From the observations of the Kasaoka bay polder after land draining, we found that the muddy soil surface was subdivided into a soil mass (column) of a tortoise shell pattern by a widely advanced large crack, the mass of which became the fundamental unit of the subsequent salt and water behavior in the soil layer. Based on the above mention, we first investigated the actual conditions of the crack development which was thought to decide the character of the early polder having occurred two times in the summer and winter using a lysimeter. Next, the yearly change in the salt behavior in the soil mass with a large crack was thoroughly investigated.
Consequently, we were able to clarify the importance of the crack which governed the conditions of the salt behavior up to this time and which is expected to control salt exclusion. Then, we classified and adjusted the movement of the salt in the soil mass based on the fundamental idea of salt exclusion, which was thought to be able to describe the actual conditions of the salt behavior under the time scale after land draining.

View full abstract

This paper describes the difference in the splash formation owing to the solidity of an water drop.
In order to clarify the influence of solidity for the difference in the splash formation, and using some basic information concerning soil detaches, an experiment was performed using four factors with two levels-large and small levels: the diameter, the surface tension, the kinetic energy and the hardness of a colliding plane. The results are as follows.
1. The number of splashes differed with each falling range. It showed the peak in the vicinity of the colliding point, and decreased in order. This peak had a tendency to be raised and pointed near the colliding point, and to become gentle and low conversely in the distance. The former was suited to the splashes by the soft colliding plane and the latter was the hard plane.
2. The diameter of the splash differed with each falling point, and it showed a tendency to increase in distance. This tendency satisfied the following relation between diameter φ(mm) and range 1 (cm).
φ=a.eb.l (b>0)
However, as for the hard plane, large splashes were observed evidently close to the colliding point.
3. The average diameter of splashes had a large value as regards the water drops, large drops and hard colliding plane.
4. The specific diameter (the ratio of a splash diameter to a colliding drop diameter) was small for the large drops, so it showed the crushability of the large diameter drop. And it became small in the case of the solution drops, so the crushability of solution drop with less surface tension was affirmed.
5. The flying range l (cm) became long in the case of the hard colliding plane and the high kinetic energy level.

View full abstract

For the purpose of nondestructive determining of soil water content near the surface of a sandy field, a surface-type neutron moisture meter with the capability of making backscatter and direct transmission measurements was introduced.
In this paper, the response on this surface-type neutron moisture meter at source positions of 0, 5, 10, 15, 20, 25, 30 cm was described by an experiment using dune sand with different water contents. To know how large the sphere of importance of a surface-type neutron moisture meter is, the influence of the presence of a paraffin mass on the count ratio was examined for the dune sand in the vertical and horizontal directions. In case of the backscatter method, the radius of the sphere of importance was obtained as the function of the volumetric water content. Furthermore, calibration curves of the surface-type neutron moisture meter for the source depths of 0, 5, 10, 15cm were determined for a dune sand in a calibration vessel. Therefore, the utility of the calibration curves to measure the soil water content in a sandy field was discussed by the experiment with a relatively higher water content in a lysimeter during drainage and by core sampling with a relatively lower water content in an actual sandy field. As a result, it was recognized that the calibration curves for source depths of 5 and 10 cm could be practically used to monitor the soil water content near the surface in an actual sandy field within the measuring error of 1% by volume.

View full abstract

In Japan, the rice suffered damage due to cold weather in 1980. We carried out a digital analysis of the Landsat MSS data (116-30, 1980. 9. 19) to obtain a geographical distribution map of the damaged rice brought about by the cold weather. The area to be analyzed (53 × 65 km) includes the Ishikari plain where is one of the very large paddy field areas in Japan. As a result of this analysis, we were able to make a distribution map of the extent of rice damage and rice yield of the analyzed area.
(1) We used the Landsat MSS data for land-use classification, and we classified the area to be analyzed into eight land-use classes (paddy fields, dry fields, grass land, forest, strip ground, towns, sea, river or lake) by a supervised classification. In this case, we used the maximum likelihood decision rule. The classification performance was 96. 7% on the average for the training fields.(Photo. 1, 2; Table 1)
(2) The color of the leaves and stalks of non damaged rice was “Koganeiro (golden or yellow color) ”, and the heads were heavily laden with seed. But, the appearance of the damaged rice was “Aodachi (poor stand and green in color) ” because of the immaturity and sterility. So, we investigated the spectral reflectance of the damaged rice concerning the data from the handy spectrometer and Landsat MSS. Then, we found that the reflectance in the red wave decreased and VI (Vegetation Index) value increased in proportion as the damage enlarged.(Fig. 2-4)
(3) We were able to make the distribution map of the extent of the rice damage by level-slice classi-fication of band 5 or VI. This level-slice classification was performed on the all pixels which were classified into paddy field by the land-use classification. This map shows the distribution characteristics of the rice damage from cold weather in the Ishikari plain.(Photo. 3)
(4) We carried out a multiple regression analysis for the rice yield estimation. Then, we made a multiple regression model of the rice yield in which the multiple correlation coefficient was 0. 91**.
Y (kg/10a) =7673.5-61.4. band 5+67.2·Eband 6-62.4 VI (R=0.91**)
As a result of the application of this model to the Landsat MSS data, we were able to make a distribution map of the rice yield of the area under analysis.(Photo. 4)
(5) Much information of various types could be extracted from the distribution map of the rice yield which was made by the digital analysis of Landsat MSS data. This shows that Remote Sensing is available for a large area survey technique.

