A mathematical model based on unstructured meshes for analyzing 2-D flood inundation is applied to the Hanoi central area. The model application includes three cases: a real heavy rainfall analysis, an assumed exceptionally heavy rainfall analysis and an assumed dike break analysis. The simulation result of the real heavy rainfall analysis shows the general agreement with the observed data. The results of the last two analyses give an overview of flood disasters if they happen to the city. These results show that the applied model could be a useful tool for the city flood control and for the prediction of flood inundation process in any areas of interest.

View full abstract

Large Eddy Simulations have been performed for fully developed turbulent flow within and above explicitly resolved simple cube arrays. The model (LES-CITY) was validated in comparison with laboratory experiments. The systematic influence of cubic density on turbulent flow characteristics was investigated through numerical experiments in a wide range of cubic area density (0 to 0.44).
The following results were obtained. 1) The intermittency of flow: the temporally and spatially averaged flow structure confirms the existence of conventionally described canyon flow regime. However, the intermittency of the canyon flow for any cubic densities is large and the stream patterns are never persistent. 2) Turbulent Organized Structures (TOS): TOS are characterized by longitudinally elongated low speed streaks and corresponding streamwise vortices. The streaks in sparse and dense canopy flows are likely to be aligned to the street-line and to the roof-line, respectively.

View full abstract

Formal explicit filtering defined by the convolution integral over a flow region has been applied to derivefundamental equations of Large Eddy Simulation (LES) for turbulent flows over complex boundaries that are common in natural environment, in which details of the flow and the boundary geometry cannot be resolved in full and modelingis needed. Formal filtering in finite and complex domain allows filtering of the flow and boundary at the sametime and indicates that extra stress-like terms appear in the equations. DNS data obtained in a flow over wavy boundary are analyzed to examine the distributions of these terms and a similarity in the filtered flow that may be useful in formulating boundary conditions in a simulation calculation that does not resolve the details of the boundary. It is found that the filtering brings out the effects of the roughness of the boundary as additional stresses, most of which come from the contribution of the pressure forces on the original boundary.

View full abstract

Presently the simulation of horizontal turbulent motions in the flows is feasible and it reduces the dependence on turbulence-closures applied in URANS i.e. simulating 2D-turbulent motions with a dominant vertical vorticity component, called Large Eddy Simulation (LES). The idea is that kinetic energy has to be conserved because the part of the kinetic energy cascaded towards smaller scales of eddy movement requires an energy-controlling subgrid-scale (SGS) model. To connect this unresolved part of 2D-turbulent motions, the additional SGS model is coupled with our existing 2D model by modification of eddy viscosity. Hence the calculated results are compared with our experiment. The SGS model has been successfully applied to simulating a flow field and shows an improvement to our existing model based on experimental data especially at area of high turbulent fluctuation.

View full abstract

This paper presents a numerical simulation method for improving precision and efficiency of flood prediction. Conventional method based on staggered leap-frog approach for two-dimension flood flows often encounters serious mass conservation error due to minus water depth during flood calculation. To avoid it, the authors introduce a method of modifying outflow rate from a mesh to preserve mass conservation at any time of calculation. It enables us to use larger time steps in calculation than those used in the conventional method, so that significant time saving for prediction without losing accuracy might be attained. The proposed method is validated by comparing with the experimental results, through the examinations on the sensitivity of time step sizes, mass conservation errors and reproducibility. In addition, the Joganji flood which took place in Japan in 1851 is used to check performance of the method for a large field problem including steep slopes. The results show us that the proposed method can perform stable calculation with large time steps without mass conservation errors, while the conventional method fails.

View full abstract

A two-dimensional numerical model, which is particularly designed for simulating the flow of a steep river with complicated geometry and hydraulic structures, is developed. The model is based on finitevolume method with unstructured grid and flux-difference splitting. Firstly, the model is verified against the existing experimetal data of the supercritical flow in converging channel and sub-/super-critical flow with hydraulic jump in diverging channel. Secondly, the model is applied to a steep urban river, and the dangerous zones of inundation from the river are specified. It shows that the numerical simulations by the model are useful in the examination of flood control functions of the urban river with steep slope.

