Abstract of Papers - JHR Volume 34,
1996
|
ISSUE NO. 1.ON THE EFFECTS OF A LIQUEFIED MUD BED ON WAVE AND FLOW CHARACTERISTICS The interaction between waves and a fluid-mud layer is studied theoretically and experimentally in the present paper. In addition the effects of a net current, or a net current together with waves are investigated. Experiments were carried out in a wave-current flume using two artificial muds. It was found that turbulence intensities in the water layer decreased when a layer of fluid mud was present. The observed wave decay and the wave-induced velocities in the liquefied bed agreed well with the those calculated using a modified version of a model due to Gade (1958). The experiments also showed that observations or measurements made at a sidewall should be interpreted with care. EROSION OF LOOSE BEDS BY SUBMERGED CIRCULAR IMPINGING VERTICAL TURBULENT
JETS This paper presents the results of a laboratory study on the erosion of sand beds by submerged circular impinging vertical turbulent jets of water for the erosion parameter Ec less than 5. The variation of the maximum scour depth with impinging distance was studied and this revealed two major flow regimes referred to as the Strongly Deflected and the Weakly Deflected Jet Regimes. The conditions under which they existed were established. The asymptotic eroded bed profiles were found to be similar except at very low values of the erosion parameter when made non-dimensional using the maximum scour depth as the length scale. Another length scale was introduced to achieve similarity for all values of the erosion parameter. Using results from this study and the literature, semi-empirical equations have been developed for the characteristic lengths of the erosion profiles in the asymptotic state. Equations that may be used for different values of the relative density difference / are proposed for the maximum scour depth and the scour hole radius using data from this study and from the literature which included data on air jets impinging on loose beds of sand and polystyrene. TRANSFER FUNCTION FOR THE DEPOSITION OF POORLY SORTED GRAVEL IN RESPONSE
TO STREAMBED AGGRADATION As a river bed of heterogeneous gravel degrades, previously buried material becomes exposed on the surface and thus available for transport. As the river aggrades, surface material and bedload becomes buried, resulting in a transfer of material to the substrate. Any predictive treatment of bed variation in gravel-bed rivers must account for this process in terms of an appropriate transfer, or exchange function at the interface between the surface layer and substrate. In the case of degradation it has long been recognised that substrate is incorporated directly into the surface layer. In the case of aggradation, however, existing formulations are based on little more than educated speculation. Here data form a large-scale experiment on the aggradation and selective deposition of gravel are used to back-calculate a form for the transfer function. The empirical relation so derived suggests that material transferred to the substrate can be represented as a weighted average of bedload and surface material, with a bias toward bedload. CHANGES TO WAVE PARAMETERS IN THE SURF ZONE DUE TO WIND EFFECTS Wind/wave interactions in the surf zone were studied using a wave tank and environmental wind tunnel. Waves were modelled at approximately 1/50 scale and broke on a model scale beach of 1/20 slope. The wind was simulated using accelerated growth techniques over a relatively short fetch, and a model scale of roughly 1/100 was achieved. Thus there was some scaling mis-match between wind and wave simulations. Results show that the wind has a significant effect on the breaking of the waves. Both breaker location and breaker type are affected, and the results agree with those of Douglas [10 and 11]. The main findings, common to both studies, were that offshore winds promote plunging waves and delay the moment of breaking (reducing surf zone width), whereas onshore winds encourage early wave breaking, generally by spilling. Wave heights at breaking were not generally affected, but due to changes in breaker depth, the height-to-depth ratio is shown to be affected by the wind. Additionally, studies of the motions of suspended particles, offshore of the surf zone, indicate wind effects on the drift velocities of these particles, although the precise nature of the wind effect was not clear. COMPUTATION OF FREE SURFACE FLOWS. PART III. STEADY SUPERCRITICAL
FLOW USING A GENERALISED LAGRANGIAN METHOD A second order Godunov-type nonoscillatory scheme is presented for solving the steady shallow water equations in generalised Lagrangian formulation and applied to compute steady supercritical flows. The Lagrangian distance and a stream function are used as the independent variables which have intrinsic flow adaptive property embedded. Numerical examples involving oblique hydraulic jumps, tangential discontinuities and negatives waves and their interactions are given. It is illustrated that all the flow discontinuities in steady supercritical flows can be accurately and effectively computed by using the generalised Lagrangian method in junction with the present high resolution scheme. A NUMERICAL MODEL FOR ASSESSING THE ADDITIONAL RESISTANCE TO FLOW
