Abstract of Papers - JHR Volume 38, 2000
> issue 1 > issue 2 > issue 3 > issue 4 [incl. full papers] > issue 5 [incl. full paper] > issue 6[incl. full paper]

ISSUE NO. 1

Formation of arches and clusters of fine material in geotechnical filters
P.G. Martinet
Two-dimensional mesoscale experiments and a theoretical analysis show that the formation of clusters in a geotechnical filter is associated with fine particles that form arches between the grains of the matrix. The analysis of the stability of such arches indicates two critical parameters. The characteristic size of a cluster is obtained as a function of the uniformity of the matrix. If the matrix is uniformly spaced, the clusters occupy large domains, whereas if the matrix is non-uniform the clusters form in small pores.

Design of Kaplan Runner Using Multiobjective Genetic Algorithm Optimization
Andrej Lipej, Carlo Poloni
Numerical flow analysis in turbomachinery is an indispensable tool for water turbines design. Following the design of axial runner, the energetic and cavitation characteristics can be predicted using the numerical method.
In this paper it is described how the multiobjective genetic algorithm aids the human decision of the best design solution, based on objective functions obtained by numerical flow analysis. Theoretical and empirical knowledge of the design procedure enables to start the optimization procedure with good initial geometry. A very effective procedure for developing water turbines has been provided merging design procedure, numerical flow analysis and multiobjective genetic algorithm.

Numerical analysis of flow past submerged vanes
Sanjiv K. Sinha, Fredrick Marelius
Details of recently undertaken numerical study to analyze the physics of the flow past a submerged vane is presented. The numerical model solves the fully three-dimensional Reynolds-averaged Navier-Stokes equations in conjunction with the standard two-equation k-e turbulence closure. The governing equations are formulated in generalized boundary-fitted coordinates to accurately resolve the bed topography and the shape of the vane. The bed roughness effects are introduced by means of a two-point wall functions approach. The predictions from the numerical model are compared against measurements from an experimental study performed in a deformable-bed straight rectangular channel. Detailed experimental measurements of all three components of velocity, both in the neighborhood as well as in the far field of the vane are compared to judge the performance of the numerical model.

Computation of regime channel characteristics on thermodynamic basis
M. S. Yalin, A. M. Ferreira Da Silva
The present paper concerns the computation of regime channel characteristics in cohesionless alluvium. The computational method proposed rests on the channel formation criterion derived from the thermodynamic principles, and on the expression of the aspect ratio determined in the earlier works of the authors on the basis of zero cross-transport rate at the regime stage. The thermodynamic considerations are based on the first and second laws, and the Gibbs' equation: the alluvial stream is treated as an isolated, uniform system. The regime channel formation process is determined by the flow rate and the physical nature of the materials involved (alluvium and water): the sediment transport rate is viewed as an outcome of the phenomenon rather than as one of its characteristic parameters. The regime channel characteristics computed from the suggested method are compared with field and laboratory data of various sources.

Calculation of Unsteady Bore Diffraction Using a High Resolution Finite Volume Method
C.G. Mingham, D.M. Causon
Unsteady bore diffraction patterns generated by a bore wave impinging on a cylindrical pier and those generated by a bore wave passing through a contraction-expansion channel are simulated using a high resolution upwind scheme of the Godunov type on a boundary conforming grid. The two dimensional shallow water equations are formulated in integral finite volume form in such a way that no explicit transformation to curvilinear coordinates is needed. Unlike its finite difference counterpart, the method can be implemented on any desired mesh topology including structured meshes, unstructured meshes based on triangular cells or patched grids. It is demonstrated that truly nonsteady hydraulic bore wave diffraction involving regular to Mach reflection, bore-on-bore collisions and their resulting complex interactions can be accurately computed by the present method.

Taylor dispersion of contaminantsdue to surface waves
Adrian W.K. Law
This study addresses the longitudinal dispersive effect due to mass transport induced by progressive surface waves. Lagrangian drift is investigated to be the mechanism creating cross-sectional mean velocity variation, an essential component for Taylor dispersion. The dispersive effect is found to be more significant if the wave height is much larger than where D is the diffusivity and T the wave period. The longitudinal dispersion coefficient with a Stoke drift profile is first established. Subsequently drift profiles with viscous effect are considered and the dispersive effect increases substantially. The coupling of the oscillatory orbital motion of the pollutant particles under linear waves with background diffusion, is further examined through a random walk numerical approach. Results show that, despite the possible modulation of the probability density with the wavy surface, the magnitude of the dispersion is not affected by the inclusion of the orbital motion.

Shear Stress Partitioning and Sediment Transport by Overland flow
Joseph F. Atkinson, Athol D. Abrahams, Chitra Krishnan and Gang Li
Grain shear stress has been used in many earlier studies to estimate the sediment transport capacity of interrill overland flow. However, this procedure may be flawed because turbulent eddies generated by large-scale roughness elements contribute to sediment transport in this type of flow, whereas they do not in deeper (river) flow. In this study a procedure is developed for estimating the proportion of total shear stress contributing to sediment transport in overland flow. Application of this procedure to 1506 flume experiments representing a wide range of discharge, slope and surface roughness conditions reveals that the proportion of total shear stress contributing to sediment transport decreases from 1.0 to 0.83 as roughness concentration increases from 0 to 0.37. Concomitantly, the grain shear stress expressed as a proportion of total shear stress decreases from 1.0 to 0.14. Clearly, the proportion of total shear stress involved in sediment transport is higher than is indicated by grain shear stress. Consequently, use of grain shear stress to predict the sediment transport capacity of overland flow on rough surfaces will result in significant underestimation of the transport capacity.

