Journal of Hydraulic Engineering

Experiments were performed to measure the mean velocity and turbulence characteristics in open-channel flows with and without suspended sediment (alluvial sand of nearly uniform size). Velocity measurements were obtained by using a newly developed technique, the discriminator laser-doppler velocimeter (DLDV), which can distinguish both liquid and sediment particle velocities. The mean velocity of sediment particles was found to be lower than that of the water. While the velocity fluctuations in the water were not changed with the addition of sediment, those of the sediment were diminished. For the range of concentrations considered, the friction velocity and the free-surface slope increased with sediment concentration unlike other bulk flow parameters, which were practically constant. A uniform method of analysis was developed to facilitate data interpretation and comparison between similar studies.

Turbulence measurements over a smooth wall in unsteady depth-varying open-channel flows were conducted by the simultaneous use of a two-component laser Doppler anemometer (LDA) and a water-wave gauge. In the present study, velocity measurements in the viscous sublayer of highly unsteady flow have been first conducted to evaluate the friction velocity independently of the log-law. The von Kármán constant κ was calculated using this friction velocity. It was proven then that the value of κ was really a constant of 0.41 even in unsteady open-channel flows. Next, statistical structures of mean velocity and turbulence characteristics were investigated across the whole flow from the near-wall region up to the free surface. In particular, turbulence measurements near the depth-varying water-surface zone of flood flow were conducted successfully for the first time with the LDA; such measurements have been impossible with conventional velocity instruments such as hot-film anemometers. Mean velocity profiles and turbulence characteristics were revealed in both the rising and the falling stages of the flood period.

Gravel-bed river turbulence is investigated on the basis of field measurements in New Zealand rivers. The data include instant longitudinal velocities obtained by two methods: point measurements and vertical profile measurements. In both cases, fast-response electronic Pitot tubes were used. Analysis of vertical distribution of velocity, velocity pulsation skewness and kurtosis, structure functions, spectra, turbulence scales, and energy dissipation is presented. In most cases, the investigated velocity spectra and structure functions are characterized by the existence of the inertial subrange. The frequency spectra and structure functions satisfy the Kolmogorov “−5/3” and “2/3” laws in the range of the scales from the smallest up to (2–2.5) river depths. Using theoretical considerations and the field data, relationships for the vertical distribution of relative turbulence intensity, scale, and turbulent energy dissipation are derived. A new characteristic turbulence scale representing the upper bound of the inertial subrange is proposed and investigated. The influence of bottom roughness on relative turbulence characteristics is shown.

The term transcritical flow denotes the existence of both subcritical and supercritical flow regimes in an open-channel system. It is similar to transonic flow in gas dynamics. Transcritical flow is encountered in many engineering problems of practical interest; irrigation networks, for example, often involve transcritical flow caused by the dynamic operation of hydraulic structures. This paper presents a two-step, predictor-corrector, implicit numerical scheme for simulating transcritical flows of practical interest. The proposed scheme is second-order accurate in time and space and requires inversion of only bidiagonal matrices. It is, therefore, computationally efficient and can readily accommodate hydraulic structures. To suppress spurious oscillations near regions of steep gradients and facilitate the resolution of hydraulic jumps, nonlinear (adaptive) second-difference artificial dissipation terms are added explicitly. The scheme is validated by calculating a series of applications of practical interest. The computed solutions are compared with previously published results and exact solutions where available. It is shown that the proposed scheme is robust and accurate, has reasonable shock-capturing capabilities, and can easily simulate hydraulic structures.

The performance and reliability of four of the more popular methods for one-dimensional flow analysis of bridges were investigated. These four methods were the HEC-2 Normal Bridge Method, HEC-2 Special Bridge Method, WSPRO, and Modified Bradley Method. The study was based on models of 13 flood events at nine different bridge sites in the southeastern United States. The sites were located on wide, flat, heavily vegetated floodplains where only free-surface, subcritical flow conditions occurred. The results showed that the HEC-2 Normal Bridge Method was able to accurately simulate the measured backwater values when the recommended 4:1 expansion ratio assumption was not applied. The application of the 4:1 expansion ratio assumption generally resulted in water surface elevations and backwater values that were higher than the measured values. The HEC-2 Special Bridge Method generally underpredicted the measured backwater, and WSPRO tended to compute the backwater to be slightly greater than the measured values. The Modified Bradley Method consistently underpredicted the measured backwater amount at each site.

The Lagrangian second-moment method is used to compute the dye-concentration distribution in turbulent recirculating flow generated in a shallow open channel by a sudden widening of flow in the transverse direction. The performance of the second-moment method is compared with the Eulerian numerical schemes of Upwind, QUICK, HLPA, and the semi-Lagrangian Hermite scheme. The results show that the Lagrangian second-moment method is an accurate and a very promising scheme for the computation of concentration distribution in flows dominated by advection. It minimizes the generation of numerical diffusion, is free from numerical oscillations, and is independent of the value of Courant number. Its efficiency in terms of computational time is high in flows where the mixing is confined to a small region of the solution domain and low in flows where the mixing is extended in the whole solution domain. In addition to an accurate numerical scheme, a necessary requirement for the accurate prediction of concentration field is a hydrodynamic model able to determine correctly the velocity field and the diffusion coefficient in shallow recirculating flows.

Processes controlling streamflow generation and suspended sediment transport for the Aberjona River watershed, a highly urbanized watershed located near Boston, were evaluated through a program of measurements on the Aberjona River and its principal tributary, Horn Pond Creek. For the Aberjona River, streamflow was associated with three distinct components (quick strong flow, slow storm flow, and long-term baseflow); and suspended sediment transport was likewise associated with these same three components as well as rainfall and channel routing effects. Due to storage effects of larger reservoirs, streamflow through Horn Pond Creek did not respond with distinct components; and suspended sediment transport was affected by long-term seasonal effects such as algal growths and the deposition of particulates. Separation of the streamflow hydrograph into components was essential for understanding transport processes for the watershed; a similar hydrograph separation technique should be useful for other rivers in urban areas whose response is not attenuated by storage effects.

This paper investigates the effect of velocity on the settling characteristics of coarse particles in a moving Bingham fluid. The experiments conducted in a recirculating pipeline show that the thickness of the deposited coarse particles contained in the Bingham fluid increases when the flow starts to move. The increase continues until a critical velocity, beyond which the thickness of deposition decreases. When the velocity is further increased to another critical value, all the particles in the flow are suspended. Further analysis of the concentration distribution of coarse particles shows that the Bingham structure is affected by the flow condition. For fluids with a low fine particle concentration, a small velocity (turbulence) increases the Bingham yield stress, resulting in a reduction of the settling velocity of the coarse particles. A high velocity (turbulence) destroys the Bingham structure of the fluid and increases coarse particle settlement. For flows with higher fine sediment concentration, the velocity (turbulence) weakens or destroys the Bingham structure and the settling velocity of coarse particles increases.

Cross-sectional velocity distribution measurements from the Sacramento Delta, Northern California, have been analyzed in order to extract values of the parameters that characterize longitudinal dispersion. These parameters are the velocity deviation intensity ratio and a dimensionless constant derived from the velocity profile by triple integration. Fischer estimated these parameters by combining laboratory measurements, theoretical derivations for ideal cases, and data from real streams. The results presented in this paper show that the parameter values used by Fischer may lead to a significant overprediction of dispersion in the delta channels. With a view to future numerical modeling, a quartic profile function was defined and shown to provide a reasonable model of measured transverse velocity profiles.