Journal of Hydraulic Engineering

The simple model for a buoyant jet in a flow is presented. The model differs from those presently available in that it is forced to satisfy the limiting cases of a buoyant jet in still water, a nonbuoyant jet in a still fluid, a jet in a coflow, a momentum vortex, and an advected thermal. It reproduces these types of flow by using a spread constant both for the Gaussian and the vortex regions and using published data to determine criteria for the transition from the advected Gaussian to the advected vortex distributions. At the transition the minimum dilution for momentum and buoyancy flux are preserved. The flow from an infinite array of merging buoyant jets from an outfall diffuser is also discussed. When the ambient flow is stationary a method of obtaining a complete solution is outlined. When the cross flow is small the merged flow has the form of an advected two‐dimensional plume. For a larger cross flow the fluid becomes sufficiently stretched such that the lower stable surface is almost horizontal, whereas convective mixing occurs on the upper unstable surface. The number of merging outfall jets that are required for the central port to behave as if it were in an infinite array is 14 when the port spacing‐to‐diameter ratio is 18.

This paper presents the research results of statistical parameters associated with turbulence in a natural river caused by the movement of navigation traffic. Movement of navigation traffic within restricted inland waterways, such as the Illinois, Mississippi, or Ohio Rivers, changes their flow characteristics temporarily in space and time. These spatial and temporal changes must be quantified in order to assess their effects on biologically sensitive areas, and they have seldom been evaluated for natural waterways. Data are collected from the Illinois River to evaluate these changes. The longitudinal and lateral components of fluctuating velocity $(u\prime \text{,}v\prime )$, Reynolds stress $(-\rho \overline{u\prime v\prime })$, turbulent intensities $\left({\sigma }_x\text{,}{\sigma }_y\right)$, turbulent kinetic energy (TKE), and maximum return velocity generated by navigation traffic are analyzed here. Results have shown that significant changes occur in all of these parameters. The largest changes take place in a zone within 10% of the channel width from the shore.

Basic knowledge of turbulent structures in open‐channel flows are now available from accurate experimental data that were obtained using a laser Doppler anemometer in laboratory flumes. Many of these experiments have, however, been conducted at moderate Reynolds numbers. Therefore, it is not yet clear whether such laboratory data of open‐channel turbulence can be applied to river turbulence at high Reynolds numbers. In this study, velocity measurements of the Biwako‐Sosui River in Kyoto and the Aichi Irrigation Channel in Nagoya have been conducted by making use of three‐component electromagnetic flow meters. Multicellular secondary currents were evident in a wide river, whereas large‐scale free‐surface secondary currents associated with velocity‐dip phenomena were generated in a narrow river. The features of such secondary currents in rivers could be explained by theory of turbulence, and also they coincided well with the laboratory experiments and numerical calculations.

An experimental study of selective withdrawal using intakes in partially submerged structures is described. This study examines circular intake structures located in two layer‐stratified bodies of still water. Results are presented for horizontally and vertically oriented intake ports and for a peripheral gap extending around the circumference of the intake structure. Data are compared with submerged, uncapped intakes and with submerged intakes provided with a velocity cap.

A comprehensive hydrometric survey of an estuary can be both expensive and time consuming. Considerable economies in both cost and time can therefore be realized if some of the more difficult‐to‐determine parameters can be related to more readily obtainable information. In particular, the tidal prism, the tidal excursion, and the estuary mean depth are found amenable to such treatment. Such relationships can be derived from an analytical solution of St. Venant's equations in a Lagrangian reference frame, provided that the estuary shape can be described by an exponentially varying cross section—in alluvial estuaries this is generally the case; and the velocity variation within the Lagrangian reference frame can be described by a simple harmonic—this appears to be no problem if the tidal range to depth ratio is not too large. The equations obtained have been calibrated on results obtained with a one‐dimensional hydraulic model and verified with prototype data obtained from 15 estuaries in different parts of the world. The relationships obtained are sufficiently accurate for use in studies at prefeasibility level.

Potentially high energy dissipation on stepped overflow spillways would imply a significant reduction of the size of downstream stilling basins. The amount of energy loss on such spillways under skimming‐flow conditions is examined experimentally for structures with a moderate number of steps and a width‐to‐height ratio equal to 0.7. The experimental results, supported by dimensional considerations, indicate that the most important parameters governing energy dissipation are the ratio of the critical depth of flow passing over the spillway to the step height $y_c/h$, and the number of steps $N$. Dissipation is highest for the small values of $y_c/h$ tested, near unity, and decreases with increasing $y_c/h$; for a certain $y_c/h$ the dissipation increases with $N$. The results are compared to those of earlier investigations referring to a large number of steps. For commonly used ranges of the step dimensions, a general relation is established between the energy loss and the parameter $y_c/h$, which may be useful in practical applications.

The inner layer of a turbulent boundary layer consists of a linear layer in which the velocity distribution is linear, the log‐law region having a logarithmic velocity distribution and a buffer layer lying in between them. Separate equations are available for the linear layer and the log‐law region. The available inner region equations pertain to flow over smooth boundary and they are explicit (containing involved expressions) or implicit. No equation is available that includes the effect of roughness. Presented herein is a generalized equation for velocity distribution in the inner law region of a turbulent boundary layer. The equation includes linear and logarithmic velocity distributions and it is valid for hydraulically smooth and rough boundaries and the transition range in between.