP measurements.

Flow visualization techniques and velocity measurements of jets in crossflow prove that the jet follows a distinct trajectory as if immiscible with the cross stream. Such observations reveal similarities between open channels receiving lateral inflow and vegetated flows. In the latter, the vegetation effect is accounted for as an increase in resistance, represented by a drag force. Using a similar approach, a new equation accounting for drag is developed for spatially varied flow (SVF) with increasing discharge. Experimental water surface profiles (WSP) from a number of studies with different arrangements of SVF were used to validate the performance of the proposed equation. A sensitivity analysis revealed that for the data considered the influence of nonuniform velocity distributions was minimal thus momentum correction factor, $\beta$ was assumed equal to unity. The Blasius equation was used to estimate friction slope. WSPs estimated from the proposed equation were a closer match to measurements than the general equation for different channel bottom slopes and lateral inflow rates. The analogous results imply that the proposed equation is generic for any conditions of SVF with increasing discharge.

Laboratory experiments were conducted on three artificial rock joints with natural characteristics to study the nonlinear fluid flow characteristics in a single joint under different contact conditions. A noncontact, three-dimensional stereotopometric measurement instrument was used to measure the three-dimensional surface morphology of rock joints before they were tested. Subsequently, the composite morphology parameters of rock joints under different contact conditions were calculated by using self-programmed software. The experimental results showed that both the Forchheimer’s equation (emphatically discussed) and Izbash’s equation could provide an excellent description for the nonlinear fluid flow in a single joint. The linear coefficient $a$ and nonlinear coefficient $b$ of the Forchheimer’s law both generally decreased, and such rate of decrease was gradually reduced along with increasing offset distance. The value of linear coefficient $a$ approached zero and the value of coefficient $n$ was close to 2 with increment of offset distance to a certain extent. A factor $E$ was used for quantitatively estimating the nonlinear characteristics of fluid flow in single rock joint. In this work, a critical value $E=0.9$ was defined to classify the regime of fluid flow, which indicated that the inertial terms were more important than the viscous terms. These findings indicated that a non-Darcian turbulent flow in the single joint occurred with the Reynolds number ranging between 1,408.2 and 5,674.4 under high water pressure. For the first time, an empirical mathematical equation was proposed to describe the apparent transmissivity by a single joint by using the Reynolds number, the root-mean square height of the joint composite morphology, and the mean aperture under different contact conditions based on laboratory observations. These findings may be useful for the computational studies of coupled shear-flow properties flow in jointed rocks.

In the present paper, three-dimensional, unsteady, detached eddy simulation (DES) of a submerged hydraulic jump with an inlet Froude number of 8.2 is performed. The volume of fluid (VOF) method with a high-resolution interface capturing (HRIC) scheme is used for free-surface tracking. The mean velocity and turbulence quantities including the Reynolds stresses are compared with available experimental data to validate the results. The three-dimensional nature of the flow in the developing and developed zone of the submerged hydraulic jump is evaluated by examining the coherent structures using the ${\lambda }_2$ criteria. Additionally, proper orthogonal decomposition (POD) analysis reveals the dominance of smaller structures in the developed region of the submerged hydraulic jump. The presence of these smaller scales is directly responsible for the energy dissipation characteristic of the submerged hydraulic jump.

Deep storage tunnels are an alternative to mitigate combined sewer overflows, but they may develop transient flow problems during rapid filling conditions. Such issues include excessive surges, formation of pipe-filling bores, entrapped air pockets, and uncontrolled releases of air. Initial decisions on stormwater tunnel designs are generally based on steady-flow considerations but can have a significant impact on the transient hydraulic performance. However, tunnel designers rarely have an opportunity to systematically assess the interplay between geometric characteristics (e.g., tunnel diameter, alignment, and junction areas) and inflow hydrographs. This work aims to provide an examination of the relationship between key design parameters and the performance of stormwater tunnels undergoing rapid filling. A flow regime transition model was used to simulate 216 different scenarios of rapid filling of tunnels, considering air–water interactions such as air pocket entrapment, compression, and expansion. Results indicate that peak surges and air outflows though vertical shafts correlated well with design parameters such as junction shaft plan areas and inflows at the time of tunnel pressurization. Other variables, such as entrapped air pocket volume, did not correlate well with geometric or inflow parameters selected in this study. While these results are not directly applicable to any existing stormwater tunnel geometry, the findings should help designers to understand potential transient flow issues and the input conditions that may lead to the most severe transient conditions.