Journal of Hydraulic Engineering

Two detailed field experiments were performed to investigate the spreading dynamics of axisymmetric, inertia‐buoyancy controlled, intrusive gravity currents in linearly stratified environments. The appropriate intrusions were successfully generated using deep‐set point‐source bubble plume devices installed within different thermally‐stratified reservoirs. The intrusive flows were found to progress at a rate predicted from a simple inertia‐buoyancy force balance with the intrusion thickness decreasing with distance from the source. The results imply that shear waves of all modes were generated at the intrusion nose and able to propagate away to modify the flow field. Turbulent activity, internal to the intrusions, was found not to influence the spreading characteristics. Scaling arguments and the experimental findings were also used to derive a parameter describing the initial intrusion dynamics. By combining the current findings with previously published results for intrusions progressing in the viscous‐buoyancy regime, a relation describing the distance at which viscous forces become dominant was derived.

Field and laboratory experiments were performed to investigate the mixing induced by energetic point‐source bubble plumes in water columns of arbitrary stable stratification. Published data from various sources were also used in the present study to increase the parameter range. It was found that the bubble‐plume surface signatures were marked by a central unsteady core consisting of an air‐water mixture, surrounded by an annular up‐thrusting flow consisting of water only. Outside this, the rising plume water spread radially, plunged below the surface, and then flowed outward as an intrusive gravity current at a level of neutral buoyancy. By equating the observed arbitrary stratification to a linear profile of equal potential energy, the radial extent of the plunge point and the net detrainment rate could be parameterized using natural scales of the problem. These formulas were verified using field and laboratory data, and may thus be used in air‐diffuser design procedures.

Simultaneous measurements of instantaneous velocities (longitudinal and vertical), suspended sediment, and salinities are performed continuously during two spring tides in the Ota River estuary. The velocity sensors are fixed at three heights above the bed and a turbidity sensor is located near the bed. Although the vertical sediment fluxes occur intermittently (as well as the vertical momentum fluxes), the longitudinal sediment fluxes are not very intermittent and the fluctuations of suspended‐sediment concentration are linked to the sign of $u$ in spite of the $w$ sign. The fractional contribution curves to the vertical sediment flux are similar to those for the Reynolds shear stress. The patterns of conditionally averaged fluctuating velocities indicate that the large positive fluctuations of suspended concentration occur immediately after the transition from the positive to negative of $u$ and the suspended‐sediment concentration decreases at the early stage of the positive $u$. The sediment suspension is related to the large‐scale turbulence together with ejection motions; this large‐scale turbulence is probably associated with the cluster of ejections.

The resistance of alluvial channels depends upon flow and sediment characteristics. This paper considers the bed features and alluvial resistance that occur at, and close to, the transition zone between the upper and lower flow regimes. New flume and river data are published and the results are analyzed to provide satisfactory predictive techniques covering the lower, transition, and upper regimes for use by river engineers. The bed features that occur in the transition regime are shown to be washed‐out dunes that diminish rapidly in height and steepness with increasing shear stress. The increase in depth with increasing unit discharge is much reduced in the transition regime due to lower hydraulic resistance. In certain circumstances, depths in the transition region can decrease with increasing unit discharge. A criterion for the upper limit of the transition regime corresponding to a moving flat bed is proposed by the writers.

A new velocity‐distribution equation based on the probability and entropy concepts is recommended as an alternative to the existing, power law and universal law equations for pipe‐flow studies. With only two parameters, the new equation can represent the velocity distribution in the entire flow field in a pipe, regardless of whether the flow is laminar or turbulent, and whether the pipe is smooth or rough. An application of this new velocity distribution equation has led to new equations relating the friction factor to entropy. A statistical method is also presented to analyze and quantify the uncertainty in the computed velocity distribution due to uncertainties in the related parameters or variables.

A rotating flume installed at the National Water Research Institute in Burlington, Ontario, Canada to undertake basic research on cohesive sediment dynamics is described. The flume is 5.0 m in mean diameter, 0.30 m in width, and 0.30 m in depth. An annular cover plate (ring), which can be transversed in the vertical direction, fits inside the flume with a radial clearance of 1.5 mm on either side and makes contact with the water surface. The flume and the ring can be rotated independently at speeds ranging from 0 to 3 rpm. The flume is equipped with a laser Doppler anemometer, a laser particle‐size analyzer and a Preston tube. A set of performance tests carried out in this flume indicate that generation of a nearly two‐dimensional flow field is possible if both the flume and the ring are rotated simultaneously in opposite directions. Flows generated by keeping the flume stationary and rotating only the ring were found to contain a strong secondary circulation cell and a large variation of bed shear stress across the flume.

The Vedernikov number $V>1$ is the necessary condition for roll‐wave formation. Using the theory of linear stability, Ponce and Simons in 1977 confirmed the Vedernikov criterion for turbulent Chezy friction in hydraulically wide channels. In addition, they identified the scale of the disturbances under which roll waves are most likely to develop. Brock, in 1967, experimentally studied roll‐wave propagation using a laboratory flume. Using Brock's data, an experimental verification of Ponce and Simons' theory is presented. Results show that dimensionless wave numbers corresponding to the roll waves documented by Brock's study plot near the peak of the theoretical functional relation of logarithmic decrement $\Delta$ versus dimensionless wave number $\sigma$. The analysis confirms that roll waves are most likely to form for large positive values of the logarithmic decrement $\left(\delta >0.2\right)$, located within a narrow center band of dimensionless wave numbers. In a practical setting, this knowledge can be used to promote or inhibit the development of roll waves.