INTRODUCED BY FLEXIBLE VEGETATION The accelerating movement to restore wetlands, and especially to return flood plains to more natural conditions, leads to the widespread introduction and propagation of reeds and other forms of more-or-less flexible vegetation. These reeds and other types of vegetation then influence flows across wetlands, and the need to control flooding sequences and intensities using numerical models and other hydro-informatics tolls necessitates a better understanding of this influence of flexible vegetation on flows. A numerical model is introduced that serves as a means for investigation the influence of the several factors that contribute to flow resistance induces by flexible vegetation. The model has been partially calibrated from data obtained in experiments with inflexible vegetation. MODELLING MIMICO CREEK AS A SURFACE DISCHARGE A steady state, length scale model and a three-dimensional stratified hydrodynamic/ pollutant transport model were used to determine the pollutant distribution in a nearshore area of Lake Ontario, Canada, due to loading from Mimico Creek. The steady state model, named CORMIX3, is a knowledge based expert system intended for surface discharges subject to constant current flowing in one direction. The 3D time-dependent hydrodynamic/pollutant transport model can simulate wind-induced circulation in the bay under both isothermal and stratified conditions. Field data taken around the mouth of Mimico Creek were compared to the models’ dilutions for cases under isothermal and stratified conditions. CORMIX3 predictions agreed quite well with the 3D model’s predictions as well as the field of data. ISSUE NO.2.STANDING WAVES AT AN OUTFALL A standing wave phenomenon associated with outfalls is examined practically and analytically. The theory of jets is used to predict the wavelength. An extension of a theory for hydraulic jumps leads to predictions of the wave amplitude and the condition for wave breaking. GROIN AREA WATER EXCHANGE The influence of coastal structures on coast’s water quality is the subject of the present paper. Water exchange in the space between groins due to breaking wave induced currents is studied by experiments, numerical and hydraulic modelling. The results give the possibility to estimate the water quality for the area of the coastal zone, enclosed by the shoreline and the pair of shore-normal groins. That may be considered as an example of estimation of the effect of typical shore structure on the water quality. WALL SHEAR STRESS IN ACCELERATION AND DECELERATION TURBULENT PIPE
FLOWS The unsteady wall shear stress in acceleration and deceleration turbulent flows of water in a smooth pipe was measured using two different approaches simultaneously: a “direct” mechanical approach, and a more conventional “indirect” approach involving measurement of the transient pressure gradient. The measurements covered a Reynolds number range of between 40,000 to 130,000, and values of acceleration parameter () from -2 to 3. The results showed a surprising trend for the unsteady wall fraction to increase in deceleration, and decrease in acceleration flows, contradiction all existing observation. It is possible that the phenomenon of relaminarisation occurring during flow acceleration could have accounted for the reduction in wall shear stress. Based on the test measurements, empirical equations for the unsteady friction factor in terms of the acceleration parameter () are derived. Limiting values of within which the flow may be approximated as being quasi-steady are also proposed. NUMERICAL SIMULATION OF AGGRADATION AND DOWNSTREAM FINING Rivers typically exhibit a tendency for grain size to become finer in the downstream direction. Data for a set of large-scale experiments on the aggradation of heterogeneous gravel have recently become available. These experiments show substantial downstream fining over several tens of meters, Here a decoupled numerical model for bed aggradation and downstream fining is developed in an attempt to test an existing gravel transport model against the experimental data. Generally good agreement is found between the predictions and the observations in the absence of all but trivial adjustments to the gravel transport model. The same transport relation does not perform as well for a corresponding case of uniform sediment. In all of the experiments the Froude number was close to unity, a condition which would suggest that a decoupled model might break down. Coupled and decoupled models for uniform sediment are thus compared for a case with Froude number very close to unity. They are also compared for cases in which the upstream water discharge, sediment feed rate and downstream water surface elevation very strongly. The surprisingly good agreement between the two models suggest that concerns in the literature about the use of decoupled models may have been overstated. POINT BAR DEPOSITS IN NARROW SHARP BENDS Meandering channels typically deposit sediments near their inner bank to form point bars. Laboratory experiments with narrow sharp bends show that under identical flow conditions the configuration of point bars depends on particle size. Explanation of this grain-size effect is found in a three-dimensional particle stability. Despite favourable secondary currents, coarse grains tend to move toward the thalweg and coarse-grained point bar deposits are eroded as the side-slope angle approaches the angle of repose of the bed material. Calculated results coincide with experimental observations. Point bars are considerably smaller for coarse-sand deposits than fine sands. MODELLING NON-COHESIVE SUSPENDED SEDIMENT TRANSPORT IN 2D VERTICAL