View full abstract

Excess agricultural drainage water in Egypt is mainly disposed into the sea through a complex network of drains. Recently, the area around the Edko Drain in the west delta region faced problem of frequent flooding by storm water during the rainy season. A solution for the problem is presented in this paper through 1-D numerical simulation for flood routing in a drainage network, where the water surface level is affected by the released drainage flow, the lake water level, and the tide in the Mediterranean Sea. From the drainage model, it is concluded that the influence of tidal flow is restricted in the lake area and not extended to the drain reaches. As for flood mitigation, lowering of drain water levels is suggested through cutting the local mound in the drain.

View full abstract

In late years, the techniques of the CFD for the free-surface flow are developing rapidly. On the other hand, study of the computer graphics (CG) to express a complicated natural phenomenon is conducted actively, and the methods to get realistic CG of natural phenomena with fluid motion, such as a cloud, a flame and a current, have been proposed. The representative of these methods in CG is the particle system. In this study, the simple motion rule of particle in the particle system is improved by introducing the MPS (Moving Particle Semi-implicit) method, which is a solver of the Navier-Stokes equation, to combine the Lagrangian physical model of fluid with the Lagrangian tool of CG. Furthermore, the sub-particle scale expression of mists is described by the Monte Carlo simulation of the mist particles, based on the concept of the DDM (discrete droplet model). Finally the physical CG of a falling water is shown.

View full abstract

The validity of the computational method for particle-laden liquid flows, MICS (Multiphase Incompressible flow solver with Collocated grid System), was shown through the comparison with experimental results in terms of unsteady wakes behind particles and interaction between particles and surrounding flows. In this paper, MICS is improved to the computational method for gas-liquid-solid multiphase flows in order to deal with the particle movements in the liquid flow with large deformation of the free surface. MICS is based on the computational method for incompressible flows, in which the treatment for density and mass conservation has been improved in the collocated grid. The external forces for the particles are evaluated from the results of the fluid computations and a model of the distinct element method (DEM). The MICS was applied to the particle movements in a free-surface flow and its applicability is discussed.

View full abstract

A fully-developed turbulent open-channel flow was computed to clarify the generation mechanism of inner secondary flow in a mixed-boundary corner formed by a vertical wall and a free surface using direct numerical simulation (DNS). The Reynolds number for the simulations was Re_τ=150 and the aspect ratio of channel width to flow depth was 2. Stable secondary currents and the maximum-velocity-diphenomena were reasonably reproduced. The significant results were the identification of inner secondary flow which rotates toward the wall at the free surface and down away from the surface along the wall. Then, in this paper, the Reynolds momentum equations were analyzed to discuss the generation mechanism. Detailed evaluation of Reynolds stress terms and pressure terms revealed that the imbalance between them causes the inner secondary flow.

View full abstract

This paper describes the numerical study on 3D turbulent flow structures around submerged spur-dikes. Effects of three hydraulic parameters, S/L (S: interval of two spur dikes, L: length of spur dikes), T/h (T: height of spur dikes, h: depth), θ(=inclination angle), are examined through the comparison of numerical results and previous experimental results focusing on both time-mean and time-dependent aspects. A non-liner κ-ε model is adopted as a turbulence model to reproduce the vortex shedding from a tip of a spur-dike and secondary currents. The numerical results show that, the flow near the bed, which is particularly important for sediment transport, is affected by both inclination angle and the ratio T/H. In the downstream inclined spur dikes, the unsteady vortex shedding is generated and the period is determined by the feedback effect of impinging shear layer. The frequency jump of the vortex shedding can be seen between S/L=2 and S/L=4 in the numerical results.

View full abstract

The Hanaguri channel, constructed by Kiyomasa Kato in Kumamoto, is an irrigation channel with uniquely shaped constrictions. A series of walls with orifices constructed crosswise in the channel was intended to prevent volcanic ash from sedimenting in this channel. Though the orifices are opened in the center of each wall now, it is also said that they were bored in right and left alternately at first. In this study, a numerical analysis of the flow through constrictions of the channel with such legendary orifices was carried out by means of LES. In addition, VOF method was used to deal with the variation in free surface area, and TVD scheme used to prevent calculation from being unstable due to shortage of grids at the place where flow changes rapidly. The results were in good agreement with experimental measurements; we found these methods to be effective for analyzing flow through constrictions such as those in an open channel with orifices bored in right and left alternately.

