In this study, the segregation and resistance law of the granular flow of mixed diameter particles are discussed from the viewpoint that the flow is the stochastic phenomena of the motion of each particle. As the results, we point out 1) the segregation occurs as percolation by the particles of the diameter dominating resistance law, and it is characterized by coefficient α on the arrangement of large particles percolating small particles and by drop-out probability P of small particles from its own layer, and 2) the resistance law is not equal, as the influence of mixing condition of different diameter particles on their moment are different, and by that, into 4 types velocity distributions are classified. In this manner, it is important that α and P are defined as stochastic quantity of the flow.

View full abstract

The authors observed the characteristics of suspended sediment in the Nagara River for three years since 1979. The suspended sediment consists of the suspended load and the wash load, and the former grain size composition is related to the bed material and its concentration is governed by the flow rate, the latter turbidity is governed by the upslope supply rate. The authors examined the suspending aspects of both the suspended load and the wash load by using measured field data. Namely, the characters of the suspended load have been examined under the idea of interchangeability with the bed material, and it has been demonstrated that the mean grain size is about 0.2mm to 0.3mm and standard deviation of grain composition is nearly constant. While a generative process of wash load has been examined with a simple mathematical expression, and the results show that the wash load is characterized by the upslope supply sources.

View full abstract

Alternate diagonal bars cause the increase in flow resistance and induced meandering flow over them forms a point of bank attack. It is, therefore, important subjects to investigate dynamic process leading to equilibrium bar height and to predict flow over alternate bars. The linear stability analysis provides a physical and mathematical explanation for occurrence conditions and evolutional process of alternate bar, but equilibrium bar height is not able to be determined because the bar increases in height with time. The present work made possible to estimate equilibrium height together with stabilization mechanism of alternate bars by considering non-linear relationship between bed profile and sediment discharge, which was divided into two non-linear relations of stream velocity versus bed profile and of sediment discharge versus stream velocity. Theoretical bar height coinsides fairly well with observed equilibrium bar height.

View full abstract

The main purpose of this paper is to describe the relation between sediment discharge, resistance to flow and bed configulation in channels with nearly uniform bed materials by appling the procedures of dimensional analysis to a vast number of flume and river channel data, and to present a estimation method of dediment discharge and resistance to flow from a set of water depth, bed material size and slope.

View full abstract

This paper treats the resistance to shallow water flows over a channel bed roughened with spheres. The drag force for spherical roughness element and the frictional resistance for the bed between roughness elements are separately measured by using a shear meter. The magnitude of the drag coefficient decreses with increasing roughness element concentration due to the shade effect which becomes more obvious as the relative water depth to roughness height increases. For the flows whose relative water depth is smaller than unity, the magnitude of the coefficient of frictional resistance increases with increasing water depth and roughness element concentration. Thus, the characteristics of resistance to shallow water flows which differ from the resistance law already established for deep water flows are revealed.

View full abstract

Coherent structure of large scale eddies and turbulence within the upper surface shear layer of a wind-driven current has been investigated by utilizing several flow visualization techniques. In the region of transition from laminar to turbulent flow, the visualization demonstrates the existence of a streak structure which can be shown in a plan view at the level close to the water surface and reveals that longitudinal vortices existing near the surface are responsible for the appearance of the streak structure. As the stream goes down into the turbulent region, the flow field related to these vortices develops into that of three-dimensional ‘hairpin’ vortices. A tentative model is presented to describe the evolution of these vortical flow field from a viewpoint of local instability due to the longitudinal vortices themselves. The flow structure of turbulent boundary layer near the surface has been shown to resemble that of usual wall turbulence.

View full abstract

In actual open channel flow, secondary flow is always generated near the side wall even in straight channel and this seems to influence the flow structure. In this study. secondary flow and turbulent structure in a rectangular open channel flow were investigated experimentally. As the results. secondary flow consists of the free-surface vortex which includes the horizontal flow from the side wall to the center of channel and the bed-wall vortex which pairs with the free surface vortex. These vortices interact with the primary flow and dominate three dimensional flow structure. The differences between open and closed channel flows are remarkable and therefore the effect of free surface plays an important role on the flow structure in open channel flow.

