Journal of Hydraulic Engineering

This paper presents the results of oceanographic drogue and current meter studies performed in the coastal waters of southern California. The purpose of the study was to evaluate, over diurnal periods, the transport and dispersing properties of the oceanographic currents at this location and relate them to extant theories of diffusion and dispersion in coastal waters. The measurements provide useful results concerning Lagrangian and Eulerian estimates of turbulent diffusivity. In particular, they show the importance of regarding diffusivity as an ensemble average property of the turbulent flow field, and that great care is necessary in applying the concept to single realization studies. The studies show that under some tide conditions significant negative diffusivities are possible. Furthermore, it is found that at this location coastal motions tend to be very coherent up to scales of 4–5 km, at which point they become rapidly dispersive. The results provide information of value to engineers concerned with the design of ocean disposal systems.

When the rate of flow in a closed conduit is changed, large‐scale conversions of mechanical energy often occur, particularly if the pipeline is carrying water or some other slightly compressible liquid. Mathematical expressions describing these transient energy transformations are motivated from first principles and derived by mathematical manipulation of the governing continuity and momentum equations. The resulting expression accounts for the kinetic energy of the fluid, the internal energy associated with fluid compressibility and pipeline elasticity effects, the energy dissipated by friction, and the work done at the ends of the conduit. The energy approach provides an integrated view of transient conditions in the pipeline and is thus a simple, efficient, and logically consistent way of comparing the transient response of different systems and solution techniques. In particular, compressibility effects are shown to be negligible when the ratio of the change in internal energy to the change in kinetic energy is much less than one. This rule helps to distinguish the “rigid water column” model of unsteady flow from the more complex water‐hammer theory.

Experimental study of clear water scouring around a circular cylinder shows that the scour mechanism is coupled to the three‐dimensional separation of the upstream boundary layer and the periodic vortex shedding in the wake of the cylinder. The first scour appears in the wake of the cylinder. The main scouring agent in the upstream region is a system of horseshoe vortices. The vortices have a periodical character that causes a triple‐scour profile to develop in the upstream region. During scouring, the number and periods of horseshoe vortex shedding undergo no appreciable change. Despite the clear water stage, the transport phenomenon is periodical. Transport of sediment takes place through turbulent scales of comparable size to macro‐length scales. The size of the horseshoe vortices are representative for the macroscale. Wake scouring is caused by the primary wake vortices and the accelerated side flow. The process is characterized by a strong periodical transport and the formation of ripples. The periodicity is controlled by the shedding frequency of the wake vortices. Collars attached to the cylinder cannot prevent the formation of the vortices.

A frequency‐related design storm with spatial and temporal variation is derived from basic statistical properties. The writers discuss the advantages of solution provided by the proposed design storm and the limitations of the current standard design storm, outline an implementation of the proposed design storm, and report on a comparison study of the current standard versus proposed design storm. The comparison study is performed using the U.S. Army Corps of Engineers' HEC‐2 Water Surface Profiles computer program and the U.S. Army Corps of Engineers, Fort Worth District, NUDALLAS Flood Hydrograph Generation Package modified by the writers to support the proposed design storm. The comparison study provides examples of information, which can be developed using the proposed design storm.

An algorithm is presented for tracing the path of steepest descent from a given starting point on a terrain model defined by a triangulated irregular network. This algorithm is then extended to solve several problems. The flow patterns for a site are generated by tracing flow paths from a large number of starting points. The approximate stream network or channel network is found by tracing the channels upstream from pits or exit points. Once the stream network is found, the source areas or contributing areas for each of the sections of the stream are delineated. The source areas are then used to delineate the watersheds of selected nodes in the channel network. The format of a possible data structure for the channel network is presented along with pseudocode examples for traversing the channel network and the source areas. Execution times on a desktop computer are presented along with suggestions for optimal use of the algorithms.

Models developed for the transient routing of intense rainfall on small armored slopes depend on expressions for flow resistance below and above the surface of the armored layer. A series of 52 experiments are conducted in two flumes to collect data on the basic relationships of flow resistance. Crushed rock of 26‐155 mm median diameter is placed on embankments with slopes ranging from 0.01 to 0.2 and subjected to steady flow in the flumes. Flow through the rock layer is found to be approximately proportional to the rock‐layer porosity, the square root of the slope, and the mean diameter of the rock, and weakly proportional to the stage. The portion of flow through the filter layer generally cannot be neglected. Flow resistance decreases and conveyance greatly increases when the stage exceeds the depth of the rock layer. A model for flow both beneath and above the rock surface on an armored slope is synthesized from formulas developed for flow in rubble dams and Darcy‐Weisbach formulas developed for flow in gravel bed streams and mountain rivers. The results of the model compare favorably to the experimental flow measurements.

A tidal channel in the Eastern Scheldt estuary in the Netherlands is closed by pumping a sand‐water mixture into the closure gap. During the closure, fields measurements of flow velocities and sand concentrations are carried out. Based on the results of the field measurements, computations with two‐ and three‐dimensional sediment transport models are carried out. The two‐dimensional computations show that the depth‐integrated transport rates can be reasonably predicted (within a factor of 2) in case of diverging flow. But less satisfactory results are obtained for converging flow. The three‐dimensional computations show quite good agreement between measured and predicted flow velocities, sediment concentrations, and transport rates, both for flood and ebb flow. A sensitivity analysis with respect to the time‐dependent concentration term, the shape of the velocity profiles, the turbulent mixing coefficients, and the bed material gradation is carried out. The results show no significant influence of these parameters. The main reason for the difference between measured and predicted results is caused by the formula modeling the sediment pickup rate (at the bottom).