In this investigation, the evaluations of progressive failure and scale effect of trap-door problems were presented, by comparing the experimental results with elasto-plastic finite element analyses with a strain-softening constitutive model including a shear band. The evaluations were performed in passive and active mode. The elasto-plastic finite element analyses had good agreement with the distribution of stress on a trap-door and shear band development observed by experiment. The scale effect of trap-door problems was also recognized by the analyses on passive mode but not recognized in active mode in the same manner as the experiment.

View full abstract

To evaluate the safety factor of a slope with respect to sliding, as expressed by the ratio of integrated shearing resistance determined approximately by Coulomb's failure equation to shearing stress along the sliding surface, the nodal displacements were analyzed by the finite element method using plane 8-noded isoparametric elements assuming an elastic model slope.
Then, normal and shearing stresses at given points on the sliding surface of the model slope were directly computed by the element stress matrix and nodal displacements for specific elements.
At the time of the stress calculation, an approximate dichotomizing search procedure for a finite element including the sampling point was used, together with the Newton-Raphson method allowing transformation from a set of global Cartesian coordinates for the model slope to a set of local coordinates for the isoparametric elements.
The safety factors of the model slope computed by the proposed finite element stress method were compared with those computed by the simplified Bishop's method and the ordinary slice method. The safety factor, diameter and position of the critical circle obtained by the finite element stress method were nearly equal to those obtained by the simplified Bishop's method and the ordinary slice method.

View full abstract

This paper developed a theory on the heterogeneity of gravity model stressing the point that the difference between the number of populational movements going out and coming into the regions depends on the size of potentiality of supply and demand of population.
It could be shown there that not only making clear the meaning of K and b as an indexes of regional characteristic but also the structure of populational movement by region, could be had using the revised K and b.
In particular the revised b provided a new explanation concerning the issues of “Map pattern” regarding the gravity model.
Moreover, considering the fitness of the gravity model, the power type model is more effective for estimating the populational movement.
Therefore, one should pay attention in choosing a suitable model, when using each model.

View full abstract

Soil transportation by the tractive force of surface water flow has been investigated in experiments using uniform surface water flow as a model, but very few papers have dealt with the mechanism of soil transportation by non-uniform and unsteady surface water flow appearing in sloped soil layer during natural rainfall. In this study, we analyzed the mechanism of soil transportation involving a non-uniform surface water flow produced by a computer controlled surface water flow generator. Experiments were performed under the same hydraulic conditions as had been set in previous experiments for the study of uniform surface water flow.
The comparison between the results of the non-uniform flow and the uniform flow reveals the following. 1) The quantity of soil transportation increased with an increase in the volume of surface water flow in both cases. 2) When the soil layer was exposed to the same volume of surface water flow, the non-uniform flow case showed smaller values than the uniform flow for the rate of increase in the quantity.of soil transportation. 3) The relation between the dimensionless quantity of surface water flow shown by q_w/q_wc (q_w: surface water flow, q_wc: critical surface water flow) and the dimensionless quantity of soil erosion q_e/u_*d_k (q_e: quantity of soil erosion, u_*: friction velocity, d_k: average clod size) had been reported in the case of the uniform flow. Therefore, we applied these relations to the results of the measurements in this experiment and obtained a new equation with a nonuniform surface flow. That equation would be expected to be applied to the estimation of the quantity of soil transportation occurring in a field subjected of a natural rainfall.

View full abstract

Stability for seepage failure of the ground subjected to an upward seepage flow is able to be judged using S_res. as the parameter; where S_res. is an area of distribution of the residual vertical effective normal stress σ_z'_res. in a critical state. The formulization of optimum design of a gradually graded filter is performed using S_res. as the parameter of stability. The formula is so difficult to be solved analitically that the following three types of distributions of coefficient of permeability k are investigated precisely here:
(1) Two-layer varying type,
(2) locally linearly varying type, and
(3) locally bi-hyperbolically varying type,
which are considered to be able to represent almost all distributions of k within the ground. The ground is approximated to a 1000-layered soil column systematically and S_res. of the ground is analyzed using the computer program: Critical State Calculation Program for Seepage Failure of Multi-Layered Ground (CRIT 1).
In this paper, the special case, k_r=5.0, is investigated precisely, where k_r is the ratio of the coefficients of permeability at the top and bottom of the ground. The following remarkable results are then obtained:
(1) The distributions of k:(1), (2) and (3) which give S_res. a maximum value S_{res. max} coincide.
(2) The distribution of the coefficient of permeability of the most stable filter is the two-layered system: l₁=0.309 L. Where L is the length of the filter and l₁ the length of the lower layer of the two-layered system.