View full abstract

The relation between Lagrangian time scale and Eulerian turbulent characteristics is considered theoretically using the simple model of space-time correlation coefficient of turbulent velocity. Introducing the non-dimensional variables, the non-dimensional parameter α composed of Eulerian turbulent characteristics is derived. Then we discuss the relation between Lagrangian time scale and the non-dimensional parameter α. Numerical simulations of turbulent diffusion are made in order to prove the theoretical results.

View full abstract

In this paper, weighted finite difference equations for one-dimensional convective diffusion equation are proposed. In these equations, a value of the desired point is represented as the sum of weighted values of the vicinity points. And these weights are obtained from the individual degree polynomial that satisfied the governing equation. This method has the characteristic that we can select at will the vicinity points which have influenced on the desired point according to the state of flow. As the change of value between lattice points is represented by high degree polynomials, this method can obtain higher order accuracy than other finite difference methods and finite element method. By those equations, composed of five vicinity points, it is indicated that the numerical analysis of the governing equation can achieve higher order accuracy than h⁴, by using non-dimensional lattice spaces Δx=h, Δt=λh².

View full abstract

In pumped storage reservoirs, a well mixed layer is often produced between the upper and lower pycnoclines by discharges from and into the reservoirs. A turbid middle layer may develop when the flood flows into the thermally stratified reservoirs. To understand the flow fields of such systems, the withdrawal from the three layered stratified systems is investigated. A theoretical analysis is performed to estimate the incipient withdrawal and the discharge in each layer. The validity of the analysis is confirmed by laboratory experiments. By using the theory, the optimum height of the outlet for the drainage of the turbid middle layer fluid is evaluated, in practical sense, which may enable the effective operation of the selective withdrawal arrangement to reduce the returdation of turbidity.

View full abstract

The flow passing through constriction in an open channel is highly three-dimensional. The center flow close to the bed, however, can be analyzed by combining the two-dimensional potential flows and by using the boundary layer analysis already developed by the authors. Although this method is applied successfully in order to calculate bed shear stress along the center line of flow, the calculation is rather tedious, and it is neccessary to know the water stage at least along the center line. Furthermore, it is not possible to analyze the phenomena of flow in the whole domain of the flow. For the practical purposes, the characteristics of flow are investigated by the aid of laboratory experiments and the appreciable results can be obtained based on the hydrodynamic consideration.

View full abstract

Self-excited oscillations of flapgate are obseved in many intake structures such as weirs and dams. Recently this phenomenon occures in rubber dam which is made by fiexible rubber sheets. This oscillation has been analised partly about ten years ago by author. But this analysis can not make clear all of the occurence of self-excited oscillation in flapgate and nappe oscillations.
In this time, another analysis has been done on this phenomenon by adding to the former amalysis and the good results can be derived which are able to explain the almost of the unstalble phenomena on the flapgate and nappe oscillations. Namely, my experimental results from 1972 to 1974. H. Ivan Schwartz's “K+1/4” law and E. Pariset's “K+3/4+β” law are certified crearly by this analysed results.

View full abstract

The purpose of this paper is, (1) to investigate the rule that divides the total rainfall into rainfall components of subsystem (groundwater, interflow and surface flow), (2) to discuss on the partial source area concept (which asserts that runoff water is produced mainly from certain portions of the watershed where the soil becomes saturated by through flow from upslope by the infiltrating water) and (3) to inquire into relationships among partial source area ratio and basin storage from time series of inversely estimated effective rainfall by the filter separation autoregressive (AR) method (the hydrologic inverse detection method proposed originally for daily and hourly runoff data analysis).