FREE SURFACE FLOWS The mathematical model for simulation suspended sediment transport under equilibrium and non-equilibrium situations is described. Using the hydrodynamic variable field obtained from a refined turbulent free surface model, the suspension transport model involves bed boundary conditions for either the sediment flux to or from the bed, or an estimated concentration level, which are tested. The Schmidt number variability is also studied for several particle types. Some 2D vertical test cases for equilibrium situations or non-equilibrium situations with net erosion or net deposition flux to the bed are presented, and the predicted development of the concentration profiles is compared with measurements. FLUSHING SEDIMENTS THROUGH RESERVOIRS The control of reservoir sedimentation through hydraulic flushing has been employed in several parts of the world to sustain the useful storage capacity of the reservoirs. However, no comprehensive analysis has been conducted to understand the sediment flushing processes. In this study, laboratory experiments were performed to investigate the flushing processes during drawdown flushing, including outflow sediment discharge, characteristic of the flushing channel and flushing effectiveness. From both laboratory and field data, it is found that outflow sediment discharge can be well related to a hydraulic parameter which is a function of outlet discharge, water-surface gradient and flushing channel width. It is also found that flushing efficiency increases dramatically when retrogressive erosion emerges. A geomorphic relationship is used to estimate the equilibrium width of the flushing channel. A 1-D diffusion model presented herein is employed only to simulate the general trend of bed profile evolution and the amount of reservoir sediment removal during flushing in order to evaluate the applicability and limitations of this model. Simulated results of this diffusion model agree well with laboratory data in a narrow flume (with essentially 1-D flow) with nearly uniform flow condition after rapid drawdown operation. However, There is a 15% average error if this diffusion model is applied to a wide reservoir (relative to the width of the flushing outlet) as performed in our laboratory. A two-dimensional or three-dimensional model is needed to study flushing processes involving strong lateral erosion in a wide reservoir. ISSUE NO. 3,9th Ippen Award Lecture, 26th IAHR Congress, London, 1995, INTERACTION BETWEEN BASIC RESEARCH AND APPLIED ENGINEERING: A PERSONAL
PERSPECTIVE Examples from the personal experience of the author are used to illustrate the synergistic ways in which basic and applied research can interact in the field of hydraulics. While maintaining such a joint effort is not always easy, the intellectual rewards can be substantial. BENCHMARK EXERCISE FOR MULTI-DIMENSIONAL THERMOHYDRAULIC ANALYSIS
CODES A benchmark exercise for multi-dimensional thermohydraulic codes was carried out related to natural convection decay heat removal in liquid metal-cooled fast breeder reactors. A total of twelve computational methods were applied to the benchmark problem which simulated mixed forced and buoyancy driven penetration flow and thermal stratification phenomena. The applicability of turbulence models and higher order schemes of convection terms was examined, and a combined method was found to be highly effective among the group of finite difference methods. The importance of turbulence models was also recognised for the finite element method. Development of turbulence model applicable to the mixed convection flow regime was also discussed. A PRELIMINARY STUDY ON THE TURBULENCE STRUCTURE OF FLOWS OF NON-NEWTONIAN
FLUID The turbulence structure of flows of clay suspensions in smooth-boundary open channels were experimentally studied. Clay suspensions are non-Newtonian fluids. Transition from laminar to turbulent flows occurs when a Reynolds number Rem is between 2,000 an 10,000. Three layers can be identified in transitional flows, i.e. a) laminar sublayer, b) turbulent layer, and c) upper layer. In the upper layer the turbulence intensity is low or even equal to zero. The flow between the turbulent layer and the laminar sublayer is alternatively laminar and turbulent. For various concentrations of clay suspension the mean velocity profiles follow the Coles’ velocity formula with negative wake parameter and smaller Karman constant. The distribution of turbulence intensity is far more nonuniform than that of water flow. The probability density and autocorrelation coefficient are discussed and it is concluded that the turbulence in the non-Newtonian flow is produced in the turbulent layer and high frequency turbulence is suppressed by the high viscosity and yield stress. FRACTIONWISE CALCULATION OF BED LOAD TRANSPORT Extensive experiments conducted by the writers on fractional bed load transport rates are reported. These data as well as those available from other sources, covering wide range of flow conditions and sediment nonuniformity, have been analysed. Firstly, the semi-theoretical model proposed by Misri et al. (1984) is modified to evolve a method of bed load calculation and finally an empirical relationship is proposed incorporating the several parameters influencing the bed load transport of nonuniform sediments. DISCHARGE PREDICTION IN STRAIGHT COMPOUND CHANNELS USING THE MIXING