The hydraulic resistance of sand streambeds under steady flow conditions.
Albert Rooseboom , Aldu le Grange
Laboratory and river data has been analysed in terms of applied power principles to determine the relationships between absolute roughness and bed deformation under steady state sediment transport conditions. It has been found that the full spectrum of bed conditions can be represented by simple relationships within a single system. As the bed forms grow, the unit stream power being applied along the bed decreases. The process of deformation continues until equilibrium is reached. Dynamic equilibrium is reached with the average rate of deposition of particles equal to the rate of re-suspension. By expressing the applied power in terms of the size of the boundary eddies i.e. the bed form sizes and comparing this with the power required to suspend the sediment particles, the link is obtained between absolute roughness and particle characteristics. Transition from lower regime conditions to upper regime conditions is initiated when the applied (laminar) power along the bed becomes greater than that which is required to maintain turbulent flow. Flow along the bed becomes fully turbulent, the (large) bed forms are rapidly broken down and fully turbulent conditions prevail until the applied turbulent power again becomes greater than that which is required for a laminar boundary layer. From this point onwards upper regime bed forms develop and equilibrium is again reached when the eddies that fit in with the bed forms grow large enough for the rate of entrainment of particles to be equal to the rate of deposition. A single diagram has been developed for determining flow resistance under all flow conditions as well as for identifying bedform types.

Dependence of dilution of a plunging discharge over a sloping bottom on inflow conditions and bottom friction
Xing Fang, Heinz G. Stefan,
A tributary flowing into a reservoir or lake is nearly always at a density different from the receiving surface water. The difference is due to temperature, dissolved and suspended materials. An understanding of density current phenomena is therefore essential to lake or reservoir water quality modeling. In this study the dependence of plunging characteristics (dilution, depth and location at plunging) on inflow conditions (densimetric Froude number and channel aspect ratio) and bottom friction is quantified by an integral jet model. The jet-like flow from a shallow channel over a sloping bottom into standing water was studied as an approximation of the flow of a tributary into a lake or reservoir. The integral jet flow analysis used similarity hypotheses for transverse (Gaussian) and vertical (power law) velocity profiles. The model especially includes momentum reductions by bottom and side wall friction and a flow development region, which were not addressed in other previous studies, and were found to have a significant effect on dilution characteristics. Depth and volumetric flow rate of the inflow were determined at the "plunge" point of negatively buoyant (sinking) inflows which in turn is characterized by a critical densimetric Froude number. These characteristics can serve as inflow boundary conditions for lake water quality or sediment transport models in reservoirs including density currents. The model predictions were validated against laboratory data and one set of field data. For slopes steeper than 0.5o, dilution at plunging (QP/QO) was found to increase linearly with inflow densimetric Froude number FO. Dilution is sensitive to the bed friction coefficient cf and the channel aspect ratio ARO, but insensitive to the bottom slope. Maximum dilution at plunging is given as a function of FO and (ARO cf).

ISSUE NO. 2

The Initiation Of Debris Flow At High Slopes: Experimental Results
Carlo Gregoretti

Debris deposit laying over an impermeable high slope surface (e.g. in steep streams and gullies) could develop a water sediment mixture flow (usually known as debris flow) when a high enough water flow is flowing over the sediment bed. In particular the rainfall first saturates the debris deposit and then mobilizes it by an overland flow.
The mechanisms which make unstable the granular material layer in such a case were studied by experimental tests performed in a tilting flume filled with uniformly distributed granular material.
The experimental results are compared with theoretical relationships on the occurrence of debris flow and with experimental findings pertinent to sediment transport phenomena.
The material employed in the experiments was nearly uniform gravel of three different sizes with mean diameters respectively equal to d = 0.023, 0.029 and 0.034 m.

Morphological Effects Of Bed Sills In Degrading Rivers
Roberto Gaudio, Andrea Marion, Vittorio Bovolin

Experimental results are presented about long term local scouring downstream of bed sills in monogranular gravel beds. Two sets of tests were performed in a tilting flume at HR Wallingford, UK, in 1996 and 1997, the main difference between the sets of tests being the size of sediments used. A dimensional analysis of the problem of local scour at bed sills was carried out and used to analyse the experimental results. The introduction of a new variable, the ‘morphological jump’, permits the derivation of two empirical formulas to predict the maximum depth and length of the scour hole at equilibrium. In addition, the Froude number seems to have no influence on the dimensions of the scour hole, but does have an effect on the development of a ‘sill step’ upstream of each sill. Non-dimensional scour hole profiles were found to be affine. As a consequence, it is possible to predict the location of the deepest scour relative to the bed sill.

Verification Of The Flow Separation In The Jet Regime
Luis A. Giménez-Curto, Miguel A. Corniero Lera,

We provide new theoretical arguments which together with recent measurements by Gong et al. [1996], strongly support the existence of the jet flow regime [Giménez-Curto and Corniero, 1996]. The theoretical arguments refer in particular to the introduction of a formal condition for flow separation from uneven surfaces for laminar and turbulent flow as well. This condition, which is a first requirement of the jet regime flow model, is compared with existing information.