View full abstract

The clarification of 2D plane mixing shear layer in an open channel flow is important for various engineering field. The flow in a mixing shear layer is characterized by vortex formations due to the K-H instabilities. Therefore, it is necessary to predict precisely not only the time-mean flow patterns but also the turbulent features, such as turbulent intensities and Reynolds stresses for quantitative prediction of mass transport around a mixing layer. In previous many works, 3D computational methods with refined turbulence models have been developed mainly in mechanical engineering field. In shallow open channels, since the dominant flow pattern has basically plane 2D structure, the plane 2D depth-averaged model is applicable to some extent. In this paper, the depth-averaged 2D open channel equations with refined expression for Reynolds stresses are applied to the plane shallow mixing layer. The computations are performed under the conditions of the laboratory tests by Chu & Babarutsi (1988) and the applicability of the model is examined through the comparison between numerical and experimental results. The numerical results indicate that the present refined model can capture the fundamental aspects of flow phenomena.

View full abstract

Construction of a dam promotes the decrease of the sediment transport, and the resultant development of the stable armor bed in their downstream river. Under such condition, the river bed tends to be less disturbed and the eutrophication develops. This river bed situation allows the sessile algae growing thickly over the stable armor bed, which has a bad influence on the environment. One of the measures to remove the sessile algae is a sand sprinkling over the stable armor bed. The sprinkled sand, which moves in saltation mode over the stable armor bed with scratching the sessile algae, makes the sessile algae removed from the armor bed. In this paper, the numerical simulation for the exfoliation process of the sessile algae under saltaion mode was simulated by the movable bed simulator based on the three dimensional distinct element method. Furthermore, the exfoliation process due to collapse of the armor coat under a high bottom shear due to a flood is simulated.

View full abstract

An entrainment of air bubbles into water during wave breaking is the one of the key issues of the surfzone hydrodynamics. Therefore, recently, the numerical models to track the gas-liquid interface are being improved actively. The Lagrangian particle method, such as the SPH (smoothed particle hydrodynamics) and the MPS (moving particle semi-implicit) method, is one of the methods which can analyze the free surface flow based on the Navier-Stokes equation. The particle method shows good performance to simulate the fragmentation and the coalescence of liquid. The difficulties in the analysis of the gas-liquid interface by the two-phase flow model is the numerical instability due to the gas-liquid density difference. To overcome this problem, Koshizuka et al. developed the two-step pressure calculation algorithm, in which the pressure fields of liquid (=heavy fluid) and gas (=light fluid) are calculated separately. In this study, the Koshizuka model, which was applied to a vapor explosion, is modified to fit the phenomena treated in hydraulic engineering. The basic characteristics of the modified model are examined in the dam-break simulation.

View full abstract

A new scheme of numerical simulation for a flow with a free water surface is proposed. First, a technique of weight residual method using “reversed weight function” is presented, with which physical condition such as “pressure is constant at a surface” can be set inside the computational domain. As a result, flow with a free water surface on a complicated topography can be simulated on a coarse grid system, which contains the surface and the bottom of water inside. Then, the theory is formulated for FEM with rectangular regular grid system, and a demonstration is made for 2-D waves in a tank with a bottom undulation. The simulation result shows a good agreement with that from CCUP method though the grid of the former is far coarser than the latter's.

View full abstract

The aim of this study is to develop a new scheme to solve long waves shoaling on a slope by following the scheme which has been developed using Finite Element Method and cubic spline function. To simplify the computation code and reduce the computation time, piecewise linear and delta function were used as shape and weight function in elements where shoreline is not included, respectively. This means that time integration at each node can be executed explicitly like Finite Difference Method. On the other hand, piecewise linear function was used as shape and weight function in the elements including shoreline. The verification was done by making comparison between theoretical solution derived by Carrier & Greenspan and computation results. They agreed well when the number of mesh over a slope is more than 25 that corresponds to the case when the wave length is twice as long as the slope length.

View full abstract

A pressure computation method for incompressible fluid flows is investigated in the unstructured collocated grid system. The governing equations are discretized with a finite volume method. The CBP (Cell-Boundary Pressure) scheme, which was proposed in the structured grid, is used to calculate the cell-center velocity components in the prediction stage. The C-HSMAC method is applied to obtain the pressure distribution which makes the velocity field satisfy the continuity equation. In the unstructured grid, the pressure gradient used in the C-HSMAC method is evaluated on the cell boundary from the line integral around the neighboring two cells. In order to confirm the validity of the proposed computational method, it was applied to the flows in a cavity and meandering channels.