LENGTH CONCEPT A method for predicting the depth-discharge relationship in a compound channel is developed and applied to two different sets of experimental results. The method uses a mixing length formulation to account for the turbulent interaction between the main channel and the floodplain and the resulting momentum exchange. This momentum transfer tends to reduce the discharge in the main channel and increase the discharge capacity of the compound channel. As a result, practical methods which can allow for the interaction effect are needed. In this formulation a variation of Prandtl’s mixing length hypothesis is applied to calculate the apparent shear stresses, indicative of the turbulent interaction, on the sides of small vertical elements which comprise the compound channel cross-section. The approach suggested is to use the mixing length approximation to calculate the correction for the momentum interaction effects that are neglected when the traditional divided channel approach is used. The traditional divided channel into three large sub-areas: main channel, left floodplain and right floodplain using a vertical division at the edges of the main channel. THE DESIGN OF SEWAGE SLUDGE PUMPING SYSTEMS A project has recently been undertaken to improve the way in which sewage sludge pipeline systems are designed. A new, more accurate method of calculation pressure losses has been developed which replaces the traditional empirical formulae. The technique uses Computational Fluid Dynamics (CFD) together with the Generalised Bingham Plastic Model to describe the sludge rheology. Improvements have also been made to how the viscometry data is measured and processed. The method has successfully been verified against experimental data and validated against site measurements. Design issues have been addressed and presented in a simple design methodology. Overall, the work has had considerable success, as demonstrated by the fact that the technique is currently being used to design major capital investment schemes. HYDRAULIC JUMPS AT POSITIVE AND NEGATIVE STEPS ON SLOPING FLOORS A prediction model for computing the sequent depth ratio of the hydraulic jumps formed in sloping rectangular channel with positive or negative step has been developed. It is based on the application of the one dimensional momentum and continuity equations. The presented model contains two defined parameters. One of them to account for the effect of the channel slope and the second for the presence of the positive or negative step. The model can predict the sole and the combined effects of the bed slope and the step on the sequent depth ration of the hydraulic jumps. Large series of experiment data have been used to calibrate and verify the developed model for both sloping as well as horizontal floors. The predictions agreed reasonably well with theoretical and experimental results of other authors. PREDICTION OF THE TRANSITION NAPPE/SKIMMING FLOW ON A STEPPED CHANNEL On a stepped channel, two distinct flow regimes may occur: nappe flow for low discharges and flat slopes, and skimming flow for larger flow rates. A simple analytical method to predict to onset of skimming flow is presented. The method is based upon the change of momentum direction at the impact of the jet on the downstream step. The results of the analysis are compared with existing experimental data. The onset of skimming flow is expressed in terms of the initial Froude number and jet angel. The generalisation of the study results enables to predict the onset of skimming flow and hence the risk of jet deflection at the first upstream step.
ISSUE NO. 4,NUMERICAL MODELLING OF THREE-DIMENSIONAL SUSPENDED SEDIMENT FOR ESTUARINE
AND COASTAL WATERS Details are given herein of the development and application of a three-dimensional layer-integrated model to predict suspended sediment fluxes in estuarine and coastal waters. The model involves using the finite difference technique to solve the equations of mass and momentum conservation and the transport equation for suspended sediment. The operator splitting technique and a highly accurate finite difference scheme have been used to solve the suspended sediment transport equation. The model has been tested against analytical solutions and laboratory measurements for different flow types and boundary conditions, and has also been applied to predict suspended sediment fluxes in the Humber Estuary, UK. TURBULENT BURSTING-BASED DIFFUSION MODEL FOR SUSPENDED SEDIMENT IN
OPEN CHANNEL FLOWS Suspension of sediment particles has been found to be in close association with the bursting process in a turbulent wall-layer flow. This mechanism forms the physical basis of the present heuristic turbulent diffusion model for suspended sediment in open channel flows. It is formulated using the bursting intensity and the mean duration of turbulent bursts. The bursting intensity as the fractional vertical turbulence intensity contributed by bursting is determined from its percentage, estimated using existing measured data, among the total vertical turbulence intensity calculated using an analytical solution in the literature. A skewed parabolic profile is obtained for the turbulent diffusion coefficient of suspended sediment. The unique parameter involved in the model is calibrated using measured data spanning wide ranges of suspension indices and sediment concentrations. It is found to increase with particle suspension index, indicating physically that more energetic turbulent bursts with longer durations are needed for the suspension of coarser and heavier particles. The developed model is shown to perform satisfactory through comparison with independent laboratory data. Further refined models can be expected, given the enhanced understanding of the turbulent bursting phenomenon. PARAMETER IDENTIFICATION FOR SEDIMENT ROUTING IN RIVERS Sediment routing in rivers can be described in terms of three one dimensional equations: the Saint Venant continuity and dynamic equations of motion, and sediment mass balance. The solution of these equations requires two supplementary relationships for friction resistance and sediment transport, which contain empirical constants in one form or the other. It is often difficult to determine these empirical constants a priori with any degree of certainty. One method of determining the constants in a particular situation is to optimise on observed river data. This paper deals with such an investigation. In this investigation a novel coupled linear implicit method is used to solve the sediment routing equations and the sediment routing parameters are identified by using the gradient techniques of optimisation. These are applied to real river data for the Missouri River. GÉNÉRALISATION DES ÉQUATIONS DE SAINT-VENANT AVEC PROFILS DE VITESSES
