Journal of Hydraulic Engineering

A model is proposed for the surface erosion of cohesive soils on the basis of the physical interaction of turbulent bursts with the sediment bed. More specifically, the model deals with floc-by-floc erosion, appropriate for conditions of relatively low shear stress. The role of turbulent burst structures on the entrainment of particles has been well-established. A number of models have been developed for soil erosion on the basis of the interaction of turbulent burst structures with sediment beds in open-channel flow. What has been lacking in this type of model and in other models for the surface erosion of cohesive soils in general is an understanding of the aggregate size distribution in cohesive sediment beds. In this study, a relationship is developed for the aggregate size distribution as a function of the bed bulk density and fractal dimension of aggregates in soils, and this relationship is incorporated in a turbulent burst erosion model. The model is applied to experimental data, and the resulting model parameters are found to compare closely to the physical values reported in the literature.

A model was developed for the prediction of the density functions of entrainment and deposition rates of nonuniform sediment. The model incorporated both statistical and deterministic parameters in its formulation. The model parameters were related to the hydraulic conditions of the flow and the properties of the sediment mixtures using dimensional and multivariate regression analyses. Laboratory experiments were conducted to identify the shape of the density functions for various grain-size fractions and to advance the theoretical formulations of the model. The experiments were also used to validate and estimate the model parameters. The experiments have shown that the density functions of the total entrainment and deposition rates can be approximated quite satisfactorily with the normal distribution curve; however, the density functions of the individual fractions within the sediment mixture depart from the normal distribution curve assuming various distributions. Therefore, the normal distribution equation was modified by introducing a bias-function to adjust the normal distribution curve to fit the distributions of various grain-size fractions within the sediment mixtures. The proposed model approximated satisfactorily the density functions of various grain-size fractions of the sediment mixtures. The deviation between the measured and predicted values was less than 25% for most of the fractions, which confirms the validity of the proposed approach for the prediction of the entrainment and deposition rates of various fractions.

The results of an experimental investigation on the generation of transverse surface waves inside an array of randomly placed circular glass cylinders are presented. The transverse waves are generated from the forces developed by periodic vortex shedding from the glass cylinders. A lock-on, or a coupling, can occur between the vortex-shedding frequency and the frequency of transverse water waves. In all the experimental runs, the amplitude of the periodic transverse waves initially increased with an increase in water depth but decreased rapidly after reaching a maximum value. The amplitude reached a maximum value at a critical value of the water depth. The maximum amplitude increased with an increase in cylinder diameter. For the lengths of all of the arrays, the maximum amplitude occurred at approximately one-fifth of the total length from the upstream edge of the array. The maximum amplitude across the array increased with increasing Reynolds number until a critical value of the Reynolds number is reached, for which surface waves form and the amplitude is at its maximum. This critical value of the Reynolds number and the range of flow velocity for which the surface waves exist are different for different cylinder diameters. A previous mathematical model developed for the prediction of amplitudes has been modified for random arrays, which agrees well with the present experimental data.

A modification of the governing equations for unsteady pipe flow is proposed based on the results of many experiments carried out for steel and HDPE pipes of various lengths. Strong damping and smoothing of the pressure wave observed in the experiments suggest that these effects are caused not only by the liquid viscosity, represented in the governing equations by an algebraic term, but also by some other processes that are not described by these equations. The phenomena observed in physical experiments, such as time-varying pressure wave celerity and smoothing of the wave front, cannot be reproduced by the standard unsteady pipe flow model even if various modifications of the formula for shear stress available in the literature are applied. For this reason, an attempt to take into consideration additional dissipative processes was undertaken. On the basis of an approximate model of the elastic behavior of the liquid and wall pipe material, an approach accounting for the variable pressure wave celerity is proposed. Moreover, a diffusive term reproducing the irreversible processes at the pressure wave front is introduced into the dynamic equations. The numerical tests showed that the proposed approach leads to better agreement between the computational and experimental results.