Merging Jets In A Coflowing Ambient Fluid - A Hybrid Approach
K. L. Pun, M. J. Davidson

A multiple-point hybrid model for predicting the behavior of merging jets in a coflowing ambient fluid is presented. The hybrid model combines the modular structure of a length-scale model with the reduced empiricism of an Eulerian-integral model. Integral models of merging behavior are cumbersome because of the need to sum the momentum of individual jets during the merging process. A length-scale approach simplifies the model significantly, but details of the merging process are lost. A significant limitation of length-scale models is the assumption that the transition between flow regimes occurs at a single point. This difficulty can be overcome by allowing the transition to occur in stages and this leads to the use of the multiple-point approach. Predictions from the multiple-point hybrid model are compared with an equivalent single-point hybrid model and an Eulerian-integral solution for the same problem. The multiple-point approach significantly reduces transition errors when compared to the single-point model. These transition errors can result in inaccurate predictions of merging behaviour. Predictions from the multiple-point hybrid model are in good agreement with the integral solution.

Three-Dimensional Numerical Model for Open-Channels with Free-Surface Variations
E.A. Meselhe, F. Sotiropoulos,

A numerical model to calculate three-dimensional turbulent flow in open channels of arbitrary cross-section is developed and validated. The model solves the incompressible, Reynolds-averaged Navier-Stokes (RANS) equations, formulated in generalized curvilinear coordinates, in conjunction with the k-e turbulence closure. The free-surface elevation is determined by allowing the computational mesh to deform during the iterative solution procedure so that the proper kinematic and dynamic conditions are satisfied at convergence. The numerical model is validated by application to simulate the flow through meandering open channels for which detailed experimental measurements are available. Comparisons of the computed solutions with experimental data reveal that the model predicts the details of the velocity field, including changes in secondary motion, the distribution of bed shear, and variations of flow depth in both the transverse and longitudinal directions.

Two-dimensional Shallow Water Flows Simulation Using TVD-MacCormack Scheme
Ming Hseng Tseng, Chia R. Chu

This paper reports a finite volume TVD-MacCormack scheme for the computation of two-dimensional open channel flows with abrupt changes. The algorithm modified the widely-used MacCormack scheme by implementing a conservative dissipation step to avoid the unphysical oscillation in the vicinity of strong gradients in the numerical solution. Compared with other numerical models, this scheme does not bring in any additional difficulty in dealing with the source terms. Furthermore, this algorithm remains second-order accuracy in both space and time. A series of simulation, such as oblique hydraulic jump, circular dambreak and two-dimensional dambreak are carried out to demonstrate its robustness and stability in capturing strong gradients and discontinuities in open channel flows. The accuracy of numerical scheme is also verified with two laboratory dambreak experiments.

A Generalisation Of Prandtl's Model For 3D Open Channel Flows
W. Czernuszenko, A. A. Rylov,

The paper presents generalisation of the old Prandtl mixing-length hypothesis (MLH) for three-dimensional (3D) flows to provide engineers with a new tool to calculate the mean velocity distribution in turbulent open channel flows. These flows are usually not isotropic, i.e. turbulence may have different length scales in different directions. Hence, at every point of the flow a symmetrical second-order tensor with dimension of length should be given. The components of this tensor are specified based on the structure of the largest turbulent eddies. To demonstrate the usefulness of the proposed model, the Reynolds equations with the new turbulence model are solved and calculated streamwise velocity contours are compared with the measurement results. The importance of turbulent structure (sizes of the largest eddies and their growth rate) in forming the streamwise velocity contours is shown.

On The Influence Of Longitudinal Mean Flow Over Langmuir Circulations
Denis Dartus, Moacyr Araujo, Philippe Maurel & Lucien Masbernat

The Craik-Leibovich equations including vortex forcing terms and a pressure gradient force in x-momentum are integrated numerically. Some typical environmental problems with strong longitudinal currents are analysed.
Results are compared with Leibovich & Paolucci’s stability analysis which allows to identify situations where Langmuir cells exist. An establishing characteristic time for the setting of cells has the same order of magnitude as in field observations. Its variation with Langmuir number (La) and non-dimensional depth is pointed out.
A scaling analysis of the momentum equations is made for vanishing La. Influence of the longitudinal current intensities over Langmuir circulations structure is showed and confirmed by numerical results: more energetic convective cells are verified for situations where stronger longitudinal velocities are present. Results indicate how the redistribution of the shear stress induced by secondary flows is important for mixing in the water column and for coastal sediment transport.