View full abstract

The characteristics of concurrence of rainfall and storm surge at Osaka are made clear by analyzing the available data of typhoons for the period 1900-1980; Distributions of typhoon positions when maximum storm surge η_max or peak rainfall intensity γ_p occurred, a histogram of time-lag between η_max and γ_p, and relationships between η_max and γ_p, and between η_max and total rainfall R_T, are shown. Predictors of rainfall and storm surge are selected by using the stepwise regression analysis, and the probability distributions of the residuals are revealed. It is probable that the flood wave following the rainfall moves downstream in a river and the storm surge invades the river, with a considerably short time-lag. On these occasions, the high water levels along the river change, depending on the length of the time-lag. Such features of high water levels are examined in terms of the behavior of storm surge propagation in the lower reach with mild slope.

View full abstract

The object basins of this study are 92 mountainous basins in Japan. At first, the effect of season on fractional-type recession constant was made clear. The basin geology was classified into ten categories. Then the standard values of recession constants for each geology and season were given using the mean values of basins covered with uniform basin geology. The authors presented a method to calculate recession constants of basins covered with various basin geology. The presented equation is as follows:
b_i=¹⁰∑_j=1B_ij·a_j
Where, b_i: calculated recession constants of i-season, B_ij: standard recession constants of j-geology and i-season, a_j: areal percentage of j-geology in the basin. Finally, suitability of the recession constants calculated by the above equation was shown.

View full abstract

In this paper a model for estimation of basin-wide snow water equivalent using snow covered area is presented. At first the relation between percent snow cover and snow water equivalent is made clear using topographical characteristics of some Upper-Tone basins and the snow distribution obtained by surveying the Takaragawa basin. The dependence of snow water equivalent to altitudinal distribution of a basin is determined using snow covered area and hydrological data. The results of the application of the model to five basins are in good agreement with observed data.

View full abstract

A simplified simulation model for predicting two-dimentional beach transformation was developed based on the recent results of research works on wave transformation, near bottom velocity and sediment transport. The wave field was calculated by the energy flux method based on either cnoidal or linear wave theory in the offshore zone, and by Mizuguchi's energy dissipation model in the surf zone. The velocity field was calculated by the approximate method of Koyama and Iwata using the cnoidal wave profile or by linear wave theory. The sediment transport formula of Shibayama and Horikawa, which includes the effect of suspended sediment caused by vortices created in the vicinity of ripples, was used to calculate transport rates. The model was found to give reasonable results. The results were compared with various laboratory data and the feasibility of the model was discussed.

View full abstract

The purpose of this paper is to present a scale-model relationship for similarity between full-scale and scale models of two-dimensional equilibrium beach profiles. A similarity comparison between full-scale and small-scale model is made by considering the degree of experimental error. The similitude obtained is graphically shown, and the scale-model relationship is also formulated. It is found that the small-scale relationship agrees with the ones derived from shoreline change, critical water depth of sediment movement, and beach profile.

View full abstract

This paper presents numerical models for wave transformation due to current-depth refraction in cases of both regular and irregular waves. The models are to solve the equations for wave number components and the conservation equation of wave action or wave action spectral density by the combination of the method of characteristics and an interpolation formula. Numerical solution agrees well with the theory in cases where a one-dimensional current exists on a uniform depth or a uniformly sloping bottom. It is also found that the wave height of irregular waves is not amplified by opposing currents as well as following currents and that the model renders a slight overestimation of wave height distribution in the case of opposing currents because of a directional leakage of energy.

View full abstract

Run-up variations due to random waves have been measured on beaches having 1/5 to 1/30 slopes and the run-up spectra have been examined. The main characteristics of run-up spectra obtained from experiments are as follows: (1) The energy saturation is seen in a high frequency region. The run-up spectra in the saturation region show a f^-4 and tan⁴ θ dependence (f: frequency, tan θ: beach slope) and the energy densities are independent of incident wave energy. (2) Only in a low frequency region, spectral energy densities increase with increase in the incident wave energy.
These chracteristics of run-up spectra are discussed from a new standpoint that a run-up wave profile consists of parabolas.