ET CONTRAINTES Les équations de Saint-Venant qui gouvernent les écoulements non uniformes non permanents à surface libre sont obtenues par intégration partielle des équations de Prandtl pour des fluided newtoniens ou non newtoniens. Les équations sont intégrées par solutions affines au voisinage de l’écoulement uniforme permanent et étendues aux écoulements turbulents en moyenne. On donne pour les écoulements faiblement non uniformes non permanents les expressions littérales a) de l’équation dynamique de Saint-Venant qui est l’intégrale d’énergie cinétique avec le coefficient de dissipation d’énergie qui ne dépend que du nombre de Reynolds et de la rugosité relative, b) du coefficient de frottement variable dans le temps et dans l’espace et c) des profiles de vitesses et de contrainte eux aussi variables dans le temps et dans l’espace. ON THE DUAL HEAD-DISCHARGE CHARACTERISTICS OF A MODIFIED CHIMNEY WEIR This paper is concerned with the dual head-discharge characteristics of a modified chimney weir. It is shown by an optimisation procedure that the modified chimney weir having an inward trapezoidal weir over an inverted V-notch can produce discharges proportional to both the logarithm of the head as well as the linear power of the head reckoned independently over two different reference planes, within a prefixed maximum allowable percentage error from the theoretical discharge. As new technique is adopted to optimise the weir parameters, in order to obtain the maximum ranges of measurement under logarithmic, linear as well as combined characteristics. In the case of linear weir it is shown that it is possible to enhance the linearity range of the chimney weir by more than 540 % and for a weir with constant indication accuracy by more than 350%. In addition, about 86% of the overall depth of the designed linear weir and over 90% of the logarithmic weir is converted as the corresponding measurable ranges. Experiments with four typical weirs give consistent constant average coefficient of discharge for each weir confirming the theory. The practical application of the weir in minor irrigation, hydraulic and other engineering fields is highlighted. WATER-SURFACE SLOPE AT CRITICAL CONTROLS IN OPEN-CHANNEL FLOW A general equation is proposed for the determination of water-surface slope of a steady, gradually varied open-channel flow at a critical section. Thereupon the different types of singular points and water-surface slopes are given at such a section caused by changes in (1) bed slope, (2) channel width, (3) surface roughness of the channel, and (4) discharge, or their combinations. Although these factors are very different they have similar hydraulic effects on the occurrence of a certain type of singular point and water-surface slope. It is shown that a saddle type singular point is the only one which can become a control. The water-surface slopes at a critical control are evaluated using the proposed equation and are verified by the experimental results of Wilkinson (1974) for cases (1) and (2). A NEURAL NETWORK MODEL OF RAINFALL-RUNOFF USING RADIAL BASIS FUNCTIONS In modelling rainfall-runoff and flows in drainage systems it can be advantageous to adopt a neural network (NN). Unfortunately traditional NN learning procedures such as back-propagation can be very slow and expensive to carry out. However, if radial basis function (RBF) networks are adopted with radial centres fixed by a suitable data clustering technique then good results may be obtained very much more rapidly. RBF networks are here shown to be very effective in modelling runoff for a large rainfall database and to give broadly comparable results to those obtained by fine-tuning the much slower back-propagation procedure. The specific model is based on the assumption that runoff depends on time, rainfall intensity I, the rate of change of I and the integral of I. A METHOD FOR MODELLING DISPERSION DYNAMICS IN COASTAL WATERS USING
FRACTIONAL BROWNIAN MOTION Recent studies have shown that the trajectories of drogued drifters on the ocean surface have a fractal structure which may be described by fractional Brownian motion. A method is detailed which allows the generation of fractional Brownian motions within a two dimensional particle tracking ocean surface dispersion model. This technique enables the modeller to more accurately mimic particle paths, enhancing the prediction of diffusion and dispersion. ISSUE NO. 5AIR ENTRAINMENT AT DROPS This paper describes the results of a laboratory study of air entrainment below drops for a range of Drop numbers from 0.0007 to 0.0046. Characteristics of the plunging flow that occurs when the tailwater depth yt is in the range of yt* to h were studied where yt* is the value of yt when a hydraulic jump is formed at the section where the deflected stream below the drop becomes parallel to the bed and h is the height of the drop. Air entrainment below the drop was studied for the jump with yt = yt* as well as for the plunging flows and approximate expressions have been developed for their air