In this paper, we research the physical stability of aggregates and some critical points of physical properties of soil particles. As the results of this experimental research, the following conclusions are obtained.
(1) From the viewpoint of physical stability, aggregates of soil particles are classified by theirparent material. For instance, though the andesitic soil has relatively more stable aggregates, the granitic soil has less stable ones.
(2) By comparing the effect of some physical methods on aggregate dispersion, they prove to be less effective in the order of kneading, boiling, supersonic irradiation, trituration and stirring.
(3) We name some critical points of physical property “Soil Particle Constants”. So, a physical basis can be given for setting up various size limits of soil particles.
(4) According to the above-mentioned Soil Particle Constants and its convenient setting, the classification of International Society of Soil Science seems most profitable for practical use.

View full abstract

This study has been done to see the effects of specific surface of soils and thickness of water film on the stress-strain behavior and strength-characteristics of compacted fine-grained soils.
From the experimental results the following matters were clarified.
1. The stress-strain behavior of soils subjected to unconfined compression varies largely by the difference of the thickness of water film, that is, the larger the thickness of water film, the more remarkable the visco-elastic property becames, regardless of the magnitude of the specific surface of soils.
2. In the same thickness of water film, the smaller specific surface gives the more remarkable elastic property.
3. The unconfined compressive strength of soils compacted at the optimum moisture content is larger in the specimen of smaller thickness of water film (that is, of larger specific surface).
4. Between the compacted load and void ratio, there is a relation of exponential function.
5. In the fine-grained soils, there is a tendency that as the water content is increased an “over-compaction” phenomenon arises.

View full abstract

The time serial changes of soil behavior and the several hydrological properties in consequence of grassland development have been continuously investigated in Shikoku Karsu mountainous district.
In this paper, the characteristics of soil before the development of grassland, namely, the initial and native conditions of soil are made clear, including comparison with those of other similar districts.
(1) The surface soil in the investigation plot is humic volcanic ash soil called “Kuroboku” containing organic matter of 18%. The subsoil is residual soil of limestone by weathering.
(2) Both the surface and subsoil include many fine-grained particles. They were highly advanced in aggregation and especially contained a large amount of coarse aggregate of more than 0.25mm.
(3) The permeability of the soil layer shows about 10^-2cm/sec.
(4) Moisture content by weight in situ gives high values of 90-180%. These values of the upper part with gentle slope (called ridge part) in the investigation plot are higher than those of the lower part with steep slope (called sloping part).
(5) Dry density of the ridge part is 0.40-0.70 g/cm³ and smaller than that of the sloping part. There are negative correlation between dry density and moisture content in situ and it seems to be similar to that of the standard compaction curves in drying process.
(6) In three-phase distribution of natural condition, the ratio of solid, liquid and vapor phases is about 2: 7: 1. And then it is found that the liquid phase occupies a very large portion.
(7) Moisture holding characteristic of soil in energy concept shows that the moisture content held by energy of less than pF 2.3 is of a small amount.

View full abstract

Water retention and hydraulic conductivity are influenced by soil structure at relatively high water content. Cultivation and compaction make the soil structure change, therefore, water retention and hydraulic conductivity are considerably affected by them.
Authors compared the hydraulic properties for several undisturbed soils and sieved soils and discussed the influence of soil structure on water retention and hydraulic conductivity. Sieved soils were made by sieving at 2.0 or 0.84mm and compacted under the condition of field density, then, these samples were in a homogeneous state.
Experimental restlts are given as follows:
1. As undisturbed soils have large pores, hydraulic conductivity at saturation is higher than that of sieved soils, but by increasing the sucking pressure, the hydraulic conductivity of undisturbed soils decreases rapidly. Therefore, in an unsaturated state, sieved soils retain more water and higher hydraulic conductivity than the undisturbed soils in the same suction. This is reason for the homogeneity of samples. Generally speaking, the above tendency is remarkable for soils with strongly developed structure.
2. Difference of hydraulic conductivity between cultivated soil and sieved soil is large near saturation, but the unsaturated flow characteristic of cultivated soil resembles that of sieved soil rather than undisturbed aggregated soil. This means that the cultivation produces large and small pores instaneously.
3. By comparison of water retention and hydraulic condudtivity among three soils, the first is the soil passing a few days after cultivation, the second is the soil passing eight months after cultivation, and the third is sieved soil, which made clear that the bulk density of the second soil is larger than the first one, and the water retention and hydraulic conductivity of the second soil resemble the third one rather than the first one. This result means that the duration after cultivation makes soil homogeneous.
4. It is possible to approximate the unsaturated flow property of undisturbed soils by making change the particle size of sieved soil. Then, it is thought that water movement occurs between millimetersized peds at relatively high water content.