View full abstract

In the three-dimensional curvilinear coordinates, a fifth-order numerical scheme (FVM-QSI scheme) is proposed for convection equations which are discretized with a finite volume method. In the FVM-QSI scheme, fluxes on the cell boundaries are evaluated with quintic spline functions. In order to prevent numerical oscillations, the calculated fluxes are corrected with the flux control techniques (DC and FP methods), which are newly proposed in the 3D space in this paper. As a result of the computations of cubic cavity flows and its scalar convections, it was shown that the FVM-QSI scheme with the DC method enables us to obtain conservative and accurate results without numerical oscillations.

View full abstract

This paper concerns a Lattice Boltzmann Method (LBM) with non-uniform grids and a higher-order spatial interpolation method, which is applicable to incompressible flows surrounded with complicatedshaped boundaries. In order to preserve numerical accuracy in the non-uniform grids, CIVA method is employed in the interpolation process. The proposed interpolation-supplemented LBM (ISLBM) was applied to a laminar flow and the numerical accuracy was investigated. In addition, the flow in the channel with a side-cavity was numerically predicted with ISLBM. As a result, it was shown that the computational time can be largely reduced with sufficient numerical accuracy by the proposed ISLBM.

View full abstract

A new method of image analysis for measuring visualized two-dimensional flow field was developed. The method is basically based on the optical flow technique which is one of the image analysis techniques for quickly detecting moving objects, e. g. in the field of robotics. Conventionally optical flow technique utilizes an advection equation for image intensity field by assuming that local velocity is kept constant within a short time interval. In the new method, constant velocity components as well as velocity gradients are taken into account to improve the accuracy in measuring shearing flow fields. The performance of the proposed method was evaluated by using synthetic images with various velocity distributions and also actual surface flow images obtained during flood. The presented method has several significant advantages over the conventional PIV (Particle Image Velocimetry).

View full abstract

3D laser scanner used on this study can not measure the landform under the water. However, the absence of data shows existence of the water in the other. It is important for the elucidation of the deformation mechanism of the riverbed to grasp the landform of the riverbed in detail. In this study, we tried to discriminate the water route from the measured data by using the trimming algorism and reflectivity. The shoal and water route was elucidated by an analysis of the trimming algorism and reflectivity. Some important characteristics of the waters area in measurement data of the landform of the riverbed by the 3D laser scanner were showed

View full abstract

The morphology change of the Nanakita River mouth is monitored by the utilization of aerial photographs colleted since 1990 to 2002. The method for detecting land and water area in the photos is investigated by the application of image analysis in order to reduce time consuming for manual digitization. From the color information in the aerial photo, hue and lightness are appropriately used to detect land and water area. The criterion can be acquired by normal distribution function. The photo. analysis results show that the Nanakita River mouth behavior trends to stabilize and has less development of the left sand bar in the recent years. In this study, the influence from seaside is examined. Breaking of wave is calculated from the wave ray model then longshore sediment transport near the river mouth can be computed from CERC formula. The computed results show the decreasing of longshore sediment transport corresponding to the river mouth behavior. For this reason changing of longshore sediment is supposed to be one of the significant factors for stabilization and diminishing of left sand bar in the recent year.

View full abstract

A table-top hydraulic model is thought to be used at the place of project explanation or agreement formation. This study examines the phenomena, which the table-top hydraulic model can express qualitatively, and reproducibility of those phenomena. In examination, basic experiments are performed both in a curved flume and a straight flume, and experimental studies are done about the table-top hydraulic modelfor real channels of the Tonegawa down-stream region and the Kamanashigawa. The channel width of the model concerned is about 10-20cm. Results of experiments shows that for a field where flow is affected by the plane shape, such as curved channels, reproducibility is relatively observed even in the table-top hydraulic model, whereas for a field where flow is influenced by bed-form, reproducibility is low because of the influence of the model scale.