entrainment rates. These results can contribute to the estimation of the reaeration of polluted streams below drops. A HYDRAULIC SHEAR STRESS MODEL FOR RAPID, HIGHLY CONCENTRATED FLOW A hydraulic shear stress model for a uniform, steady flow on a slope of a mixture of water and a granular material at high concentration is presented here. The model includes stresses generated by particleparticle interactions along with those generated by fluidbed interaction. Some results from the kinetic theory for granular flow are introduced to modify Bagnold’s model for dispersive stresses. The model was verified experimentally by means of an inclined channel where silicasand, supplied by a hopper, mixes with a uniform, steady water flow. Channel slopes range from 19 to 23 degrees. An interesting finding was obtained in the experiments presented here: there is not a continuous increase in velocity with depth of flow as stated by Manning’s formula for the whole range of concentrations. The hydraulic model is able to reproduce the shape of the velocity vs. flow depth curve and gives some insight about the mechanical behaviour of the flow. DEPTH RATIO OF HYDRAULIC JUMP IN RECTANGULAR STILLING BASINS A theoretical prediction for computing the sequent depth ratio of the hydraulic jump formed in rectangular sloping and roughened stilling basins is developed. It is based on the application of the onedimensional momentum and continuity equations. The derived equation contains two non-dimensional experimental coefficients, one for the effect of slope and the other one for the effect of bed roughness. When the two coefficients are used together, the combined effect of both slope and roughness on the sequent depth ratio of the hydraulic jump can be computed. Experimental data verify the developed equation reasonably well. A FULLY CONSERVATIVE BEAM AND WARNING SCHEME FOR TRANSIENT OPEN CHANNEL
FLOWS A fully conservative Beam and Warming scheme that uses Roe’s approximate Jacobian is presented for solving transient open channel flows. A general treatment for satisfying entropy inequality condition is incorporated which always leads to a physically realistic solution. Details of the Beam and Warming scheme, inclusion of the source terms and proposed modifications for achieving full conservation are presented. Improvement in accuracy achieved with the proposed treatment is demonstrated by comparing numerical results by different versions of the scheme with analytical solutions. The applicability of the scheme is examined by comparing numerical results with experimental data and numerical results from existing studies. The scheme’s response to the Courant number and Manning’s coefficient is also investigated. OPTIMIZATION OF WATER DISTRIBUTION NETWORKS In this study, a computer program is developed which enables us to design a water distribution system which satisfies all constraints including pipe diameters, low velocities, and nodal pressures with a minimum total cost. The hydraulic analysis of the water distribution system is performed by the NewtonRaphson method. This analysis is coupled with a nonlinear optimization technique in order to minimize the design total cost. Accuracy of the developed computer program is approved by applying it to branched water distribution systems for which the optimum design can be calculated analytically in a simple manner. Comparison of the results indicates that the accuracy of the method used in this study is quite satisfactory. A PERTURBATION SOLUTION TO THE TRANSIENT PIPE FLOW PROBLEM The analysis of hydraulic transients in pipes is traditionally carried out numerically. An approximate analytical solution of the nonlinear problem of unsteady flow in pipes is presented herein. The continuity and momentum equations are combined to give a nonlinear secondorder hyperbolic partial differential equation whose approximate solution is derived using a perturbation technique called the delta expansion. It is shown that the zerothorder perturbation solution gives ample accuracy for practical application. Useful series expansions for small values of the waterhammer parameter k are also obtained which compare extremely well with the results of the method of characteristics especially at small times. The explicit and simple form of the solution is suitable for further mathematical analysis and manipulation such as determining the maximal values and their time of occurrence. They are especially useful in determining the minimum time of closure possible while keeping the transient pressure in the pipe system within a specified limit. A NUMERICAL STUDY OF DIVISION OF FLOW IN OPEN CHANNELS A two-dimensional numerical model which employs the depthaveraged forms of continuity and momentum equations along with ke turbulence closure scheme is used to simulate the flow at the open channel divisions. The model is generalised to flows of arbitrary geometries and MacCormack finite volume method is used for solving governing equations. Application of cartesian version of the model to analyse the flow at rightangled junction is presented. The numerical predictions are compared with experimental data of earlier investigators and measurements made as part of the present study. Performance of the model in predicting discharge distribution, surface profiles, separation zone parameters and energy losses is evaluated and discussed in detail. To illustrate the application of the numerical model to analyse the flow in acute angled offtakes and streamlined branch entries, a few computational results are presented. MULTILAYER MODELLING OF THREEDIMENSIONAL HYDRODYNAMIC TRANSPORT PROCESSES A multilayer watersediment