View full abstract

The authors investigated into the strength properties of peats, especially the tensile strength properties which are peculiar to peats in relating to the moisture stage.
1) The tensile strength of peats of the natural stage decreases with the progress in decomposition. Although the density increases with the decomposition, the resistance is thought to decrease due to destruction of fibres.
2) Observing the change in tensile strength of peats of the same structural materials, by drying gradually with drainage, it is thought to change with the drainage properties. At all specimens, the strength increases gradually with the drainage and increases suddenly at a certain point.
The inflection point lies at pF 3.2-3.5 which is thought to be the moisture stage of the rupture of capillary bond for the ordinary soils.
3) The physical factors influencing the tensile strength by the drainage are thought to be the moisture content and the density. However, their correlation is not obvious.
Observing the change in strength of the peats of natural stage, by increasing the density, it increases gradually. Considering the fact that the strength per unit tensile area is almost constant, the strength is thought to increase, solely with the increase in density. With the drainage, the moisture decreases, the density changes and the friction is applied to the fibres. These accumulated effects increase the strength.
4) In order to clarify the relation between the structural materials and the tensile strength, it is necessary to clarify further the quantative judgement of the floristic constitution. The peats consist mainly of the sedge show the largest increase in strength by the drainage processes.

View full abstract

Runoff phenomena in a low-lying basin composed mainly of paddy fields hardly allow direct computer simulation based on a mathematical model of unsteady flow because of the complexity of drainage channel network system.
The methodology for lumping the channel networks in the natural drainage system has been developedto follow a previous study for the system having pumping activities, from the viewpoint of enginneering practice. The possibility of the lumping is examined stepwise through a “paddy field lot-ditch” system, a “ditch-branch channel” system and a “branch channel-main channel” system under a given condition of water level at the downstream end of each system.
As a result, it was disclosed that the complicated channel networks are able to be reduced to a second-order channel system by fully taking into account of the flow conditions in channel.

View full abstract

The water movement from water-supply ditches buried in soil related to the sub-irrigation method proposed by the authors was considered experimentally.
The rising velocity (V) of wetting front from the watersupply ditch is closely related to the height (H), of wetting front as expressed by Eq.(1)
The rising velocity of wetting front is also affected by the initial soil water content (M) above the watersuppy ditch as shown by Eq.(2). By using the integrated formula of Eq.(2), we can estimate the irrigation time required until the wetting front reaches a certain height.
Furthermore, it is found that the amount of irrigation water supplied to soil from the water-supply ditch per unit time decreases with the height of the wetting front

View full abstract

In this study, the estimation of functions of the sprinkler irrigation system and the relaxation of the conditions for designing the system were investigated.
As the conditions which should be relaxed, the allowance of time difference in each sprinkler and the lowest limit of time required to sprinkle chemicals were treated.
The results obtained are as follows:
1) The mean depth of sprinkling acts as the index of the upper limit of its time and the minimum depth in the sprinkling is a primary factor to determine the lowest limit of the sprinkling time.
2) The optimum sprinkling time exsists between 4 and 7 minutes, and the time increases linearly according to the scale of the irrigation facility (See Fig.7).
3) The ratio of the time difference and the sprinkling time is practically allowable up to 30-40% for the facility having a remarkable scale effect relating to the cost.