Dissolved gas supersaturation downstream of a spillway II:Computational model
Joseph J. Orlins & John S. Gulliver

The increase in dissolved gas concentration downstream of hydraulic structures such as dam spillways can be harmful to most fish species. Such increases have been noted at numerous hydroelectric dams at the Columbia and Snake Rivers in Washington State, and there is fear that this total dissolved gas (TDG) "supersaturation" will increase mortality in juvenile and adult salmonids. Modifications to the spillway and/or its tailrace at some of these dams have or will be installed to help lower the concentration of total dissolved gas downstream of the dams. At Wanapum Dam on the mid-Columbia River, spillway modifications were designed and evaluated using a combination of physical and numerical models. The physical model provided information about the hydraulics associated with different spillway modifications. The numerical model calculated the concentration of total dissolved gas based upon hydrodynamic data from the physical model and mass transport relations developed for air-water flows. This article describes the numerical model development and application. A companion article (Mannheim and Weber, 1998) describes the physical modelling efforts made to evaluate the performance of the proposed spillway modifications at the dam.

ISSUE NO. 3

Integrated Two-Dimensional Macrophytes-Hydrodynamic Modeling
J. Morin, M. Leclerc, Y. Secretan and P. Boudreau,

Lake Saint-François is the first fluvial lake downstream of the Great Lakes. Aquatic macrophytes are abundant because of water level stabilization and increased nutrient loads. The influence of plants on flow modification is very important and must be considered in order to simulate hydrodynamic conditions. The spatial distribution of plant species is linked to their instream flow preferences. Several abiotic variables are of importance: light penetration, wave energy, current velocity, nutrients in substrate and physical characteristics of the substrate are controlling the species and their biomass.
Field characterization of macrophytes was performed using echosounder transects in association with a submersible video camera. This technique allowed the calibration of each echofacies for species identification and their relative proportion, height and density. The main eleven assemblages appear to be strongly correlated with abiotic conditions. A basic interpretation key was set up in order to describe plant distribution over the entire lake. Species, relative proportion, density and plant height were interpolated between transects. This information is used to adjust the Manning's friction coefficient for each assemblage. Simulations of the flow fields with plants and in absence of plants show a contrasted pattern. During the summer (with plants), the flow is mainly concentrated in deep channels where velocities are clearly increased by approximately 20% compared to spring-fall simulation (without plants).

A generalisation of Prandtl's model for 3D open channel flows
W. Czernuszenko, A. A. Rylov,

The paper presents generalisation of the old Prandtl mixing-length hypothesis (MLH) for three-dimensional (3D) flows to provide engineers with a new tool to calculate the mean velocity distribution in turbulent open channel flows.These flows are usually not isotropic, i.e. turbulence may have different length scales in different directions. Hence, at every point of the flow a symmetrical second-order tensor with dimension of length should be given. The components of this tensor are specified based on the structure of the largest turbulent eddies. To demonstrate the usefulness of the proposed model, the Reynolds equations with the new turbulence model are solved and calculated streamwise velocity contours are compared with the measurement results. The importance of turbulent structure (sizes of the largest eddies and their growth rate) in forming the streamwise velocity contours is shown.

Simulation Of Transient Cavity Flows Driven By Buoyancy And Shear
By Dragoslav L. Stefanovic and Heinz G. Stefan

The unsteady two-dimensional flow of a viscous incompressible fluid in a rectangular cavity, driven by effects of shear and/or buoyancy, is modelled numerically. This physical problem is studied from the viewpoint of obtaining accurate and efficient numerical solutions to the Navier-Stokes equations, applicable to the simulation of convective exchange processes in stratified water bodies, where flows are driven by forced (wind-induced) and/or natural (buoyancy-induced) convection, e.g. in lakes, ponds and reservoirs. The hydrodynamic model, based on the vorticity-stream function formulation, was found to be highly accurate and economical in obtaining iterative solutions to nonlinear coupled systems for a wide range of Reynolds/Grashof numbers and height/width cavity ratios that can be encountered in natural environments.

Numerical Simulation And Experimental Verification Of Dam-Break Flows With Shocks.
F. Aureli, P. Mignosa, M. Tomirotti,

New experimental results of 1D Dam-Break flows with shocks and results found in the literature are compared with those obtained by means of a numerical model based on the well-known McCormack shock-capturing scheme. In implementing the numerical scheme care has been taken to treat the source terms of the equations so that they are compatible with the resolution of shocks introducing artificial dissipation terms or Total Variation Diminishing (TVD) corrections.
To verify the numerical model under severe test conditions, laboratory experiments have been carried out in which shock formation, reverse flow and wetting and drying conditions in the flow field were induced. The good agreement between experimental and computed results confirms the validity of the numerical model even in limit conditions that could occur in nature, for which the St. Venant hypotheses are not completely verified.

Large Eddy Simulation Of Periodic Flow Characteristics At River Channel Confluences
K.F. Bradbrook, S.N. Lane, K.S. Richards, P.M. Biron, A.G. Roy

This paper describes an application of Large Eddy Simulation methods to the flow in a laboratory-style confluence of parallel channels of unequal depths (Reynolds No. *13,500), and a natural river channel confluence that also exhibits pronounced bed discordance (Reynolds No. The aim is to investigate the formation of periodic flow features inherent in the flow dynamics in river confluences. The laboratory-style confluence predictions are compared with experimental data, and suggest that the simulation was capable of capturing some of the key flow periodicities. The model provides detailed flow and mixing data which help inform understanding of the production and evolution of large scale turbulence features in confluences.