predictive model has been developed to study the effects of coastline reconfigurations on the threedimensional (3D) flow and sediment fields in coastal waters. The model is gridded with finite elements in the horizontal domain for easy adaptation to complex boundary configurations. The nonuniformity of flow and sediment distributions in the water column are represented by multiple vertically averaged layers. The LaxWendroff twostep scheme is used for time marching. This allows the flow field and sediment field to be explicitly coupled so that the density of the watersediment mixture and the flow depth are updated without matrix inversion for the advanced time level. Because of the adoption of the explicit timemarching scheme, the allowed time increment of the present numerical scheme is relatively small. However, this is beneficial to studying short life activities such as sediment entrainment and deposition. The model predicted results were tested against field measurements in two case studies and good agreement was found in each case. SPREADING OF THREEDIMENSIONAL INCLINED GRAVITY PLUMES Geometric and kinematic behaviour of threedimensional (3D) negatively buoyant gravity plume spreading on a steep sloping surface has been investigated in a laboratory tank. Simple analytical expressions are presented based on the balance of driving and resisting forces, i.e., gravity, buoyancy, inertia, and friction, in both the longitudinal and lateral flow directions. These analytical expressions for spreading depend on the bottom slope, the initial buoyancy flux, the Richardson number, and the geometry of the plume at the source. Similar relations for a twodimensional (2D) gravity plume are also presented for comparison. Mathematical solutions show satisfactory agreement with measured experimental data. Key words: Density current, gravity current, inclined dense plume, underflow, stratification, geometry, analytical expressions, modelling. FLOW STRUCTURE IN TURBIDITY CURRENTS This article presents a series of experiments with turbidity currents using two different types of sediments. During these experiments, the distributions of the velocity and of the concentration were measured simultaneously at three stations. The results show that the non-dimensional velocity distribution is relatively insensitive to large variations in the concentration distribution. The article proposes some characteristic ratios, which allow a rough estimation of the flow structure in the body of a turbidity current, based on average physical quantities which can be measured in practice.
ISSUE NO. 6SPECIAL ISSUE BY SECTION II.2 IAHR/IAWQ JOINT COMMITTEE ON URBAN WATER MANAGEMENT OVERVIEW OF URBAN DRAINAGE: ENVIRONMENTAL IMPACTS AND CONCERNS, MEANS
OF MITIGATION AND IMPLEMENTATION POLICIES Recent developments in urban storm drainage are reviewed starting with drainage impacts on receiving waters, followed by a brief analysis of selected components of urban drainage systems hydro-informatics, and regulatory programs. Further advancement in drainage planning will require considerations of the catchment, drainage system, groundwater, treatment plant and receiving waters as one entity, and ensuring sustainable integrated development and ecological enhancement of urban streams and corridors. Such processes should be based on comprehensive modelling, which is well advanced for hydrological aspects, but further development is needed in water quality, sediment quality, transport and impacts, and ecological modelling. Mitigation of stormwater impacts is accomplished by best management practices, which form a part of the treatment train, and their sustainability has to be ensured by proper operation and maintenance. Impacts of combined sewer overflow are mitigated either conventionally by storage (and return to the central wastewater treatment plant), or more recently, by satellite treatment or pollution separation at the overflow points. An improved utilisation of the available spare (idle) capacities in the drainage system can be achieved by real-time control. Finally, regulatory programs in various countries indicated a great deal of uncertainty and some aspects of such programs may even impede effective environmental practices. RAINFALL-RUNOFF PROCESSES AND MODELLING Hydrological studies of rainfall-runoff processes provide the basis for estimating design flows for urban stormwater drainage systems which control floods and the transport of sediments and pollutants. This paper outlines the theory of urban rainfall-runoff processes and describes the development of modelling practice and the current use of computer models. It concludes that rainfall-runoff theory has reached a mature stage, and highlights deficiencies and dilemmas which require further research including rainfall variability, scale effects and interfacing between models. MECHANICS OF SEWER SEDIMENT EROSION AND TRANSPORT Knowledge about the behaviour and effects of solids in sewer systems has advanced considerably in the past decade, with new guidance for the control of sewer solids and associated pollutants emerging to protect the environment and for the better operation of wastewater systems. Sediment transport in sewers, whilst being controlled by the same physical processes as prevail in natural waters, is affected much more by the limited physical scale. Transport is influenced by the cross-sectional shape in particular and limited sediment supply, with the effects of spatial and temporal variations in hydraulic conditions being particularly significant in sewers. This paper describes new theories that are providing the means to estimate when and by how