Journal of Hydraulic Engineering

Underflows generated from the impingement of descending dense cohesive particle-laden buoyant jets on a horizontal bottom of a water body were investigated theoretically and experimentally. Laboratory experiments were conducted by discharging fluid mud (suspension of cohesive clay particles in water) at a constant volume flow rate from a submerged vertical pipe into a body of water. The experiments involved a wide range of initial parameters (e.g., height of the discharge pipe from the bottom, flow rate, and concentration of the fluid mud at the time of discharge). The discharged fluid mud descended as vertical buoyant jets and flowed away as axisymmetric underflows after impinging on the bottom. The underflows exhibited two well-documented propagation phases: (1) a radial momentum-driven wall jet phase near the impingement region, and (2) an inertial gravity current phase after the completion of the wall jet phase. Experiments revealed that, depending on the initial discharge parameters, underflows underwent a viscous propagation phase in which the propagations were influenced by the non-Newtonian rheology of the discharged fluid mud. The dynamics of the propagation phases of underflows were analyzed and compared with previous studies on saline underflows. The radial wall jet lengths (radial extent of underflows at which they transitioned from the wall jet phase to the inertial gravity current phase) were found to scale on the characteristic jet length scale and discharge source-bottom separation distance when the buoyant jets impinge on the bottom as momentum-dominated jets. When the buoyant jets impinge on the bottom as buoyancy-dominated plumes, however, they scaled with only the source-bottom separation distance. Radial wall jet length parameterizations for both types of the impingements were developed. The predictive capabilities of two different shallow water model solutions were tested for the inertial propagation of fluid mud gravity currents. The transitions of fluid mud underflows into a viscous propagation phase were analyzed in terms of the discharge parameters and their non-Newtonian rheology. A box model solution for the viscous propagation of axisymmetric non-Newtonian gravity currents was developed. Predictions of the developed box model were then evaluated using the authors’ experimental observations. Of those underflows in which viscous propagation occurred, few showed an abrupt settling phenomenon after propagating some distance in the viscous propagation phase in which clay particles abruptly settled en masse from the underflows and only a turbidity cloud propagated. Discussions on the abrupt settling phenomena are also provided.

The paper describes an investigation into a water hammer with column separation. Special attention is given to the volume of cavity as a result of a pressure drop. Its volume has been determined by Gibson’s method and using images taken by a high-speed camera. The water hammer was then addressed numerically as a one-dimensional flow. Air bubbles affecting the speed of sound in the system were considered. The pipeline’s flexibility and attenuation were taken into account in the computation.

Check valves are normally installed at pumping stations to prevent reverse flow through pumps after pump shutdown. Check-valve slam appears to be among the most common causes of water hammer in water supply systems. A methodology for the selection of a suitable check valve for a new pumping station to prevent water hammer after pump shutdown appears to be inadequate at present. A method for analyzing pressure surges caused by check-valve closure after pump shutdown, taking into account valve characteristics as well as characteristics of the pumping system, is proposed in this paper. This method can be used to select the most economical check valve at the design stage that will prevent slamming by limiting the pressure surge to a desired value.

The hydrodynamics and contaminant transport in the nearshore region of Lake Ontario, from Port Hope to Cobourg, were simulated. The model results were comprehensively validated against observations of water level, temperature, and currents collected during April–September, 2010. The model generally agrees well with the observations. A RMS error of $\sim 2°\mathrm{C}$ was simulated and normalized Fourier norm (between 0.45 and 1.17), indicating that both the thermal stratification and currents are well-simulated, respectively, and are comparable to other model applications. Internal Kelvin waves were not observed and the internal Poincaré wave oscillation was observed offshore, but not modeled. Rather, up-welling and down-welling events caused by southwesterly and northeasterly winds, respectively, were both modeled and observed to be the dominant large-scale hydrodynamic processes. The episodic events lasted for 4–5 days with the upwelling front extending $\sim 10\mathrm{km}$ offshore. The up-welling and down-welling events generated geostrophic alongshore currents or coastal jets of $\sim 20\mathrm{cm}·{\mathrm{s}}^{\mathrm{-1}}$. The influence of these dynamics on the transport of river and wastewater-treatment plant plumes, toward drinking water intakes, was investigated using tracer release simulations. Tracer concentrations in the range of 10–0% (i.e., 70–90% dilution) were found at the Port Hope and Cobourg drinking water intakes. The tracer concentrations were primarily influenced by the proximity of the intakes to the plume origins and the wind direction, which governs the direction of the alongshore currents resulting from the up-welling and down-welling events. These results will help municipalities better understand the transport of contaminants in the nearshore zone relative to drinking water intakes.

