Journal of Hydraulic Engineering

A slow draining reservoir on the U.S. East Coast was monitored to identify the processes governing channel evolution upstream of a dam removal. Channel evolution was documented through cross section surveys, sediment size analysis, discharge measurements, and visual assessments of vegetative growth. The reservoir drained slowly, allowing for an analysis of channel evolution and identification of the morphometric parameters defining the path and time required for a channel to reach dynamic equilibrium. Channel evolution was a multidirectional process, and evolving channel reaches actively migrated laterally while alternating between aggradation and degradation. Channel formation was dominated primarily by the hydrologic regime at the time of dam removal and secondarily by the ability of vegetation to establish and stabilize the channel form. The importance of seasonal site hydrology over the minimum time required for channel evolution and the process through which the channel evolves indicates the complexity of channel formation within the first year following dam removal. Existing channel evolution models (CEMs) were modified and a new CEM that explicitly incorporates local hydrology and vegetative growth in the channel evolution process is presented. The modified CEM is applied to dam removals in Illinois, Virginia, and New Hampshire to illustrate its application beyond the immediate study area. In all cases, the seasonal flows were a dominant factor over the time frame and process of channel evolution. The CEM can be used to improve predictions of the length of time needed for a channel to evolve and the magnitude of channel change during the evolutionary process, such that it can be used to aid in planning and facilitating dam removals in similar regions.

Coming from the zero inertia (ZI) equations, an analytical model to describe sheet flow phenomena with a special focus on rainfall runoff processes is developed. A slight modification of the ZI equations, which draws upon the concept of a momentum-representative cross-section of the moving water body, leads—after comprehensive mathematical calculus—to an analytical solution describing essentially one-dimensional, shallow overland flow. In a test series, the analytical ZI model is applied together with three numerical models, one based on the Saint-Venant equations, one on the kinematic wave equations, and another one on diffusion wave equations. The test application refers to a typical rainfall runoff situation, i.e., rather shallow overland flow on a hillslope as a consequence of excess rainfall. Contrary to the analytical model, the comparative analysis clearly shows the difficulties of the numerical solutions in terms of exactness and robustness when approaching typical shallow water depths. This problem of numerical models is tackled by applying small time and space discretization, which, however, comes along with higher CPU execution times. Besides the good computational efficiency and freedom of any numerical inconvenience, the new analytical model outperforms the numerical models for typical overland flow simulations. This particularly refers to a highly satisfactory fulfillment of the mass balance and a nearly perfect match of peak flow rates.

A stochastic Lagrangian approach is proposed for the analytical derivation of a longitudinal dispersion coefficient that accounts for both transverse and longitudinal flow field variability in straight-axis open channels characterized by longitudinal large-scale bed heterogeneity, that is, by longitudinal bed irregularities over a wide range of representative lengths, from the simple grain roughness to large undulations (up to order of kilometers), possibly related to topographical discontinuities or to extended and inhomogeneous depositional processes carried out by natural or anthropogenic agents. The resulting dimensionless expression, involving the average Chezy coefficient and the ratios of river width and longitudinal heterogeneity correlation length to the average flow depth, is obtained at the first order in the depth fluctuations as a time-dependent function given by the sum of three distinct components. The first main component is related to the transverse velocity distribution and would apply even for truly uniform flows; the second component comes from the moderate longitudinal nonuniformity and under certain conditions can reach values of comparable order of magnitude; finally, the third component, originating from the cross contribution of transverse and longitudinal velocity variability, is always quantitatively insignificant. While the numerical validation of the linearized formulation was already illustrated and discussed in a previous paper with reference to ideally two-dimensional streams, the present work includes, among other things, a comparison of the performances provided by the asymptotic version of the transversally generated dispersion component and three earlier empirical or semianalytical formulas, based on field measurements documented in the literature.

A tidal bore is an unsteady flow motion generated by the rapid water level rise at the river mouth during the early flood tide under macrotidal and appropriate bathymetric conditions. This paper presents a study that physically investigates the turbulent properties of tidal bores. Results from some experimental measurements of free-surface fluctuations and turbulent velocities conducted on smooth and rough beds are reported. The free-surface measurements were conducted with Froude numbers of 1–1.7. Both undular and breaking bores were observed. Using an ensemble-averaging technique, the free-surface fluctuations of breaking tidal bores are characterized. Immediately before the roller, the free-surface curves gradually upwards. The passage of the bore roller is associated with some large water elevation fluctuations; the largest free-surface fluctuations are observed during the first half of the bore roller. The turbulent velocity measurements were performed at several vertical elevations during and shortly after the passage of breaking bores. Both the instantaneous and ensemble-averaged velocity data highlight a strong flow deceleration at all elevations during the bore passage. Close to the bed, the longitudinal velocity component becomes negative immediately after the roller passage, implying the existence of a transient recirculation. The height and duration of the transient are a function of the bed roughness, with a higher and longer recirculation region above the rough bed. The vertical velocity data presented some positive, upward motion beneath the front with increasing maximum vertical velocity with increasing distance from the bed. The transverse velocity data show some large fluctuations with nonzero ensemble average after the roller passage that highlight some intense secondary motion advected behind the bore front.