Evaluation of some approximate Riemann solvers for transient open channel flows
A. I. Delis, C. P. Skeels And S. C. Ryrie,

This paper constructs, presents and compares four approximate Riemann solvers for use with solving one-dimensional unsteady free surface flows. These solvers are usually the building blocks towards the construction of higher order accurate shock-capturing schemes. Details of the governing equations are presented. One dimensional dam-break flood wave propagations are analysed to demonstrate the applicability and the validity of the Riemann solvers. A treatment of source terms such as friction, slope and non-prismaticity of the channel is also presented. Computed results obtained with these solvers are compared with exact or published numerical solutions to define their characteristics and give an insight into the limitations of the methods, their relative strengths and weaknesses.

ISSUE NO. 4

An evaluation of directional analysis techniques for multidirectional, partially reflected waves: Part 1 numerical investigations.
Suzana Ilic, Andrew Chadwick, Jacob Helm-Petersen
download the full paper [in Adobe PDF format]

Recent studies of advanced directional analysis techniques have mainly centred on incident wave fields. In the study of coastal structures, however, partially reflective wave fields are commonly present. In the near structure field, phase locked methods can be successfully applied. In the far field, non-phased locked methods are more appropriate. In this paper, the accuracy of two non-phased locked methods of directional analysis, the maximum likelihood method (MLM) and the Bayesian directional method (BDM) have been quantitatively evaluated using numerical simulations for the case of multidirectional waves with partial reflections. It is shown that the results are influenced by the ratio of distance from the reflector (L) to the length of the time series (S) used in the spectral. Both methods are found to be capable of determining the incident and reflective wave fields when L/S>0.5.The BDM provides, in most cases, more accurate estimates of incident significant wave height, average reflection coefficients and directional spreading.

An evaluation of directional analysis techniques for multidirectional, partially reflected waves: Part 2 application to field data.
Andrew Chadwick, Suzana Ilic, Jacob Helm-Petersen
download the full paper [in Adobe PDF format]

Based on the findings of a numerical investigation, presented in the Part 1 companion paper, two methods of directional analysis, the maximum likelihood method (MLM) and the Bayesian directional method (BDM) are applied to over 80 field data sets. These cover a wide range of environmental conditions, for which multidirectional, partially reflective sea states exist. The results show that trends similar to those found using the numerical simulations are observed in the field estimates of relative predictions of incident significant wave height, average reflection coefficients, main directions and directional spreading. It is concluded that overall the BDM produces the more accurate results when applied to real sea waves.

Modeling the mixing of heated water discharged from a submerged multiport diffuser
Dae Geun Kim, Il Won Seo
download the full paper [in Adobe PDF format]

The mixing of heated water which is discharged from a submerged multiport diffuser was analyzed using the three-dimensional grid-based numerical model. The discharge momentum flux is properly incorporated into the model using the concept of a numerical slot diffuser, along with multi-jet theory. Plane or line patch was introduced to impose discharge momentum flux which plays an important role in mixing of the thermal effluent in the near field. The laboratory experimental work has been conducted to investigate mixing characteristics of the coflowing diffuser. A comparison of model simulations with laboratory experiments show that the proposed model properly simulates the shapes of thermal plumes and the distributions of excess temperature. The proposed model properly simulates velocity induced by a coflowing diffuser in proximity to a shoreline boundary, and this is confirmed by Lee's experiments.

Turbulent characteristics in a baffled contact tank
K. Shiono, E. C. Teixeira
download the full paper [in Adobe PDF format]

Turbulence measurements were undertaken using a two-component laser Doppler anemometer (LDA) in a serpentine contact tank (CT) commonly used for water chlorination. A detailed examination of turbulent parameters was carried out and showed the significant effects of the baffle lee and the inlet and outlet of the tank on the CT's hydraulic efficiency. Turbulence levels were quantified along the tank and were found to be considerably high in the tank inlet, first and second compartments and to be decaying rapidly in the latter compartments. The decay rate of turbulence along the tank is similar to that of the grid generated turbulence. The turbulence generation mechanisms in the first compartment were identified using comparative studies of the turbulence characteristics in the backward-facing step flow obtained by Eaton and Johnston (1981) and in the plane wall jet obtained by Rajaratnam (1976). The main contribution of the high turbulence level in the compartment was found to be due to shear generated turbulence in the upper layer. Tracer measurements were also carried out and an optimum dispersion coefficient at the outlet of the tank was found to be 16.2 . From the Marske and Boyle's equation (1973) in this tank, for b=37, the dispersion coefficient is 2.9 , which is one order smaller than that obtained from this study.

Prediction of flow patterns in local scour holes caused by turbulent water jets
O. A. Karim, K. H. M. Ali
download the full paper [in Adobe PDF format]

The paper describes a testing procedure for investigating the suitability of the FLUENT CFD package in simulating flow patterns, generated by a turbulent water jet impinging on rigid horizontal and scoured beds. The scoured beds were formed at various times of the scouring process.
The three turbulence closure models incorporated in FLUENT are: the Standard k-M Model, the Reynolds Stress Model and the ReNormalisation Group Theory-Based Model. Results obtained using these models were examined.
In general, flowfields, velocities and shear stresses predicted by FLUENT showed close agreement with relevant experimental results.