Cross-flow turbines are very efficient and cheap turbines that allow a very good cost/benefit ratio for energy production located at the end of conduits carrying water from a water source to a tank. In this paper, a new design procedure for a cross-flow turbine working with a variable flow rate is proposed. The regulation of the head immediately upstream the turbine is faced by adopting a shaped semicircular segment moving around the impeller. The maximum efficiency of the turbine is attained by setting the velocity of the particles entering the impeller at about $2\times$ the velocity of the rotating system at the impeller inlet. If energy losses along the pipe are negligible, the semicircular segment allows always a constant hydraulic head and a constant velocity at the impeller inlet, even with variable flow rate. The decrease of the turbine efficiency along with the inlet surface reduction is first investigated; a design methodology, using also computational fluid dynamics simulations, is then proposed for both the cases of negligible and not negligible energy losses along the pipe.

The Sanmenxia Dam built in 1960 in the middle reach of the Yellow River has experienced severe sedimentation problems, not only in the reservoir area itself but also extending hundreds of kilometers in the backwater region. Morphological responses in the middle Yellow River and its tributary, the Wei River, upstream of the dam have generally lagged behind the causal changes in upstream and downstream controls. Based on the geometrical characteristics of a wedge-shaped sedimentation body, a method was derived for calculating equilibrium values of the accumulated sedimentation volume $V_e$ in the channel reaches. Combining the method for calculating $V_e$ with rate law models for channel morphological adjustment, a method for calculating the accumulated deposition volumes was derived. The results showed that the proposed method can accurately simulate sedimentation processes in the study reaches during 1960 and 2011, in the response to the integrated impacts of base level fluctuations and changes in incoming water and sediment discharges. The characteristics of the potential adjusted (equilibrium) morphology was analyzed and it is shown that while the nature of the adjusted (or equilibrium state) of the fluvial system varies dynamically with external conditions, it generally acts as the target for the observed channel adjustments, although the actual values of variables defining the fluvial system seldom adjust to their corresponding equilibrium values due to the effects of multiple disruptions, continuously varying water and sediment inputs, and delayed responses in the fluvial system.

Based on numerical simulation of hydraulic transients in a pump storage power station (PSP), different closure laws are investigated for use when a high-head pump-turbine load rejection occurs. According to the characteristics of the pump-turbine and main inlet valve (MIV), a novel closure law, combining closure of wicket gates (WG) and MIV, is presented. The proposed closure law can reduce the effects of S-shaped characteristics of a pump-turbine, and the maximum pressure in the spiral case, the maximum rotating speed of the runner, and the minimum pressure in the draft tube during the transient process can be mitigated. A simultaneous load rejection test on a double unit is conducted, and the numerical predictions agree well with the field test data.

In 2008 and 2009, a series of bed load impact plates was installed across a channel spanning weir on the Elwha River, Washington. This is the first permanent installation of its kind in North America and one of the largest anywhere. The purpose of this system is to measure coarse bed load during and after the removal of Elwha and Glines Canyon Dams. It is estimated that $21\mathrm{million}{\mathrm{m}}^3$ of sediment have accumulated behind both dams, with 45–50% of the total accumulated volume expected to be eroded through natural processes. The impact plate system consists of 72 plates installed at a diversion weir downstream from both dams and 5 km upstream from the river mouth. Of the 72 plates, 46 are instrumented with a geophone and 26 with an accelerometer. Collection of physical bed-load data for calibration of the geophone plates has begun, with additional measurements to be collected in the future. This paper describes the specifics of the Elwha impact plate system and the ongoing process to collect bed-load measurements for system calibration.

A nonintrusive method to calculate water depth and discharge in partially full pipes using data from a single ultrasonic Doppler velocimeter (UDV) profiler is presented. The position of the free surface is identified from the velocity profiles measured with the UDV. The flow discharge is computed from an approximated parameterization of the velocity field in the cross-section, using a single measured velocity profile. The proposed methodology was applied to steady and unsteady flow conditions in two different pipes with diameters of 90 and 200 mm, and depth-to-diameter ratios up to 0.35. Under these conditions, the water depth and discharge were measured with mean absolute errors of 1 mm and $0.1\mathrm{L}/\mathrm{s}$ in the 90-mm pipe and 0.5 mm and $0.05\mathrm{L}/\mathrm{s}$ in the 200-mm pipe. These errors are almost independent of the discharge.