View full abstract

Recently, in order to obtain basic data for the integrated management of the sediment transport in a river basin, several monitoring of the bed load by indirect estimation in actual rivers called the hydrophone method have been conducted and examined for practical use. In hydraulic model experiments to evaluate the function and efficiency of installing sabo facilities in rivers, the bed load is measured by trapping directly the sediment using boxes, sieves, etc.
In the present study, we measured the temporal variations of pulse counts reflected from collisions the sediment particles with the various shapes of hydrophones which are laid in a rectangular channel. We then compared the results of the measurement of bed load at the downstream end of the geometric model by means of hydrophones with those obtained from the conventional way of measuring bed load by using sieve. As a result, it has been confirmed that the results of actually measured bed load are in good agreement with those obtained from pulses by mean of hydrophones and therefore that hydrophone is indeed applicable to hydraulic model experiments.

View full abstract

Float type towed vehicle is required for the field measurement of surface water quality distribution, especially in bay, estuary and other shallow stratified water body. The authors developed a new float type towed vehicle with a fin for submerging the body while running. The submerged body makes few bubbles which disturbs optical measurement; turbidity, chlorophyll, etc.
This paper represents the specification of this vehicle, characteristics of submergence and the comparison with the existent sinking type vehicle. The settings of fin angle, towing angle, length of towing rope and towing velocity define the submerging depth. The relationship between the velocity and the depth in some casesare clarified. Suitable settings for the measurement in required depth can be defined from it. This towing vehicle is more stable in constant depth than the existent vehicle.

View full abstract

Recently, the vegetation and sand bar have been preserved from the viewpoint of maintaining the nature of river. On the other hand, preserved natural environment in the rivers may disturb the flood flow. In the past, most of the vegetation and sand bar had been removed in order to prevent the damage of floods such as the overflow of the bank. Therefore, the control of vegetation and sand bar becomes an important river engineering problem. The purpose of this study is to propose an accurate numerical model which can examine the effect of vegetation and sand bar in the Asahi River. The numerical results compared to high water marks of the flood of typhoon No.10 in 1998 show the usefulness of the presented numerical model.

View full abstract

An acoustic doppler current profiler with low noise data and high resolution (HR-ADCP) recently developed has high potential to measure shallow-water flow like flood flow in river. In the present study, to examine the vertical structure of flood flow in river, we have attempted to apply the HR-ADCP into a flood-flow measurement in the Edo River. The observed data shows characteristic vertical structure of the velocity in the main channel that the relation between water elevation and velocity in the surface layer is quite different from that in the bottom layer. It is also noted that the friction velocity and vertical current profile in the main channel vary considerably in the rising and falling stages of the flood.

View full abstract

Near-bottom turbulence and flux of suspended sediment were measured in the Tone River Estuary to study a relationship of resuspension and settling of cohesive sediment with turbulence. An acoustic Doppler velocimeter was deployed because of its advantage of measuring suspended sediment concentration besides turbulence by utilizing acoustic intensity data. Vertical profile of velocity and spatial distribution of suspended sediment were measured with an ADCP and water quality meters.
Three phases for sediment suspension were observed: First phase was characterized by quasi-equilibrium under low Reynolds stress, second phase by resuspension of comparatively fine sediment accompanied by a growing turbidity front moving upstream, and the third phase by strong resuspension of comparatively rough sediment under strong Reynolds stress. Budget analysis of suspended sediment suggested that the settling velocity at the third phase was 10^-3[m/s] while the value at the first phase was 10^-5[m/s].

View full abstract

In the Tone River Estuary, turbidity maximum (TM) is formed in the reach of 5km below the estuary barrage. The TM region is covered by fine sediment having the mean diameter of 10μm. The sediment repeats resuspension and deposition following the variation of flow field. Time variation of SS distribution in vertical showed almost the same tendency at many points in the TM region. This fact means that the measurement and analysis in one-dimension in vertical can be applied to investigate the sediment behavior in the region. Sediment pick-up rate obtained from the one-dimensional analysis on the budget of suspended sediment was found to have a high linear correlation with the shear velocity obtained by the log-low fitting for measured velocity profile. The critical shear velocity for resuspension was shown to be a small value of 1.8×10^-3m/s, which was the same order as the value of ‘non-cohesive’ sediment. It would be caused by the surface bed sediment having a high water content due to the repeating motion of resuspension and deposition.