View full abstract

The bedload theory developed by the author in PART I and Part II and the formula of the movable bed which was deduced from this theory for the straight channel of the uniform rectangular section were applied to the results of experiments which the author had long been carried out and the validity of the theory was proved.
Namely, the observed values of the sand surface height at the movable bed experiments carried out on the channel 40cm wide and 16 m long with variable bed gradient through varying the discharge, the diameter of sand particles and the bed gradient were compared with the predicted values of the sand surface height which were obtained through applying the observed water depth to the abovementioned semi-theoretical formula. The frequency analysis was carried out on the absolute values of these differences and the confidence limit was obtained through the analysis of variance using the design of experiment. The following knowledges were obtained. The predictions were made with fairly good accuracy, independent of the time interval and the distance interval. As results of the frequency analysis, in almost 50% cases the differences are within 3 mm. This is less than 1 mm, which is thought to be almost negligible, in 16% cases. In 60-70% cases this is less than 5mm which is regarded as the allowable limit. As results of the analysis of variance, the contribution coefficient of the factors such as gradient, discharge, time, distance for the variation of the differences is 1% at the largest, and less than 0.5% in most cases.
This shows that the author's theory is universally valid. Moreover as a result of the confidence limit analysis it can be stated with the probability of 95% that the absolute values of difference between predicted values and observed values using simple point gauge lie between 3mm and 5mm.
It is thought to be possible to highten the accuracy through improving the measuring method. Thus the validity of the author's theory was established.

View full abstract

In this paper, comparison of results calculated by the finite element method presented in Duncan-Chang papers for non-linear, character of stress-strain relation, and the field data in performance of two test embankments are discussed to predict the undrained behavior in compressive foundation. In addition, the author attempts to examine whether it is possible to predict the sliding failure of foundation in performance by lateral movement below the toe of embankment or not.
The conclusions are summarized as follows: Application of the initial tangent modulus E_iu obtained by the unconfined compression test is superior to E_it obtanined by the triaxial compression test from the viewpoint of approximation, which is on the safety side in design for the purpose of predicting displacement in compressive foundation.
The critical value δ_c of lateral movement below the toe of embankment is proposed to control embankment performance from the load-lateral movement relation obtained by application of E_iu in the analysis.δ_c is defined on the basis of Tayler's yield point. It seems from the two field tests that δ_c is effective to predict the sliding failure of foundation in performance approximately.

View full abstract

Concrete is chemically basic having a pH of about 13 and attacked easily by acids which have pH values less than 7. Therefore, experiments were carried out for the last three years on the influences of the corrosionof a 5% sulfuric acid solution on the periodic characteristic of variation of roughness on concrete surfaces, as well as on the changes of strength and weight, and the following factors were measured and analyzed every week over a 9 to 20 week period:
a. Volume and weight.
b. Unit weight.
c. Bending strength.
d. Compressive strength.
e. Dynamic modulus of elasticity.
f. Carbonization of concrete.
g. Variation of pH value of 5% sulfuric acid solution.
h. Periodic characteristic of roughness on concrete surfaces.
Especially, new attemps were adopted for measurements of carbonization of concrete and roughness on concrete surfaces, and good results could be obtained. In summary, the results are as follows.
1. Volume and weight decreased linearly 8 to 10 weeks later and after 12 weeks those became nearly constant.
2. For the factors c, d and e, the test pieces curing in a 5% sulfuric acid solution showed lower values than those curing in water.
3. Among the factors c, d and e, the influence on corrosion by a 5% sulfuric acid solution was minimum for the facotr e.
4. The carbonization of concrete reached a level of 15% and more 5 to 6 weeks later.
5. Periodic characteristic of roughness on concret surfaces could be analyzed properly with respect to wavelength by means of Fourier spectral analysis.

View full abstract

The effect of creep history on the shear strengh of clay was studied by the undrained triaxial creep test.
Specimens were KIBUSHI clay which had been used in a series of experimental studies by the authors.
Cylindrical specimens of 3.5 cm in diameter and 8.75 cm in height were used. The method of this test was as follows:
(1) When the stress level in the specimen amounted to some percentage of the shear strengtb., the loading to the specimen was stopped and creep distortion was made to occur in the specimen for 24 hours. The creep stress levels (q_c/S₀) that occurred in the creep distortion were about 50%, 60%, 70%, 80%, 90% of the shear strength.
(2) After the specimen had been subjected to the creep distortion, the stress applied to the specimen was released.
(3) The shear strength was measured by the undrained triaxial compression test.
The experimental results indicated that the increase of shear strengh (σ_cf/S₀) after experiencing the creep history on the specimen increased linearly with the creep stress level (q_c/S₀).
However, it may be thought that there is not linearity when the creep stress level of 75-80% or not less than the shear strength was applied to the specimen.
The relationship between the quantity of the creep strain and the creep stress is described by the following equation:
ε_c=Exp(6.53q_c-3.09).

View full abstract