Journal of Hydraulic Engineering

The nonstratified steady flow hydrodynamics of rectangular sedimentation tanks is simulated using a combination of the strip integral and finite element methods. The strip integral method is generally a forward marching scheme in which the partial differential equations of continuity and momentum are reduced to a set of ordinary differential equations in terms of certain preselected parameters. These parameters, along with a set of shape functions, describe the velocity distribution. In this study three shape functions are chosen, corresponding to the boundary layer, the potential core, and the free mixing and recirculation zone. The shape functions were: (1) Power law in the boundary layer; (2) uniform velocity in the potential core; and (3) Gaussian distribution in the free mixing zone. The shape functions are chosen to allow recirculation above the mixing zone. A modified mixing length approach is used to introduce the effect of recirculation into the solution. A Runge‐Kutta method is used to integrate the set of ordinary differential equations; a finite element solution is used in the withdrawal zone. The numerical results compare favorably with experimental velocity distributions obtained by laser anemometry and with available field data. The convective and mixing characteristics of the flow can be obtained from this model and used with a transport model to determine the removal efficiency of the tank.

The thermal front phenomenon was observed during field surveys at the Ginna Nuclear Power Plant. Temperature oscillations were observed with amplitudes of over 3 °C measured from the mean with periods ranging from approximately 100–700 sec. A physical scaled model of the Ginna discharge was also used to study the thermal fronts and plume characteristics. The model was 31 ft (9.4 m) long and 16 ft (4.9 m) wide and geometric scale ratios of 1–50 and 1–100 were used to model the plume. Temperature versus time for several positions inside the model thermal plume exhibit a clear oscillatory pattern with amplitudes varying from approximately 1–5 °C (average differences between the discharge and ambient were 11 °C). The discharge and ambient temperature traces revealed no significant temperature fluctuations. Predominant periods ranged from 10–70 sec in the model and approximately 100–700 sec in the prototype. The power of the dominant oscillations tended to increase with depth at the same position inside the plume as did the magnitude of the temperature variations. The mechanism responsible for the thermal fronts is believed to be due to a Kelvin‐Helmholtz instability. This instability results in vortex formation at a shear layer due to the movement of the heated surface plume across the cooler ambient lake water. The Richardson number (J) computed from the measured temperature and velocity profiles, for locations in the near field plume with temperature oscillations, fell well within the unstable range when compared to J for the neutral curve developed for the Kelvin‐Helmholz instability.

The meander curvature and other geometric features for rivers are analyzed using an energy approach together with relations for flow continuity, sediment load, resistance to flow, bank stability, and transverse circulation in channel bends. The analysis establishes the maximum curvature for which a river does the least work in turning. This curvature, stated as the radius of curvature to channel width ratio, has an average value of 3; it shows only minor variation within the meandering range. Meandering development is explained by the river's tendency to seek minimum channel slope for the given conditions. The analysis demonstrates how uniform utilization of power and continuity of sediment load are maintained through river meanders.

The boundary integral equation method is formulated to calculate hydrodynamic pressures in a reservoir Impounded by a dam and subject to horizontal ground vibrations. The dam is assumed to be either a rigid body or a thin elastic structure. The radiation condition for infinite reservoir is treated using a hybrid method. Good agreement between present numerical results and analytical solutions is observed for the cases where water depth is a constant. Numerical results are also obtained for nonuniform water depth cases.