The application of a two-dimensional (2D) finite volume numerical model to real flood events in the Ebro River is presented. The hydraulic model used is based on the 2D transient shallow-water equations on the irregular bed that are able to compute flow advance over a dry bed. This study involves the reliable simulation of not only the flood wave advance but also the drying process in a series of events of different magnitude. The importance of the correct characterization of the roughness coefficient and the topography is emphasized in the study. The former is estimated from a previous classification of structurally homogeneous habitats, and the latter is defined by merging the digital terrain model data with a hydraulic river bed elevation reconstruction algorithm. The calibration of the full model resulting from the roughness, bed river, and flow simulation models is based on field measurements of the flooded area for two steady discharges. The validation is performed by comparing the numerical results with the water levels measured during five flood events at certain times, with the flooded area and time series of continuous point measurements of water depth during different situations throughout the year. Because the model provides correct predictions of the surface processes both for low and high flow discharges, the simulation results are used to analyze the present floodplain hydrodynamics. In the same way, different topographic scenarios, on the basis of changes in the hydraulic river-floodplain connectivity, are generated to analyze their potentially beneficial effect in the floodplain geomorphic dynamics.

An engineered containment facility (ECF) that would cap and contain some of the most contaminated sediments in Hamilton Harbour has been proposed. Because of the size and shape of this structure, the ECF will alter local flow hydraulic conditions and isolate local discharges and intakes from free interaction with the open water of the harbour. To investigate the impact and functionality of the ECF, a three-dimensional (3D) hydraulic model was adopted in this study. The model was verified against vertical velocity profiles measured with an acoustic doppler current profiler in the study region. Results of the verification reveal that under the driving force of wind, the best agreement between simulated and measured currents occurs in the surface layer. Considering the complex nature of the hydraulic conditions in the harbour, the model performs reasonably well. Overall, the model should provide a useful tool considering and comparing hydraulic conditions under different ECF configurations and the existing condition. The functionality of a newly created connection channel used for the purpose of improving local water quality was studied in detail with some unexpected discoveries.

Side weirs are widely used hydraulic structures typically designed and studied in the case of fixed bed conditions. In the case of subcritical flows, the hydraulics of side weirs can be modelled by using the classical De Marchi hypothesis. In the present work, a generalization of this hypothesis is developed for the case of side weirs over movable beds. Experiments showing the effects and feedbacks between the spilling discharge and the bed morphodynamics are presented. The application of the experimental observations to the generalized De Marchi hypothesis clearly show that the functioning of side weirs on a movable bed can be modelled by using this hypothesis. These findings could be instrumental for the design and verification of these structures.

Determination of local boundary shear stress is an important topic in hydraulic engineering. When attempting to determine this from a very thin boundary region, it is a difficult one because it requires special skills and instruments to treat the measured data. With the advent of new equipment, like the acoustic Doppler velocity profilers (ADVP), all the velocity profiles in the primary flow region can be simultaneously measured, and it is necessary to develop a new method to estimate the boundary shear stress using the primary flow data. This paper presents such a method called the momentum balance method (MBM). A theoretical relationship between the boundary shear stress and parameters of the primary flow region has been established. The results obtained from MBM agree reasonably well with other methods, indicating that the new method is workable in closed duct flows, and possibly its applicability can be extended to a wide range of boundary conditions.

Weir submergence can influence head-discharge relationships for weirs used in channel applications when high tailwater conditions exist owing to a downstream control. Weir submergence describes a condition in which the water level downstream of the weir exceeds the weir crest elevation. When a weir becomes submerged, the driving head required to pass a specific discharge over the weir can increase significantly relative to a free-flow condition. In this study, the effects of tailwater submergence on laboratory-scale piano key weir head-discharge relationships were evaluated experimentally and compared with previously published data for labyrinth and sharp-crested linear weir submergence. The results of this comparison show that for relatively low levels of submergence, the piano key weir requires less upstream head relative to the labyrinth weir to pass a given discharge. This increase in efficiency was minimal ($<6\%$) and was reversed at higher submergence levels.