Journal of Hydraulic Engineering

Turbulence measurements are made in the inlet region of a rectangular laboratory settling tank with a two‐component laser‐Doppler system. The flow fields from a plane jet impinging on two types of deflector, at constant depth and discharge, are studied. Effects due to density differences or the presence of sediment are not considered. Although the two flow fields exhibit marked difference in terms of the extent of their recirculation regions and the extent of regions of high turbulent transport, flow‐through curves from tracer tests indicate little difference. A comparison to the case without deflector shows improved hydraulic efficiency, attributable to the inhibition of short‐circuiting effects or to increased initial mixing. Strong and complex three‐dimensional effects are found. Implications for design as well as numerical modeling are discussed.

The instantaneous velocity of particles driven by an oscillating flow can be determined by means of Fourier analysis. A simple numerical method based on the principle of least square error is presented to obtain the solutions of the Fourier coefficients. The computed zeroth harmonic term, which is the effective fall velocity and is less than the terminal velocity in still fluid, is in good agreement with experimental observations. The reduced velocity is caused by the nonlinear modification of the drag force exerted on the particle by the fluid because the relative velocity between the particle and the fluid is no longer the constant terminal velocity. There are three parameters that determine the effectiveness of fall velocity reduction: The terminal velocity Reynolds number, the ratio of the flow velocity amplitude to the terminal velocity, and a quantity characterizing the frequency response of the particle to the unsteadiness of the flow motion.

Field surveys and experiments in two tilting flumes on hyperconcentrated sand‐water‐mixture (swm) flows were conducted. This paper is limited to supercritical flows over a flat bed. It is shown that an equilibrium slope can be defined at which sedimentation and erosion are in balance. For laminar flow conditions this slope is only dependent on the average sand concentration, whereas for turbulent conditions the flow rate per unit width is the main parameter. The sand concentration near the bed measured about 35% by volume for most test conditions. The vertical gradient in sand concentration decreased with increasing mean sand concentration, indicating a decrease in the damping of turbulence. The velocity distribution behaved fairly logarithmically, showing a variation in slope with varying sand concentration. A minimum value of the effective von Kármán constant was found at concentrations of about 20%. Finally, it appears plausible that at high concentrations a considerable growth of the viscous sublayer occurs, resulting in an increase in bed shear stress.

Shear and velocity distributions near a sloped bank in a 270° open channel bend are studied. Data, which include detailed velocity and turbulence measurements in both the longitudinal and transverse flow directions, are collected with a laser Doppler anemometer system. The flow adjacent to a 2:1 (horizontal: vertical) and a 3:1 sloped bank is examined. The purpose is to examine in unprecedented detail the developing flow near a sloped bank in a curved channel, including the effect of varying side slope. It is found that the longitudinal flow redistribution is fully developed at the downstream end of this 270° bend, although the spiral flow is found to develop much more quickly. In the straight reach upstream of the bend it is found that the friction factor on the sloped bank is more dependent on channel shape than a local velocity‐shear‐stress relationship. Within the bend, velocities and shear stresses on the sloped bank increase significantly, with the shear stress correlating more directly with the local velocity.

The depth‐averaged velocity and bottom shear stress distributions in a rectangular channel near a groyne are computed by using a 2‐D depth averaged model. The model uses a hybrid finite difference scheme and an iterative method to solve the governing equations of flow and turbulence transport. Due to streamline curvature effects in the region near the groyne tip, a correction factor is incorporated into the k‐ε turbulence model that significantly improves the agreement between the computed and experimental data of the velocities and of the streamline pattern compared to previous numerical methods. In this region the bottom shear stress is found to be largely influenced by the 3‐D effects. A 3‐D correction factor is introduced which considerably improves the computed bottom shear stresses. Sensitivity analysis is made on the k‐ε model coefficients and on the correction factors of the streamline curvature and the 3‐D effects. The experimental errors in the velocity and bottom shear stress measurements are analyzed. The average errors between the computed and previous experimental results are presented with confidence intervals.

Bank protection is often employed to stabilize an alluvial channel and to reclaim land from a flood plain. For the sake of channel‐bed stability the hydraulic geometry of the channel must be designed so that its sediment‐transport capacity matches the supply from the upstream natural reach. Construction of a flood‐control channel in a flood plain usually increases the flow velocity and sediment transport. To maintain an approximate sediment equilibrium, the slope of the channelized reach is reduced using drop structures. The channelized reach and the natural reach have dissimilar geometries; their sediment‐transport rates are usually different under the changing discharge, and the channel as designed will undergo scour and fill during the design‐flood event. To account for the time‐dependent processes and general channel geometry, a mathematical model for water and sediment routing is used to determine the hydraulic geometry of erodible‐bed channels. The simulated sediment yield and its spatial variation are employed to arrive at the equilibrium (nonscouring and nonsilting) geometry at the peak discharge. Sediment transport and fluvial processes are also modelled to determine the depth of toe entrenchment. The computer‐aided design of channelization is illustrated by its application to Santa Gertrudis Creek in California.

Laboratory studies are conducted on a rectangular broad‐crested weir to determine the effects of entrance and exit conditions on the discharge and required energy loss. In doing so, a method for visually determining whether downstream conditions affect the flume calibration is developed. The current mathematical model of these flumes satisfactorily predicts both discharge and required head losses. Approach‐ramp slopes of 3:1 and 2:1 are shown to be acceptable for accurate discharge prediction, although a slight rounding of the approach ramp corners will improve predictions. At high ratios of energy head to throat length and with a sudden downstream transition, the model slightly underpredicts the required energy loss. As long as the water‐surface profile is dropping continuously from the approach ramp to the end of the throat, the downstream water level is not influencing discharge prediction. Field observations and measurements support these laboratory results, which should apply to most long‐throated flumes, since a rectangular broad‐crested weir represents the worst‐case conditions for head losses.

The circulating water‐pump intake at Iowa Power's Council Bluffs Power Station Unit 3, located along the left concave bank of the Missouri River near Council Bluffs, Iowa, has suffered from chronic fine‐sand shoaling problems both within the intake bays and just outside the intake structure. In order to maintain the pump‐intake bays sediment‐free, an innovative scheme of promoting a deep scour trench in front of the intake is developed using a 1:36‐scale movable‐bed model. On the basis of the model testing results, an array of submerged vanes, comprising two rows of 13, 9‐ft (2.7‐m) wide sheet‐pile vanes, were installed in the prototype in August 1985.The prototype performance of these vanes has been found to be excellent during the past 3.5 yr, and no dredging has been required. The positive riverbed‐scouring approach developed in this study may be a viable solution to resolve persistent sediment‐deposition problems at numerous riverside water intakes.

Hydrographs of ephemeral streams in arid and semiarid regions usually decrease significantly in magnitude downstream, unless augmented by tributary flows. This process, termed transmission loss, is important not only in its obvious effect on flow reduction, but also as a source of ground water recharge to underlying alluvial aquifers. Most of the previous work in transmission loss was carried out on the semiarid regions of the United States. In this paper, three regression equations were developed for relating transmission loss to channel characteristics for an arid region in southwest Saudi Arabia. One of the equations seems to be reliable predictor of small losses associated with small upstream volumes. The other two equations were better predictors for large losses associated with large upstream volumes. Transmission loss data were also compared with some previously published. Although there was some scatter, the data is generally consistent with that previously published. This paper also adds to the data base of published data on transmission loss.