much sediment beds will erode, how the particles and released pollutants will be transported, and as a consequence their potential impact on downstream systems. URBAN DRAINAGE IMPACTS ON RECEIVING WATERS The paper deals with the causes and consequences of urban storm drainage impacts on receiving waters and discusses available methods and tools for implementation of structural as well as non-structural measures to improve surface water quality from the deleterious effects of such intermittent urban discharges. Pollution of receiving waters must be understood on the basis of the characteristics and processes of controlling interactions within the urban water and wastewater system in order that an operational procedure for improved quality of the surface waters and adjacent corridor can be achieved. The main aspects of such a procedure involve water quality and biotic criteria as well as modelling systems which enable the user to link together the effect of changes within the catchment of surface water system with a specific quality and ecological improvement of the receiving waters. REAL TIME CONTROL OF WASTEWATER SYSTEMS The application of real time control (RTC) within a wastewater system may be useful if operational problems exist which vary in time and space and if idle capacity in the system is simultaneously available. In a first section in this paper general features of RTC systems are discussed. Typically a RTC system requires hardware that can work under harsh conditions. In addition, simulation programs are very useful when the RTC systems and their respective control strategies are to be developed. Finally, in a second section the development of the Malmö-Klagshamm RTC system is described in more detail. This is one of the few wastewater systems where integrated RTC of, both, the sewer system and the treatment plant is applied in full scale. STORMWATER DETENTION & BMPs The paper outlines the considerations involved in stormwater hydrological and pollutant mobilisation processes, and the selection of stormwater detention and pollution control Best Management Practices (BMPs). In addition, it provides a summary of the major BMPs commonly used, including material on their purpose, description and performance. A table is included providing broad indications of pollutant removal and flow attenuation capacities of each BMP. The non-structural range of BMPs summarised includes the modification of the use and disposal practices of household chemicals, land uses and management practices, on-site programs of runoff management, the management of pollutant build-up, and sewer infiltration management. The structural range of BMPs summarised includes infiltration and local disposal systems, retention or restoration of natural vegetated channels, inlet controls, detention basins, retention ponds and wetlands, and in-pond treatment. INTEGRATED DESIGN AND ANALYSIS OF DRAINAGE SYSTEMS, INCLUDING SEWERS,
TREATMENT PLANT AND RECEIVING WATERS Integrated design of urban drainage systems aims at the abatement of water pollution in the receiving water. This paper outlines the basic approaches in order to formulate and assess appropriate water quality criteria. The analysis of a hypothetical system is performed by means of a deterministic model of the total system. It is shown that the discharges of both the sewer system and the treatment plant into the receiving water have to be taken into account for the assessment of waterpollution. INFILTRATION PRACTICE FOR CONTROL OF URBAN STORMWATER Infiltration stormwater locally into the ground instead of discharging to conventional pipe sewers is increasingly considered as a means of controlling urban stormwater runoff. This paper reviews the most recent developments within this field and points out some of the major problems remaining. Easy-to-use methods for designing storm-water infiltration structures are available but methodologies for determining the design parameters based on local conditions and technologies for clogging prevention are needed. No evidence so far points at a high risk of groundwater contamination, but the quality of surface soils will decrease due to long-term infiltration of polluted storm-water runoff. Simplistic modelling approaches bases on conceptual process descriptions are needed for assessing the impact on soil and groundwater in local areas. The perspectives in relation to control of stormwater runoff are clear, the runoff peaks and volumes are decreased, and the urban hydrological cycle is returned to a more natural state. In some aspects stormwater infiltration is more effective for runoff reduction and abatement of pollution discharges than detention basins. In the future, measures to promote urban stormwater infiltration need to be developed. NATIONAL STORMWATER QUALITY REGULATIONS AND STANDARDS This paper reviews the storm water quality regulations of six major industrialised nations: The United States, Canada, United Kingdom, Germany, Australia and Japan, the order being set by the principal authors (both North America) who consider this to be a logical geographical order. The main purpose of the paper is simply to introduce the interested reader to these rules as a matter of broadening his/her interest knowledge base. Due to space limitations much is left out, but hopefully the essence of the laws and regulations are presented. No attempt has been made to compare and contrast these different results with the objective of determining which are the best rules or what the “best rule” might be. The reader will observe that the differences are great, and to a large degree reflect the culture of the individual country. |