Abstract Generally, from the viewpoint of soil physics, the cultivated land where soil is abundant in aggregates, especially water-proof ones, has high agricultural productivity. On the other hand, the formation and maintenance of soil aggregates is one of the effective method of the land conservation. In this paper, at first, we propose a method that the degree of aggregation is represented by the variation between the gradation analysis curves before and after dispersion, and secondly investigate especially the factors contributing to aggregate formation and the following conclusions are obtained.
(1). In the natural aggregates, the range of diameter where the aggregates are of a maximum amount is between 0.01 and 0.025 mm.
(1). In the natural aggregates, the range of diameter where the aggregates are of a maximum amount is between 0.01 and 0.025 mm.
(2). Though cation (Ca⁺⁺) is effective to the formation of relatively micro-aggregates, organic matter is more effective to the macro-ones.
(3). It is desirable that the expression of soil structure by the three phase is converted to four phases by dividing the solid phase into the phases of soil particle and organic matter.
(4). It is suggested that aggregate formation is dependent upon an interaction between cation and organic matter. Furthermore, when the aggregates are created artificially, the ratio of 15% Ca to 5% organic matter seems the most profitable addition for a virgin soil.

View full abstract

Quantitative analysis on the unsaturated flow in layered soils with so-called closed system where the air in porous media is not connected with the atmosphere was tried. In this case, the phenomenon of unsaturation is caused by that the dissolved air is released as air bubble in the pores of percolating layer. Some presumptions are set for the execution of analysis. As to hydraulic conductivity, saturation, pressure head, temperature, air volume and others, there are relations shown in Fig. 1, and these individual relations were discribed in detail. In case the air volume in the upper layer is obtained, the volume of released air is considered separately with the part caused by lowering of solubility and that by oversaturation, and the agitation coefficientαandβshowing the readiness of air release to above each part was introduced. Similarly, γfor the agitation coefficient to the under layer was used.
For example, the theoretical caluculation on the isothermal process was performed. As the result, it was recognized that the calculation corresponds nearly to the distribution of pressure and the change of percolation flux, therefore the propriety of the theory to explain the phenomenon was ascertained.

View full abstract

Laboratory experiments were carried out to examine the validity of Darcy's law for unsaturated flow in soils. Hydraulic conductivity was measured by the apparatus of Richards and Moore (1952) under the suction pressure below 200cmH₂O, and hydraulic conductivity of the order from 10^-6 to 10^-8cm/sec was applied in the experiments. Hydraulic gradient was increased from a small value to a large value and the maximum value was 10-20 cmH₂O/cm. Three kinds of soils sieved through 0.84mm (SiCL, S, LiC) and two kinds of undisturbed soils (CL and HC) were used. Within the range considered in this study, the relation between the flux (cm³/sec/cm²) and hydraulic gradient was linear for all samples except one. Deviation from linearity for one sample was thought to be due to the displacement of particles under high hydraulic gradient.
These experimental results showed that the Darcy's law is valid and the effect of interaction between pore water and soil particles was negligible.
Using a moisture characteristic curve, hydraulic gradient was converted to moisture gradient. As the result of this, the moisture gradient was smaller than 1%/cm at maximum hydraulic gradient and under any suction for most samples. The above values were much less than the value of Swartzendruber (1963) as pointed out about the non-Darcynian behavior during infiltration.

View full abstract

Temperature distribution in soil is generally affected by moisture movement and moisture distribution. These phenomena can be expressed by a general differential equation (2) proposed by de Vries (1958) and others. However, it seems that in many cases the analysis of experimentally measured temperature distribution have not been sufficiently analyzed.
This paper is aimed to clarify the effects of moisture movement and moisture distribution with respect to the temperature distribution in a cylindrical sand column under a one dimensional temperature gradient. An estimate of the effect of heat escape through the heat insulating material in relation to the temperature in the sand is also made.
Fig. 1 shows the experimental apparatus. The sample is sand, the heat conductivity of which is shown in Fig. 2 as a function of water content (g/g soil). Fig. 3 shows the temperature and moisture distribution measured experimentaly and Fig. 4-6 show the calculated temperature distributions.
It is apparent by solving equation (2) with some assumptions that ; The temperature distribution is determined by the effect of the phase transition of water in the sand, by the variance of heat conductivity of the sand by water content, and by heat escape through the column wall.(ii) If the water content is equal to a steady state under the temperature gradient, water vapour transports the sensibleand latent heat. This phenomenon, however, has little affect on the temperature distribution.(iii) If the water content is not equal to the steady state under the temperature gradient, evaporation or condensation occures in the sand which influences the temperature distribution.(iv) Thh variance of heat conductivity of the sand by the water content markedly affects the temperature distribution.(v) The lower the water content of the sand, the greater the influence of heat escape through the heat insulating material on the temperature distribution.(vi) When the water content is relatively high, the escape of heat and the latent heat of evaporation in the sand cancel each other out.