View full abstract

A long-duration observation of stratification has been conducted in the Ohta Estuary. During the spring tide, the data demonstrate a semi-diurnal stratification cycle driven by tidal straining of freshwater-induced horizontal density gradient. The wind-driven current induced by a land and sea breeze brings about diurnal stratification cycle during the neap tide. During the neap tide, the change of stratification is mainly caused by the variation of surface density Tn addition to the short cycle of stratification, the spring-neaps mixing cycle and the variation of river discharge bring about the longer period of stability. The current profiler mooring observation shows that the velocity gradient during the ebb is larger than the flood velocity gradient. This results in the stronger tidal straining during the ebb tide. The well-mixed condition around the low water slack is associated with the small depth and strong tidal stirring.

View full abstract

The wave propagation process caused by the penetration of long-period waves into the mouth was investigated through field investigation, taking the Segami River flowing into Hitachi Port in Ibaraki Prefecture as an example. Five wave gauges and current meters were installed around the river mouth and in the river in September, 2002. Long-period waves were measured during two typhoons and one storm. Long-period waves of the period 100-120s selectively invaded the river channel and wave amplitude increased two times that of the river mouth. Numerical simulation was carried out in order to reproduce the long-period waves and to examine measures against long-period waves. As a result, the effectiveness of the pond damping the long-period waves was confirmed.

View full abstract

This paper describes the application of the MEL1D-MB to duplicate the discharge and salinity at the upstream end of River Takase which has a large shallow water area in the middle of the river. Because of the large tidal amplitude compared to the maximum water depth of the river that is almost one third of the maximum water depth, it may be difficult to compute the flow field of River Takase by using ordinal three-dimensional model based on finite difference method. However, the model based on finite volume method may enable us to compute the flow field with high accuracy of the conservation of mass. Therefore, we decided to apply the MEL1D-MB based on finite volume method to compute salinity distribution in River Takase. The results obtained from the MEL1D-MB agreed well with the field observation results and the application of the MEL1D-MB is found to be possible to reproduce the salinity distribution in River Takase.

View full abstract

A permeable groin constructed with natural stones would be a nature-friendly river structure with minimum negative impact on natural environment. For making suitable design of the structure, hydrodynamics of the flow inside and around the permeable groin was investigated in the present study. Physical models of the permeable groin were installed in an open channel with various dimensions, porosity and rubble diameters. A numerical model was developed in order to simulate a two-dimensional flow field both in the open channel and the porous body. A flow resistance inside the groin was formulated by applying a non-Darcian formula that has already been verified in the authors' previous study on a rubble mound weir. Effects of grain diameter, porosity, geometry of the groin on the flow structure and flow force were discussed based on the experimental and the numerical data.

View full abstract

The flow structures around the continuous spur dikes were investigated by using a PIV method. In this study, the orientation angle of spur dike was chosen as a control parameter and it was varied in three ways. The visual images were recorded by a high-speed camera and analyzed with the cross-correlation method. Combining the two-dimensional PIV analysis of horizontal and vertical planes, we were able to verify the time-averaged three-dimensional flow structures. In the forefront section, obvious differences were observed among the three cases. The flow structure in the second spur-dike zone is controlled by the interaction with the separated flow from the first spur dike. In the backward section, the flow deflection as seen in the forefront section was not observed, but still slight differences are recognized. Particularly, the over flow velocity on the downstream-oriented spur dike becomes much larger than that in the other cases.

View full abstract

Groins are mainly used in curved channels to protect river bank against erosion. Arrangement of the groins has been investigated through laboratory experiments, numerical analysis and field observation in river. The groins are often set up at right angle or toward upstream from bank, because scouring is prevented in the downstream and sediment deposition occurs between the groins in these arrangements. On the other, it is important that the weight of groin materials is enough not so as to be washed away.
In this paper, flow and hydrodynamic force distributions for different groin angles accessed by experiments and 3-D numerical flow model and suitable groin angle is discussed.

View full abstract

Although many researchers have studied the local scour phenomenon around a spur dike experimentally and analytically, almost all the studies have been carried out for the steady flow state. This paper focused on a local scour around a spur dike during a surge pass and bed variations around the spur dike is investigated numerically by using a two dimensional numerical model with flux difference splitting technique under unstructured grid system. At first, the applicability of the employed numerical method was examined under the steady flow state. Secondly, rapidly varied flow such as surge flow condition was applied. The numerical results show good agreement with the experimental results.