Experiments on suspension flow in open channels with bed forms
M. Cellino, W. H. Graf
download the full paper [in Adobe PDF format]

The influence of bed forms in open-channel flow on the suspended sediment concentration distribution is experimentally studied. At one single measuring section the flow and the concentration profiles have been investigated in order to put the effect of bed forms into evidence. From these measurements the sediment- and momentum-diffusion coefficients, eq. 4, have been calculated; subsequently the -value, which appears in the Rouse relation, eq. 2, could be evaluated.
With these results it could be shown that the concentration distribution, given by the Rouse relation, is altered due to the presence of bed forms. The -value for suspension flow over bed form, being , appears to be larger than the one for comparable suspension flow over a plane bed, being .
Also investigated is the evolution of the flow structure, measured along part of a bed form. The velocity distributions reveal a separating shear layer and a recirculating region. The turbulence and Reynolds-stress profiles show clear peaks in the vicinity of the shear layer.

On the negative weighting factors in the Muskingum-Cunge scheme
Sándor Szél, Csaba Gáspár
download the full paper [in Adobe PDF format]

The Muskingum-Cunge scheme applied to the one-dimensional unsteady advection-diffusion equation is investigated. To eliminate the numerical diffusion, the coefficients of the scheme are defined in such a way that the scheme does not contain the weighting parameters explicitly, but the Courant and Péclet numbers only. If one of the weighting factors is prescribed, the other should be necessarily negative in a lot of cases, which does not affect the applicability of the scheme. It is shown that the accuracy can be increased further, the numerical oscillations can also be eliminated by prescribing a simple relationship between the Courant and Péclet numbers. Sufficient conditions for strong stability are also presented.

Hydraulic jump in a suddenly widened circular tunnel
Bachir Achour
download the full paper [in Adobe PDF format]

The evolution of a hydraulic jump in a suddenly widened circular cross-section is investigated theoretically and experimentally. The flow upstream is not developed. A functional relationship is defined linking the various parameters which influence the phenomenon. The analysis makes it possible to compute the extent of the widening of the tunnel which is necessary to ensure development of the jump. The characteristics of the jump such as its length and axial surface profile are quantified.

ISSUE NO. 5

Sediment Re-Suspension By Turbulent Jet In An Intake Pond
Meilan Qi, K. Fujisaki and K. Tanaka
download the full paper [in Adobe PDF format]

Modern thermal or nuclear plants need cooling water from river or sea. Therefore an intake pond for the water delivery system is a requisite component where sediment frequently deposits. An impinging jet is a functional measure for sediment re-suspension and a precaution against sediment deposition. This paper predicts the consequences of this new measure using theoretical analysis and laboratory experiments. The jet diffusion discharged into an ambient flow field, and the jet velocity at bottom, are predicted by the governing equations using an integral method. Equilibrium scour is obtained after the process of scour asymptotically finished for a limited depth sand layer. The non-dimensional equation derived can predict well the particle re-suspension. Critical jet condition is obtained for the sediment re-suspension by experiments. The effect of the jet angle and jet height on sediment re-suspension is also included and discussed. The results obtained for the scour to a limited sand bed by an impinging jet through cross flow is especially significant to the engineering design of intake ponds and culverts etc.

Sand Wavelets In Laminar Open-Channel Flows
Stephen E. Coleman, Burkhard Eling,
download the full paper [in Adobe PDF format]

The results of experimental investigations indicate that sand wavelets can be generated from plane-bed conditions in open-channel laminar flow, the lengths, shapes and patterns of generation for these wavelets being consistent with observations for alluvial flows. Wavelets are of a preferred wavelength which is relatively insensitive to the characteristics of the applied flow and primarily a function of the size of the sediment, these wavelengths ? for alluvial and laminar open-channel flows over beds of quartz and lightweight sediments of size d = 0.2 mm to d = 1.6 mm being simply described by ? = 175d0.75, where ? and d are expressed in millimetres. The laminar-flow sand-wavelet data present significant implications to contemporary understanding of bed-form mechanics, with both ripples and dunes being postulated to subsequently develop from these wavelets for alluvial flows. The data also raise significant questions as to whether the generation of ripples and dunes in alluvial flows can be attributed to an organised structure of turbulence within the flow.

In-Stream Flow Impact On River Water Temperatures
Bashar A. Sinokrot, John S. Gulliver,
download the full paper [in Adobe PDF format]

The Central Platte River often experiences high water temperatures during sunny, hot summer days. A 128-km reach of the Platte River downstream of two hydropower dams (Kingsley Dam and North Platte/Keystone Diversion Dam) was studied to determine the relationship between river summer water temperatures and river flow-rate, and the impacts of in-stream flow requirements upon peak water temperatures. This reach serves as a habitat for eight federally listed or endangered species, as well as over 300 species of migratory birds, including 500,000 sandhill cranes and 7-9 million ducks and geese. Hourly water temperatures were simulated using a dynamic numerical model (MNSTREM) with and without in-stream flow requirements. It was found that a clear relationship exists between river water temperatures and river flow-rate. In addition, it was found that the occurrence of high water temperatures can be attributed to low river flow-rate and can be reduced, but not eliminated, with minimum in-stream flow requirements.