View full abstract

This paper presents the results of a resonant column test and an ultrasonic wave propagation test on statically compacted specimens of cohesive soil, describes some of the more meaningful phenomena observed, and discusses the effects of water content on the dynamic elastic constants of the soil.
In this dynamic experiment, the Young's modulus, shear modulus and Poisson's ratio of compacted specimen of cohesive soil were investigated for a range of vibration strain of 10^-6-10^-4.
An attempt was made to relate the dynamic elastic constants of cohesive soil to the water conditions around the soil particles by introducing an equivalent thickness of water film: D (water content/specific surface). Based on this investigation, we obtained information about the influence of water conditions of the specimen on the dynamic elastic constants of cohesive soil. The results are as follows:
(1) The dynamic elastic constants of cohesive soil change by increasing and decreasing the degree of saturation while holding the dry density constant.
In most cases, the dynamic elastic constant-water content curves show that as the water content is increased, the constants increase to a peak and then decrease.
(2) The values of water content which correspond to the peak values of the dynamic elastic constant differ in each sample soil. However, the values of D are approximately equal. The water films ranged from 30Åto 50Åin thickness.
It appears that the thickness of water film represents the boundary between adsorbed water and free water on the soil surface. The inflection point on a dynamic elastic constant-water content curve represents characteristic water condition of the soil.
(3) The observed velocities of longitudinal wave propagation using an ultrasonic pulsating method are different from the results of resonant column method with respect to the degree of specimen saturation. However, the differences in the test results of the dynamic elastic constant can be considerably reduced when the relation between propagation wave length, radius of specimen and Poisson's ratio ocompacted cohesive soil is considered.
(4) The dynamic shear modulus of compacted cohesive soil can be derived from the propagation velocity of the transverse wave in the specimen. The values are not noticeably affected by the water conditions of the soil.
However, the dynamic shear modulus shows a tendency similar to that of the dynamic Young's modulus with respect to the water conditions in the soil, and has a peak value which is related to the change of water content in the specimen.

View full abstract

This paper presents the results of an experimental study of the shear strength of KI-SUSHI CLAY under the influence of loading velocity. Experiments on the unconfined compressive strength of KIBUSHI CLAY, in which kaolinite is the main ingredient, were carried out by a stress control apparatus.
Water content, liquid limit, plastic limit and plastic index of the specimen were 26. 0%, 51. 0%, 15. 7%, 35. 3% respectively, The relationships between shear strength, the upper yield value in the rheological model, Henkey's elastic energy and the loading velocity are discussed in this paper. Analysis of the results obtained from the tests indicates that these three factors remained constant when loading velocity wasgreater than (α) =6. 0×10^-4 kg/cm²/min, and that these three factors increased exponentiallyas loading velocity' decresed below this valve calculated. Stress-strain-time corves were classified into 2 types as described below.
(1) Stress-strain-tim e curves of samples under a loading velocity of (α) =2. 5×10^-5 kg/cm²/min, behaved elastically. The upper yield value was 92% of the shear strength. When the load applied to the sample exceeded the upper yield value the sample instantly collapsed.
(2) Stress-strain-time curve were similer to viscoelastic curves when the sample was under a loading velocity of (α) =2.4×10^-3 kg/cm²/min. In this case strain in the samples was accumulated during loading, and the yield strength of the sample decreased to 70% of the shear strength. The yield strain was approximately twice that in (1).
Therefore, the following conclusions can be niade.
(1) Shear strength is affected by the loading velocity.(2) The upper yield value varies from 70% to 92% of the shear strength depending on the loading velocity.

View full abstract

In the present paper, the three-dimensional characteristic of seepage through an earth dam is analyzed by the finite element method.
We adopt various dam types and investigate the discharge and location of the exit point.
A summary of the results obtained is shown below:
(1) The discharge is approximately proportional to the ratio of permeability coefficients in the horizontal' and vertical directions irrespective of the dam type.
(2) When the upstream-and downstream-slopes and the crest width of dam are held constant, the abutment-slope doesn't influence very much on the ratio of discharge to area of the upstream face.
(3) We propose a simple method to estimate the three-dimensional discharge of the earth dam by using the Casagrande's method.
(4) The location of the exit point becomes higher as approaching the abutment and this characteristic is remarkable in a dam with isotropic permeability coefficient.
(5) By comparison of the Casagrande's method with the numerical results, the ratio of the location of exit point by the numerical results to that by the Casagrande's method is found to be from 1. 6 to 2. 0, in case of the isotropic permeability coefficient, and from 1. 3 to 1. 5 in case of the anisotropic permeability coefficient.
Furthermore, we analyze an earth dam having its crest arches toward upstream and compare the results with the above-mentioned results.
As a result, with respect to the location of exit point, there is no difference between the two cases. And regarding the concentration of the equipotential surfaces toward the downstream, the dam type with arching crest is slightly distingushed and the discharge of the dam type with arching crest is 1.05 times as much as that of the dam type with straight crest.

View full abstract

The channel networks system in a low-lying drainage basin composed mainly of paddy fields is so complicated that the direct pursuing of flood flow by a computer applying the mathematical model of unsteady flow is hardly possible. A study is then made of possibility of lumping the channel networks system with pumping activities from the viewpoint of engineering practice. The examination is performed stepwise such as a paddy field lots-ditch system, a ditches-branch channel system and a branch channels-main channel system, under a given condition of discharge at the outlet of each system. As the result, it has been disclosed that the channel networks system can be simplified to the secondorder channel systems.

View full abstract