View full abstract

A mathematical model including conditions of irrigation flow rate, irrigation water quality and fertilizing as operating variable was built for estimating the phosphorus outflow from paddy fields. This model consists of balance equations and rate equations such as adsorption, desorption, bioturbation and other relevant rate processes.
Six experiments using lysimeters under various conditions of paddy fields were carried out to estimate structure and verify validity of this model.
The model simulated the phosphorus concentration in surface water under various conditions with enough accuracy, reflecting the annual changes and the effect of fertilizing. High stability of the model was checked by a sensitivity analysis. Phenomena inside of experimental paddy fields were estimated to be described well. This model will be applied to actual paddy fields after verification tests have been carried out in large-scale fields.

View full abstract

The cross section of fill dams is generally triangular or trapezoidal in shape. If the analytical system contains such geometry, the finite difference or finite element procedure is the most effective for analyzing such problems for handling the boundary condition and obtaining accurate numerical solutions. However, it is inevitable that the computational efforts need to be increased for obtaining more precise solutions using these procedures. From this reason, a more simplified analytical method is proposed in this paper. It is characteristic to sum the finite difference eqation of a given equation under boundary conditions. Basically, this approach is similar to the procedure from which the closed-form analytical solution is derived, whereas, solutions obtained are reduced to a discrete function. Applying this method to a two-dimensional consolidation equation which is Terzaghi-Rendulic's simplified equation, the pore pressure distribution in the impervious zone of fill dams can be computed.
To ascertain the validity of this method, the forward finite difference method is employed here. It is indicated that the computational results by this procedure agree with the one by the conventional finite difference method with a reasonable degree of accuracy. In the case of two types of cross sections with symmetry and asymmetry, the computational results of the pore pressure distribution. are presented for the isotropic condition of ratio of horizontal to vertical coefficient of consolidation, c_h/ c_v=1 and the anisotropiC condition of c_h/c_v=30. Based on these results, the characteristics of geometrical cross sections and differences between isotropic and anisotropic conditions are estimated.

View full abstract

Water qualities were monitored regularly in the investigated paddy field area during two irrigation periods and one non-irrigation period. The inflow and outflow loadings were calculated.
1) The amounts of inflow water during the irrigation periods in 1987 and 1988 were 4, 170 and 5, 150 mm respectively as a result of the flowing irrigation and a large percolation quantity.
2) The peak concentrations of surface outflow and percolation during the non-irrigation period were much higher than those during the irrigation periods, except for the phosphorus concentrations of percolation.
3) The T-N and T-P loadings of the surface outflow calculated by the separation method, in which the loadings in dry weather days and rainy days were calculated separately, were less than those by the non-separation method during the irrigation periods.
4) The net outflow loadings of T-N, T-P and T-COD were 45.7, 8.72 and 98kg/ha·year, respectively. Among them, the values of the non-irrigation periodwere T-N=23.6, T-P=1.24 and T-COD=35kg/ha.

View full abstract

The nitrogen and phosphorus contents in the paddy soil were measured six times during the irrigation period in 1988. Specifically, mineralized nitrogen (the amount of ammonium nitrogen after 28 days flooded incubation at 30° of air-dried soil) and available phosphate (the amount of phosphorus extracted with the Bray No.2 solution) were measured in the three paddy fields in the area investigated. The relationships were discussed between the nitrogen and phosphorus contents in the paddy soil and the pollution loading outflows from the area presented in the authors' previous paper.
1) The value of mineralized nitrogen in the first layer reached the maximum on Jul. 20, and the second layer on May 25. It was calculated that the amount of nitrogen supplied from paddy soil and fertilizer.was approximately equal to the uptake by the rice plants during the irrigation period, . and the amounts of nitrogen fixation and denitrification were also approximately equal. The amount of net outflow loading was probably caused by these relationships.
2) The values of available phosphate were high level compared with previous studies. It was thought to be that the excessive accumulated phosphate in the soil caused a large amount of net outflow loading.

View full abstract

The previous stress-strain theory for various types of soil has been developed employing the incremental theory. The strain increments are described by means of the mean principal stress increment and deviatoric stress increments with the deformation moduli which are the simple functions of stress. The theory is applicable to the soil behavior of general three-dimensional stress conditions in work-hardening process up to failure during loading with no rotation of principal stress axes under orthogonal anisotropy. The values of the parameters involved in the moduli can be determined entirely from the results of conventional triaxal compression tests. The true triaxal tests for loose Toyoura Sand were performed using various stress paths with a series of constant values of the mean principal stress. These experimental results of the stress-strain behaviors were predicted with good accuracy using a set of unique values of the parameters in the increment equations.