The experimental observations are presented on flow patterns and turbulence characteristics measured by an acoustic Doppler velocimeter within the developing (transient stage) scour hole around circular uniform and compound piers. Four series of experimental runs were conducted under the clear-water approach flow conditions. One experimental run was conducted around a uniform circular pier of diameter 114 mm, whereas the other three runs were conducted around a circular compound pier of diameter 114 mm and footing diameter 210 mm. In the series with a circular compound pier, the top surface of the footing was placed at three different elevations with respect to the general level of the channel bed, i.e., above the bed level, at the bed level, and below the level of the channel bed. Detailed measurements are presented on components of time-averaged velocity, turbulence intensity, and Reynolds shear stress around the pier in radial planes at 0°, 90°, and 180° from the flow axis. Flow structure around a circular compound pier in the presence of a scour hole was compared with flow structure similarly observed around a circular uniform pier by utilizing the observations made in radial planes at 0°, 30°, 60°, 90°, 120°, 150°, and 180° from flow axis. Diameter of the principal vortex upstream of the compound pier, when top surface of the footing is above the general level of the channel bed was 1.11 times as large as that for the circular uniform pier. However, size of the principal vortex is 0.85 times its size for the uniform pier, whereas the top surface of the footing was below the channel bed level. The components of turbulence intensities and Reynolds shear stress profiles in different radial planes are also compared around each of the pier models.

Turbulent suspensions of sediment are investigated to establish the characteristic length and time scales on which they adjust from one state to another. The suspensions are modeled by using a simple closure for the turbulent fluctuations in which the average flux of sediment is treated as a diffusion process. A key dimensionless settling parameter, which is closely related to the Rouse number, measures the magnitude of the settling to diffusive fluxes of particles. It is shown how the length and time scales on which the suspension responds are a function of the settling parameter and the assumed form of the eddy diffusivity, and that the predictions are broadly in accord with laboratory experiments. It is further established analytically that, in the regimes of the settling parameter much greater or much less than unity, the timescale of response is independent of the form of the eddy diffusivity. This motivates the use of simple eddy diffusivity laws to provide generic insight to the unsteady evolution of complex suspension and sedimentation problems.

Crossbar block ramps are a common solution for a naturelike channel. Current investigations deal with energy dissipation processes on block ramps, where large stones are arranged on the complete slope. Contrary, crossbar block ramps build several basins by arranging stones in a row transversal to the flow direction. Lower stones guarantee a minimum water level in each basin and allow for an overfall for small discharges and fish climb capabilities. With increasing discharges, the water also overflows the large stones and the basin flow regime changes over a waved flow regime into a channel flow regime. This paper deals with physical and numerical crossbar block ramp models. The physical model allows for the variation of discharges from 1 to $50\mathrm{l}/\mathrm{s}$; numerical simulations are run up to $200\mathrm{l}/\mathrm{s}$ with varying slopes, crossbar heights, and distances. Water levels, velocities, and drag forces on single stones are measured. New approaches to relative energy dissipation and the friction factor to evaluate water depths for channel flow regimes are developed. Additionally, drag forces on single stones in various crossbars are measured and analyzed.

Stream stabilization designers are often faced with the challenge of selecting effective bank stabilization measures. The potential benefits of stream stabilization measures can be economic, environmental, or social. Depending on the level of the potential benefit, a designer may be willing to take higher risks in implementing a given measure. A risk-benefit analysis is presented here that involves a qualitative analysis of risk and benefit (using failure modes and effects analysis) and risk and benefit quantification in terms of cost. The initial result of the method is the establishment of risk priority numbers (RPNs) and benefit priority numbers (BPNs), which provide a relative qualitative measure of the potential risk and benefit and can be used to prioritize and rank measures. The results of the qualitative analysis are then used to estimate risk and benefit quantitatively in terms of cost. These quantitative values are then compared for several stabilization alternatives to provide justification and guidance on selecting the most effective alternative. The risk-benefit method is applied to select a stream stabilization measure for a stream rehabilitation project in Indiana.

Sills and other appurtenances are often used to aid the formation of a hydraulic jump in a stilling basin. The drag force acting on a sill in forced hydraulic jump depends on the initial Froude number ${\mathbf{F}}_{\mathbf{1}}$, the height of the sill $S$, and its distance $x_s$, measured from the toe of the jump. The forward flow in the main body of the hydraulic jump behaves like a wall jet. Therefore, based on observed data, simple relationships were developed to predict the maximum velocity $U_m$ and the half-width $b$ of a free hydraulic jump. These are coupled with a dynamic pressure model that expresses the local drag coefficient $C_{d\delta }$ of a sill in a plane turbulent wall jet. The drag coefficient for the sill in a forced hydraulic jump $C_d$ was obtained by a simple transformation involving the wall jet parameters.

Vortex separators are increasingly utilized to remove suspended solids from stormwater runoff. Understanding flow patterns in a confined vortex chamber is crucial when investigating the mechanisms of liquid-solid separation. The angular velocity of fluid motion around the vertical axis is a dominant parameter for determining fluid flow patterns and particle trajectories in a rotational flow field. In this paper, an angular velocity formula, based on the angular-impulse momentum principle coupled with the Rankine vortex model, is derived under steady flow conditions. Comparison of predicted results using this formula to experimental data from a previous study indicates a general agreement. The derived formula can be used to estimate the angular velocity of the rotational flow field in vortex separators with tangential inlet and without internal components.

This paper deals with the study of the geometry of the scour hole and topography of the bed around a T-shaped unsubmerged spur dike located in a 90° bend. The experiments were carried out in a channel with a 90° bend. Uniform sediments having an average diameter of 1.28 mm were used under clear-water scour conditions. The effects of parameters like the length of a spur dike, the wing length of a spur dike, the location of a spur dike in the bend, the radius of bend, and flow intensity on the scour around a T-shaped spur dike were investigated. A new equation for scour parameters at a T-shaped spur dike is developed.