View full abstract

It is very important for bride safety to prevent a local scour around a bridge pier. The down flows from horseshoe vortex have an important role on the scour around the pier. Local scours should be controlled if we can handle horseshoe vortexes. Saito et. al. have investigated the local scour around the pier with a horizontal rectangular groove that is called a horizontal riblet. They have reported that the horizontal riblet can reduce the maximum scour depth by 20-50% comparing with the depth obtained by pier with no riblet. However, the effect of the riblet on the prevention of the local scour decreases as time goes.
We have attempted to add not single riblet but many riblets to the pier in order to improve the effect of the riblet on the local scour and studied the characteristics of the local scour experimentally. It is found in this study that the pier having the riblets under the sand bed surface is most effective for the local scour.

View full abstract

Nature-friendly hydraulic structures such as a rubble mound weir have attracted attention in recent years. Therefore, it is important to establish the analytical model to analyze the flow around such a porous weir. The VOF method is used for modeling the flow around a trapezoidal porous weir and Porous Body Model is employed to express a porous part of the weir. The treatment of the resistance force is also examined to express the difference of permeability of a porous weir. The numerical results show that the VOF method can be applied to the flow over an impermeable trapezoidal weir. Moreover, the flow of the porous part was approximately obtained by using the Porous Body Model.

View full abstract

Bed protection works are constructed for protection of local scour and as energy absorber in downstream of falling works. If hydraulic gradient in bed material were over standard value, seepage control seat would be designed for piping under the falling works. The hydraulic gradient is estimated by use of difference of water levels in upstream and downstream of the river crossing structures. But the hydraulic gradient may increase by bed deformation in down stream of the river crossing structures after the construction. And in the case of low density of concrete blocks of bed protection works, pressure would come down by separate flow in the downstream end of apron.
In this paper, it is examined that causes of piping under falling works and countermeasures of gravels among bed protection works in downstream of apron by large-scale hydraulic experiment.

View full abstract

The bed protective work is installed to prevent structural damage due to bed scouring. However, bed scouring can also occur just downstream of the bed protective work and the last row elements suffers the largest hydrodynamic forces due to large water surface profile accompanied by a drop of water level. Therefore, we need to review the design method.
The purpose of this paper is to clarify the failure mechanism of bed protective works based on dynamical basis. First, we examined the failure mechanism of bed protective blocks due to drop in water level downstream of the blocks. Then, we measure the hydrodynamic forces acting on the last row elements in the threshold of block movement and consider the threshold of block movement through experiments.

View full abstract

It is important to evaluate the effect of vane works not only in the simple uniform bend but also in the reach of various forms in river. In this study we investigate the effects of vane works reducing scour depth in the compound channel whose plan form changes longitudinally. In this paper, at first we describe the idea of the design of vane works based on the recent topics of the flow and bed topography characteristics in a compound meandering channel, and show vane works fulfill their functions through flood from vane's effects in simple and compound channel flows. Bed variation in a channel with vane works is reproduced by using 3D-flow model considering drag and lift forces exerting on vanes.

View full abstract

Since a local scouring around spur-dikes or a groin decreases the stability of these structures, it is a very important problem on disaster prevention. Now, in the Meghna river in Bangladesh, the local scouring around the groin in the left bank of the just upstream of the Meghna bridge poses the problem. The feature of this local scouring is that the scouring hole is formed not around the upstream edge of the groin but in the more downstream by the influence of the low slope revetment. In this paper, we aim at the elucidation of the mechanism of this local scouring and discuss this phenomenon by the numerical analysis based on MacCormack scheme to the model river. The flow field is analyzed by 2-D and 3-D model, and the flow structure is discussed. Moreover, the local scouring is analyzed by 2-D model which can be applied to the bed evolution due to both bed and suspended loads, and the scouring process is discussed.

View full abstract

This study aims to clarify the flow structure in curved rectangular open channel with a side concavity. Surface flow velocity profile is measured using PIV technique and bottom flow direction is obtained by a new method using image analysis of bottom streamlines visualized by dye injection method. In curved channel flow, developing process of secondary flow is examined. The structure of channel flow with side concavity in each side is investigated and their differences are clarified. The results show that water level is lager in the case the side concavity is installed in the outer bank. On the other hand, surface fluctuation is greater in the inner bank case. The aspect ratio of channel width to water depth changed the structure of the flow around side concavity.