View full abstract

On-offshore and vertical water particle velocities in surf zones on beaches in wave tanks are measured with a hot-film anemometer and a two-component laser-doppler velocimeter. Turbulence generated by wave breaking is estimated. Vertical distribution of turbulence intensity inside the surf zone is discussed in comparison with the vertical distribution of eddy diffusion coefficient in surf zone. Time variation of turbulence intensity during one wave period can be explained by extending the turbulent wake theory. Time variation of Reynolds stress during one wave period shows a similar trend as that in oscillatory pipe flow. Reynolds stress, which is defined here, may not play an important role in the on-offshore momentum transfer in a time scale shorter than the wave period.

View full abstract

A simulation model of oil spread due to tsunamis is developed. The equations for the motion of oil are the Navier-Stokes equation and the continuity equation. Since no effects of surface tension is included, the method is only applicable for a short time, in which the inertia and viscous effects dominate. This method is applied to Miyako Bay, on the North Sanriku Coast in Japan. It is assumed that 2000m³ oil spill from tanks due to earthquake. Input tsunami is the same magnitude as Tokachi-oki Tsunami hit in 1968.

View full abstract

Runoff rate of pollutant load (COD, N, P) was studied, as a planning parameter for water quality control of lakes or of inner sea areas. With three primary or secondary rivers, Ibo R., Kako R. and Kotoh R., daily sampling surveys more than one year, were practiced. Also, the discharge load in the river basins was calculated in detail. It was confirmed that the runoff rate of pollutant load had better be divided into two stages in convenience, i. e., flow-out and flow-down stage. Good accordance was observed between measured and calculated values of runoff load of pollutant, by setting the former value of flow-out rate 0.7-1.0, the latter value of flow-down rate as an expornential formula of flow-down distance. The application of this method to another several rivers was also satisfiable.

View full abstract

Based on the experimental research and in-situ survey in eutrophic sea areas, the numerical modelling of nutrients behavior between water and sediments was investigated.
The fundamental elements in the model are phosphorus, nitrogen, oxygen and iron. Sulfide ion is regarded as debt of DO, and ferric oxide as store of DO respectively. Ferric oxide constituting oxidized zone is formed by the reaction of ferrous sulfide in the upper layer of sediments, with DO in the lower layer of water, and is reduced by sulfide ion in the upper layer of sediments. Then, a part of the adsorbed phosphates are equivalently released. Using this model, the characteristic patterns of seasonal variation of phosphate discharge from sediments, DO concentration in water and so on, were satisfactorily simulated.

View full abstract

In this paper a new graphical method of judging whether or not a particular distribution adequately describes a set of observation is proposed. This method needs only an arithmetic-scale paper and a visual judgment of goodness of fit. The test makes a comparison between the plots of the actual data and the straight line having a gradient of one through the origin on the arithmetic-scale paper. To show the effectiveness of this method, it is applied to random numbers and actual precipitation data. Results show that it is easier to judge visually the goodness of fit and to apply to any continuous or discrete distribution than with the usual probability-paper method. The properties behind this visual test are discussed in detail.

View full abstract

The appropriate description of fluid flow pattern in fluidized bed biofilm reactor is necessary for the prediction of its system performance. The experimental correlation equation of the longitudinal dispersion coefficient was estimated by the tracer impulse response method using the various particles which are similar in physical properties to the bioparticles, that is, biofilms coated onto growth supporting carriers. The presented correlation equation between longitudinal dispersion coefficient, void fraction and particle diameter was obtained by introducing well-known Taylor's and Kozeny's models. It is proved that the presented correlation is more valid within the range of particle Reynolds number used in this experiment, compared with the prediction by Krishnaswamy's and Chung's experimental correlations.

View full abstract