Journal of Hydraulic Engineering

Concise equations for the tangential, radial, and axial velocities as well as the water‐surface profile of air‐core vortices at hydraulic intakes have been derived. The equations were developed by modifying the equation for tangential velocity originally proposed by Rosenhead in 1930. Laboratory experiments were conducted on strong air‐core vortices near a water intake. The results indicate the equations agree with experimental measurements and are applicable to vortex motion in general. The equations for radial and axial vortex velocities contain an eddy‐viscosity term, which should be proportional to vortex circulation. Dimensionless parameters describing the circulation and intake Froude numbers and the intake submergence are also used in the analysis. Using these equations, the depth and three velocity components of an air‐core vortex at certain hydraulic intakes can be predicted. This information may be helpful to the practicing engineer in determining the submergence required to avoid air‐entraining vortices at intakes.

Computer‐based terrain models have become increasingly popular as a tool for automating much of the geometrical data acquisition necessary for hydrologic analyses. An algorithm is presented that uses triangulated irregular networks (TINs) to accurately delineate drainage‐basin boundaries. Boundary delineation can be performed for areas that drain to individual points or stream networks defined on the TIN. A special feature of the algorithm is that the segments of the stream network need not correspond to channel edges. This allows greater flexibility in defining stream networks on the TIN and makes it possible to apply the boundary delineation algorithm to urban areas where the stream networks may be dominated by manmade structures such as streets and canals or to rural areas where the terrain exhibits little relief. Once the basin boundaries are established, geometric parameters such as basin areas, slopes, and maximum flow distances are easily computed. These tools can be combined with hydrologic modeling software such as HEC‐1 to create a comprehensive hydrologic modeling environment.

As computation speed increases, Monte Carlo simulation is becoming a viable tool for engineering design and analysis. However, restrictions are often imposed on multivariate cases in which the involved stochastic parameters are correlated. In multivariate Monte Carlo simulation, a joint probability distribution is required that can only be derived for some limited cases. This paper proposes a practical multivariate Monte Carlo simulation that preserves the marginal distributions of random variables and their correlation structure without requiring the complete joint distribution. For illustration, the procedure is applied to the reliability analysis of a bridge pier against scouring.

Supercritical flow below a sluice gate in a horizontal smooth rectangular channel is investigated to explain the effect of the inflow condition on the characteristics of the hydraulic jump. The movement of jump location to variations of tailwater level is also examined so the position of the jump in a horizontal apron can be stabilized. The distributions of mean velocities and turbulence intensities were obtained with a laser Doppler velocitymeter for a wide range of supercritical Froude numbers. The development of the turbulent boundary layer, the water‐surface profile, and the location of the critical point (the point where the limit of the boundary layer crosses the water surface) are predicted by an approach to the boundary‐layer analysis, and the agreement of those observations with the experimental data is presented. Using the analyzed water‐surface profiles, the movement of jump location to variations of tailwater level is predicted accurately and verified by laboratory experiments.

In 1991, the Tennessee Valley Authority (TVA) began implementation of a Lake Improvement Plan to provide increased minimum flow and dissolved oxygen in releases from several dams and reservoirs in the Tennessee Valley System. Reregulation weirs, constructed in the first few miles downstream of dams, are one means adopted by TVA to sustain continuous improved minimum flow and wetted area for aquatic life downstream of hydroprojects during off‐generation periods. A series of low‐level pipes through the weirs, where some are fitted with regulating float‐actuated valves, maintain essentially constant releases over a fullrange of weir pool elevations. Previous designs of such float‐control mechanisms were based on empirical measurements from physical modeling. The present paper provides a functional description of such a flow‐regulating system and presents a quick and reliable analytical technique to determine float movement and pipe discharge rate at different headwater conditions. Data obtained from a full‐scale model of the TVA South Holston labyrinth weir pipe and valve assembly were used to develop this analysis. Results of the analytical model were found to compare well with experimental data.

Results comparing six column‐separation numerical models for simulating localized vapor cavities and distributed vaporous cavitation in pipelines are presented. The discrete vapor‐cavity model (DVCM) is shown to be quite sensitive to selected input parameters. For short pipeline systems, the maximum pressure rise following column separation can vary markedly for small changes in wave speed, friction factor, diameter, initial velocity, length of pipe, or pipe slope. Of the six numerical models, three perform consistently over a broad number of reaches. One of them, the discrete gas‐cavity model, is recommended for general use as it is least sensitive to input parameters or to the selected discretization of the pipeline. Three models provide inconsistent estimates of the maximum pressure rise as the number of reaches is increased; however, these models do give consistent results provided the ratio of maximum cavity size to reach volume is kept below 10%.

For unsteady open‐channel flows having N interacting unknown variables, a system of $N$ mutually independent, partial differential equations can be used to describe the flow‐field. The system generally belongs to marching‐type problems and permits transformation into characteristic equations that are associated with $N$ distinct characteristic directions. Because characteristics can be considered “wave” or “disturbance” propagation, a fluvial system so described can be viewed as adequately definable using these $N$ component waves. A numerical algorithm to solve the $N$ families of characteristics can then be introduced for formulation of an $N‐\mathrm{component}$ flow‐simulation model. The multimode method of characteristics (MMOC), a new numerical scheme that has a combined capacity of several specified‐time‐interval (STI) schemes of the method of characteristics, makes numerical modeling of such $N‐\mathrm{component}$ riverine flows feasible and attainable. Merging different STI schemes yields different kinds of MMOC schemes, for which two kinds are displayed herein. With the MMOC, each characteristic is dynamically treated by an appropriate numerical mode, which should lead to an effective and suitable global simulation, covering various types of unsteady flow. The scheme is always linearly stable and its numerical accuracy can be systematically analyzed. By increasing the $N$ value, one can develop a progressively sophisticated model that addresses increasingly complex river‐mechanics problems.

Sediment entry into irrigation and hydropower canals may be controlled by a sediment excluder at canal headworks and/or a sediment ejector on the canal downstream of headworks. These devices take advantage of the fact that the bed‐load part of transported sediment in a stream moves near the bed and the suspended‐load part is distributed nonuniformly in a vertical column, with heavier concentrations near bed. The sediment ejector consists of a horizontal diaphragm slab, positioned slightly above the canal bed, and of ducts below the diaphragm flush the heavy sediment‐laden water into an escape channel. The present study concerns determination of diaphragm height of the ejector for uniform sediment, based on principles of sediment transportation. The diaphragm height should be such that suspended sediment load in the canal approaching the ejector above the diaphragm level should be equal to the total sediment transport capacity of the canal downstream of it. A correction factor is suggested to account for the effect of flow disturbance at the nose of the ejector. A design example has been worked out.

A small, robust, and relatively inexpensive current meter was developed to measure uniform and oscillatory water flows in both open‐weather and enclosed‐physical‐model facilities. The 10‐MHz acoustic‐doppler volocimeter (ADV) measures three‐dimensional current velocity over the speed range of 0 to 2.5 m/s at a sampling rate of up to 25 Hz. Measurements are made in a small sampling volume located 5 cm from a slim transmit‐receive assembly. Comparison tests of the ADV in several engineering model facilities together with other velocity‐measurement instruments and in a precision tow‐channel facility were highly successful. The ADV is rugged and portable, and it can remain installed in facilities for extended periods without recalibration.