View full abstract

Relationship between the upstream flow depth and downstream one of a hydraulic jump with and without abrupt rise is derived from the continuity equation and momentum equation. On the other hand, such theoretical equations cannot give some information on the streamwise length of the hydraulic jump. In this study, the length of the hydraulic jump with abrupt rise was measured with changing both of the inlet Froude number and the ratio between the inlet flow depth and the height of abrupt rise by making use of a point gauge. The velocity measurements in and around the hydraulic jump were also conducted with two-components type of electromagnetic current meters. It was found that the length of the hydraulic jump with abrupt rise is affected both of the inlet Froude number and the ratio between the inlet flow depth and the height of abrupt rise. As a result, a new formula about the length of the hydraulic jump with abrupt rise is suggested.

View full abstract

The stepped channel is effective for the energy dissipation of the supercritical flow. Systematical investigation on the energy loss due to the stepped channel flows is required. As the flow condition in the stepped channel, Skimming flow, Transition flow, and Nappe flow are defined. But the clarification of the energy loss has been made only for Skimming flow. In this paper, the residual energy for various flow conditions has been investigated systematically. The effects of the step height, the channel slope, and the total drop height on the residual energy are clarified for various flow conditions. The comparison of the residual energy for various flow conditions shows that the difference of the residual energy due to the flow conditions is negligible for S/d_c≥0.5 (d_c: critical depth, S:step height). The relative energy loss for various flow conditions can be predicted. Also, the equilibrium condition for each flow condition is shown.

View full abstract

Hydrodynamic force exerting on a submerged square pillar in steady free surface shear flows were investigated experimentally. The ratio of pillar length l to pillar height k and the ratio of depth h0 to pillar height k were changed in the range of 0.5-2.5 and 1-2.5, respectively. It is found that the drag coefficient C_d of a square pillar is affected by h0/k, Reynolds number based on pillar size and Fruoude number when h0/k<2.0, while the values of C_d at given l/k becomes constant when h0/k>2.0. The value of C_d at given l/k in a shear flow are different from those in uniform flows. Diagrams to describe the relationships between C_d and l/k is presented.

View full abstract

Laboratory experiments were performed to study the wave forces acting on a model plant (Zostera Japonica Ascheras) on the bed of two dimensional wave flume. The wave forces acting on the both directions of wave propagation (in line direction) and gravity were measured by using a two components load cell installed under the bed of wave flume. In order to study the wave damping capacity of a crowd of model plants, theenergy dissipation per unit time due to a model plant were evaluated by measuring the wave forces and incident flow velocities.

View full abstract

Governing equations of debris flow include correction factors due to vertical distributions ofvelocity and sediment concentration which vary corresponding to bed slope. In this study, the correction factors in governing equations of a two layer model for debris flow proposed by the authors were examined under theequilibriumstate of the flow over a wide range of bed slope. Numerical simulations were carried out on simple conditions to examine effects of the correction factors on erosion and deposition process of debris flows. The term of sediment flux in a mass conservation of the two layer model modified by the correction factor influences stronglyon the erosion and deposition processes of debris flows.

View full abstract

Debris flow characteristics such as flow discharge, mean flow velocity, mean sediment concentration, sediment runoff volume and so on are usually predicted by solving the governing equations numerically. Formulas for erosion/deposition rate and flow resistance play important roles in mass and momentum equations. In addition, potential erosion depth and sediment correction factor also affect largely the debris flow characteristics.
The present study emphasizes the importance of the correction factor for sediment concentration and velocity profiles in predicting sediment transport rate and sediment runoff volume. According to Egashira et al.'s theory, sediment is concentrated near the bed, and this non-homogeneity increases with decreasing bed slope. This means the flux sediment concentration of a debris flow differs from the volumetric cross-sectional mean concentration. Correspondingly, sediment runoff volume simulated without a correction factor is very different from actual runoff volume. In this paper, we present a method for predicting the correction factor for sediment discharge, and discuss simulated results of runoff process of debris flow in a flume.

View full abstract