Investigation Of The Compressibility Of Extra-High-Velocity Aerated Flow
J. F. Zhao, W. Li
download the full paper [in Adobe PDF format]

Some fundamental topics of the compressibility influence of the extra-high-velocity aerated water flow in hydraulic engineering are studied in the present paper. A closed system of the simplified basic equations of the aerated water flow is deduced using the two-fluid model. A general formula of the sound velocity in the aerated water flow is derived using the sound analysis. It is found that the Wood adiabatic sound formula is a good approximation for calculating the sound velocity in the aerated water flow. Using the analysis of the order of the magnitude, it is found that the compressibility is important in the following cases. 1) The sound motion in the aerated water flow. 2) The aerated water flow with extra great depth. 3) The steady extra-high-velocity aerated water flow in which the mixture velocity is of the order of the sound velocity in it. It is shown that the adiabatic Mach number may act as the compressibility criterion in the steady aerated water flow. For the case of the steady homogeneous aerated water flow, a detailed compressibility standard is obtained, which is just the same as that in aerodynamics.

On The Effects Of A Negative Step In Pressure Fluctuations At The Bottom Of A Hydraulic Jump.
V. Armenio, P. Toscano, V. Fiorotto
download the full paper [in Adobe PDF format]

Experimental evidence of the statistical structures of turbulence pressure fluctuations at the bottom of hydraulic jumps over a negative step is brought out in this paper.
Special attention is paid to the definition of the extreme values and of the spatial correlation structures of the anisotropic field of fluctuating pressures in view of their relevance in the structural design of the lining in spillway stilling basins.
The analysis is performed in the case of B-jump and Wave-jump considering two different heights of the drop for Froude numbers ranging between 6 and 9.5. Moreover, the effect of the shape of the drop on the hydraulic jump has been investigated using the abrupt step as well as the rounded one. The experimental results herein reported may be helpful in the design of a stilling basin with a negative step, with particular reference to the thickness of the concrete slabs required to ensure the stability of the linings.

Implicit High-Resolution Methods For Modelling One-Dimensional Open Channel Flow
Delis, C. P. Skeels, S.C. Ryrie
download the full paper [in Adobe PDF format]

Three implicit high-resolution total variation diminishing (TVD) schemes are presented for solving the Saint-Venant equations. The applicability, performance and validity of these methods are investigated. Recently created benchmark solutions are reproduced for a wide range of cases, which include friction, non-uniform bed slopes, transitions between subcritical and supercritical flow, non-prismatic cross sections and hydraulic jumps. The tests produce satisfactory results without resorting to excessively fine grids. All the methods also produce satisfactory results for an idealised dam-break problem.

Modelling Flow Over An Initially Dry Bed
Abdul A. Khan
download the full paper [in Adobe PDF format]

The one-dimensional groundwater flow equations are coupled with the St. Venant equations to simulate the flow resulting from a sudden removal of a dam over an initially dry downstream bed. The St. Venant equations are solved using a Petrov-Galerkin finite element scheme, while the groundwater flow equations are solved using the Bubnov-Galerkin finite element scheme. The comparison of the computed water surface and discharge per unit width profiles with the corresponding analytical solutions, for a dambreak over frictionless horizontal bed, show that the model possesses excellent phase accuracy, for both positive and negative waves, and can predict the discharge distribution accurately. Also, the computed water surface profiles are compared with the available measured data for the dambreak flows over smooth and rough surfaces of horizontal and sloping channels. In addition, the results obtained from only the St. Venant equations with minimum depth criteria are presented for comparison with the above model.

Applicability Of Euler-Lagrange Coupling Multiphase-Flow Model To Bed-Load Transport Under High Bottom Shear
Abbas Yeganeh (Bakhtiary), Hitoshi Gotoh, Tetsuo Sakai
download the full paper [in Adobe PDF format]

Bed-load transport at high shear stress is numerically simulated with the aid of the two-phase flow model, in which Euler-Lagrange coupling of the governing equations of the fluid and sediment phases are implemented. Fluid phase is described by the vertically two-dimensional turbulence model in unidirectional flow condition, while the sediment motion is expressed by the numerical tracing of the saltating particles. The fluid/particle interaction, as the main interaction mechanism, is explicitly introduced into the governing equations in the present model. The characteristics of the mean-flow velocity profile of the saltation dominant flow, namely two-layer profile, is reproduced well with the present model. Experimental results show the existence of the three-layer type velocity profile under the high bottom shear, while the present model cannot reproduce such characteristics. The limitation of the assumption in the present model by neglecting the existence of the interparticle collision will be discussed on the basis of the results of the simulation.

A Simple Method for Measuring Shear Stress on Rough Boundaries
S. Wu and N. Rajaratnam
download the full paper [in Adobe PDF format]

This technical note presents a simple method for the real time measurement of bed shear stress with a LabView Program for turbulent flow over uniformly rough boundaries, based on the classical logarithmic velocity distribution equation. The method is based on a step-wise linearization of the additive coefficient in the classical logarithmic velocity distribution equation.