View full abstract

The strain increments of the incremental equation presented are composed of one part due to the mean principal stress increment and another due to the deviator stress increment. Therefore, the incremental equation must be examined to see whether it can estimate the strain behaviors. depending on the stress path that the mean principal stress and the deviator stress change coincidently or not. For the stress path with a constant lateral pressure, the measured strains are well explained by those calculated. For the stress path with constant principal stress ratios, the measured volumetric strains changing linearly to axial strain are also estimated by those calculated. The measured strains induced by changing the stresses from one state of stress to another depend not only on the end point but also on the stress path. This experimental fact is well represented by the calculation.

View full abstract

Barrages constructed over these last ten years were built using the New Design Standardsauthorized by the Ministry of Construction. Some of them have caused by deposition of sand andgravel on the sill of barrages during big floods.
The study of typical cases by analysing (1) surveying maps and/ or (2) aerial photograghs revealed such conclusions as follows:
(1) The New Design Standards enforced engineers to design and construct barrages with their sill elevation equal to that of the “planned” river bed, which is lower than natural one.
(2) In most observed cases, the dredged and mined reach of river bed around a barrage is not so sufficient that sand and gravel accumulate on the sill of the barrage during big flood.
(3) If deposition on a sill of a barrage consists mainly of sand, deposition under gates can be flushed by the operation of sluice gates during the falling stage of a flood.
(4) If deposition on a sill of a barrage consists of gravel, gravel is too heavy to be flushed. So some precaution against deposition is needed. The most likely technical coutermeasure is to dredge a sufficient reach length around the barrage. The most practically proposed dredging reach is about six times as long as the width of the river channel upstream of the barrage and two times the downstream.

View full abstract

The behavior of buried flexible pipe is influenced to a considerable extent by the properties of the backfill material and pipe stiffness.
Especialy, in case of twin parallel buried flexible pipes, the interaction between the backfill material and pipes is more complicated because of reaction of each pipe. In this paper, the authors discuss the behavior of twin parallel buried flexible pipes that was obtained by experiments using the steel test pit and nonlinear elastic finite element analyses.
The pipes for experiments are fiberglass reinforced plastic pipe (D=φ200mm, t=3.4 mm, EI=89.87N/m²·m⁴/m, EI/D₂³=10, 900N/m²). Its ring stiffness factor (EI/D₂³) is comparable with that of commonly used FRPM pipe.
The backfill material is dry Seto siliceous sand.
The results obtained are summarized as follows:
1) The behavior of buried flexible pipe can be estimated by nonlinear elastic finite element analysis, if the backfill material is dry sand.
2) The deflections of twin buried flexible pipes for 0.5R separated between pipes become 77% in vertical direction and 72% in horizontal direction compared with single buried condition.
3) In case of twin buried flexible pipes, the circumferential strains of pipes concentrate on the points that inclind 30° upper and lower from the spring-line of the adjacent side. Even so the maximum strain under this condition is smaller than single buried condition.
4) The vertical earth pressure on the pipe is depressed distribution. And it is not influenced by buried conditions.
On the other hand, the horizontal earth pressure acts on the broad area of the adjacent side at the twin buried flexible pipes.

View full abstract

In this paper, the method of predicting soil flux and water flux in the vertical direction after puddling is proposed. The model consists of Darcy's equation including the relative. velocity between water and soil as well as a continuity equation for water and soil. The settlement of soil surface, flow rate from the soil column and hydraulic head differences calculated by the method were compared with the experimental data using three types of soil. The results are summarized as follows:
1. The results, especially in clayey soil, obtained by this calculating method agreed approximately with the experimental results.
2. Variation of soil flux and water flux with time obtained by the method were dependent on soil texture.
1) In clayey soil, soil and water moved always downward after the mixing.
2) In medium-textured soil, they moved downward just after the mixing.
However, water at the upper part of the mixed layer reversed upward after the flow rate from the soil column decreased suddenly by filling the macropores with fine soil particles. Moreover, water reversed downward again as soil flux decreased.
3) In coarse-textured soil, water moved upward just after the mixing. Further more, it reversed downwared as soil flux decreased.
3. Area of the hindered settling just after the mixing was confirmed by the soil flux obtained by the calculating method. It was considered that soil flux in the hindered settling was decreased as the flow rate from the soil column was decreased.

View full abstract