ISSUE NO. 6

Sixth Issue 2000 abstracts

Turbulent characteristics in a baffled contact tank
K. SHIONO, E. C. TEIXEIRA
download the full paper [in Adobe PDF format]

Turbulence measurements were undertaken using a two-component laser Doppler anemometer (LDA) in a serpentine contact tank (CT) commonly used for water chlorination. A detailed examination of turbulent parameters was carried out and showed the significant effects of the baffle lee and the inlet and outlet of the tank on the CT’s hydraulic efficiency. Turbulence levels were quantified along the tank and were found to be considerably high in the tank inlet, first and second compartments and to be decaying rapidly in the latter compartments. The decay rate of turbulence along the tank is similar to that of the grid generated turbulence. The turbulence generation mechanisms in the first compartment were identified using comparative studies of the turbulence characteristics in the backward-facing step flow obtained by Eaton and Johnston (1981) and in the plane wall jet obtained by Rajaratnam (1976). The main contribution of the high turbulence level in the compartment was found to be due to shear generated turbulence in the upper layer. Tracer measurements were also carried out and an optimum dispersion coefficient at the outlet of the tank was found to be 16.2 cm 2 /s. From the Marske and Boyle’s equation (1973) in this tank, for b = 37, the dispersion coefficient is 2.9 cm 2 /s, which is one order smaller than that obtained from this study.

Numerical simulation of sediment mixture deposition part 1: analysis of a flume experiment
Ph. BELLEUDY, SOGREAH
download the full paper [in Adobe PDF format]

This paper reports the numerical simulation of previously published laboratory experiments concerning deposition of a sediment mixture in a flume.The results of the simulation are analyzed and their interdependencies are discussed with reference to flume observations and previously published analyses. Cross-comparison of different parameters of the system (deposition rate, sediment transport, grain size) allows some validation of the system of equations of our modeling system. It may also give some clues and directions for further measurements and numerical experiments.

Non-uniform sediment transport in alluvial rivers
WEIMING WU, SAM S.Y. WANG and YAFEI JIA
download the full paper [in Adobe PDF format]

A correction factor has been developed in this paper to account for the hiding and exposure mechanism of non-uniform sediment transport. This factor is assumed to be a function of the hidden and exposed probabilities, which are stochastically related to the size and gradation of bed materials.Based on this concept, the formulas to calculate the critical shear stress of incipient motion and the fractional bed-load and suspended load transport rates of non-uniform sediment have been established. These formulas have been tested against a wide range of laboratory and field data and compared with several other existing empirical methods. The predictions by these newly proposed formulas are very good.

1D Mathematical modelling of debris flow
P. BRUFAU, P. GARCêA-NAVARRO, P. GHILARDI, F. SAVI
download the full paper [in Adobe PDF format]

Debris flow is modelled using the equations governing the dynamics of a liquid-solid mixture. An upwind finite volume scheme is applied to solve the resulting differential equations in one dimension. These equations have a structure similar to those of the monophasic water flow, differing from them by the presence of some terms characteristic of the bifasic nature of the mixture, such as granular bed erosion velocity, sediment concentration, bed shear stress, etc. The model and the system of equations to be solved are presented with the description of the implementation of the upwind scheme for the resulting hyperbolic conservation system. The numerical method is first order in both space and time. The treatment of the source terms is specified in detail and some comparison with laboratory experiments are presented.

Parameter quality conditions in open-channel inverse problems
RAHMAN H. KHATIBI, PETER R. WORMLEATON, JOHN J. R. WILLIAMS
download the full paper [in Adobe PDF format]

Some open-channel empirical parameters lack exact values due to (a) not having definitive measurement methods and/or (b) not rendering a "properly-posed" system of equations in the formulation of their inverse problems. The values of these parameters are therefore susceptible to data-errors, imperfections in governing equations or insufficiency of the gauged data used in inverse problems. Best values of such parameters in a mathematical sense can be identified by the implementation of inverse problems using optimisation methods but there are potential pitfalls. A comprehensive review of inverse problems is presented in this paper outlining some of the pitfalls in their implementation and illustrating the parameter quality conditions of "identifiability, uniqueness and stability". The uniqueness condition was approached by statistical methods in this paper. Other considerations of equal importance were found to include the compliance with the assumptions underlying optimisation methods and a careful selection of the objective function.

Key words: improperly-posed, properly-posed, empirical parameters, inverse problems, quality, assumptions, statistics

Bed slope effect on the dam break problem
BLAISE NSOM, KHALED DEBIANE, JEAN-MICHEL PIAU
download the full paper [in Adobe PDF format]

We consider the dam break problem upon an inclined plane. The flow of a series of Newtonian fluids is generated by the collapse of a dam in a completely transparent channel with variable slope. The stage at given abscissa and the front wave evolution are accurately investigated using ultrasonic and image analysis facilities, respectively. While the Navier Stokes and the continuity equations are properly rendered non dimensional in the viscous regime and solved with the 1D shallow water approximation. A successful comparison is shown between the experimental and the theoretical results.

Free-Surface Effects in Open Channel Flow at Moderate Froude and Reynold's Numbers
J.SHI, T.G. THOMAS AND J.J.R. WILLIAMS
download the full paper [in Adobe PDF format]

A large-eddy simulation of flow in an open channel has been carried out at the moderate Froude number, Fr = 0.66, and Reynolds number, Re = 20,800. A resolution has been used that allowed the simulation to be carried out on a workstation; thus enabling the results obtained to be viewed within an engineering context. The code used treats the free-surface exactly and allows it to freely deform with the underlying turbulence. The turbulent features found in this study are consistent with those found both by experimental observations and direct numerical simulations at lower Reynolds numbers where the free-surface has been treated as a stress free rigid lid.