View full abstract

This paper examines the results of an experimental study on the basic problems hydraulics in order to develop and make a rational design of a torrent intake works of a collecting conduit type.
The results are summarized as follows.
The relation equation for the percolation velocity (V) and friction head loss per unit length of layer (ΔH/L) can be expressed by the Forchheimer law, when the state of the flow passing through packed columns of sand and gravel shows a transition zone between laminar flow and turbulent flow.
ΔH/L=aV²+bV
But if the constant range of ΔH/L is limited in the range of 0.4-2.0, the relation between V and ΔH/L is shown by the following equation.
ΔH/L= (V/K) n
K= [k^-1·φ^3-n·Dp^3-n·ε³·(1-ε) ^n-3·μ^n-2·ρ^n-2] ^1/n
The value of 1/n in the above equation expresses the gradient of a linear equation which can be obtained from the logarithm value of V and ΔH/L, and it can be expressed as the function of the modified Reynolds number. This relation is given in Fig. 8. And it became evident due to the experiments that the values of n change within the range of 1.0-1.9. There is a tendency for the values of n to become larger as the percolation velocity of water passing through the packed columns of the sand and gravel increases and the mean diameter of the sand and gravel becomes larger.
For the transition zone between the laminar flow and turbulent flow, the values of experimental constant k in the fluid head loss equation can be expressed as the function of the modified Reynolds number. This relation is given in Fig. 10.
The experimental results we obtained gave us a clear understanding between the percolation velocity and head loss in the transition zone between the laminar and turbulent flow.

View full abstract

Up until now it has been believed that there is a control section at the crest of a weir even in an unsteady flow.
In this study, an attempt is made to analyze theoretically various concepts of a control section operative (present) in hydraulic systems under unsteady conditions.
Verification of the results were made by numerical simulation using the characteristic method. In this analysis, however, for simplisity, it was assumed that the flow is one dimensional and non viscous. Also in an unsteady flow the critical section (where u=√gh) at the crest is called the “control section”.
The following results were obtained:
(1) When a negative small surge runs over the crest, the control section moves accordingly downstream near the crest. In the case of a positive surge, the section moves upstreamnear the crest. These relationships are represented by Eq.(25).
(2) In an unsteady flow, the position of the control section is these different from that of the minimum specific energy section.(see Table 1)
(3) A small disturbance occurring near the downstream part of the control section can be transferred in the upstream direction as shown in Fig. 6. In other words, the state of the flow between points x_D and x_C in Fig. 3 influences the upper flow profile.
Based on these findings, control section concepts in steady hydraulic systems do not seem particularly significant in unsteady hydraulics.
Based on the results of this study, the so-called “control section” under steady conditions cannot be applied directly to an unsteady flow.

View full abstract

Turbulence is caused by the roughness of a river bed due to sand and gravel of various sizes and forms.
Bed-loads are produced by means of sand and gravel being picked or chosen out in a river bed that are equal to or smaller than the vortexes which are caused by turbulence.
My theory that is based on such an idea has been proven with some of the representative data that have been recognized as valuable by some popular papers in this field. But there is no example of my own research applied to this field, so I have tried to prove this theory through an analysis concerning the movement of a river bed as function of time that has been introduced by this theory to the down stream of the Yoshii River in Okayama.
There are two different sides regarding the analysis of the movement of a river bed.
One of them is the variation with time in the profile of a river bed, and the other is one about the profile of a river bed at a given time. The variation concerning the profile is influenced by the hydraulic characteristics at the beginning point that is going to be calculated. This paper discusses the former. The following data must be gathered or calculated: the real quantity of a river bed movement and the hydrograph of a flood during the term, and discharge in critical tractive force in each section, its frequency, etc. The movement of a river bed in some points down stream is calculated on the basis of the hydraulic characteristics at the beginning point. The coefficent of the movement of a river bed at each point from the beginning point of calculation can be calculated by using the above-mentioned data. And the following results have been obtained, that is, the theoretical results agreed well with the values of the survey.
The profile of the latest river bed seems to be decided by a rather small flood, that is a near critical tractive discharge. And the following will be able to be expected, that is, this theory may be sufficiently accurate in the present level that approves of coincidence with rate.

View full abstract

Every rational method of pavement design has many simplifying assumptions but requires the use of elastic properties for each layer of pavement. The purpose of this paper was to evaluate the elastic moduli of subgrade and subbase course of a farm road experimentally. Three kinds of subgrade soil were molded in test pits at different places. Unscreened crushed stone was laid on the surface of the subgrade soil, and cement stabilized weathered granite soil was laid on it. By the use of a circular disk, a deflection test was performed on the surface of each layer. An unconfined compression test and a bending test for the test pieces were also performed for the cement stabilizing soil.
Test results showed that the moduli of deformation of the subgrades changed slightly with load intensity on the plate. The moduli of deformation of two subbase courses were evaluated by three methods. The first was the linear method, in which the modulus of deformation of subgrade was assumed to have a constant value. The second was the quasi-linear method, in which the moduli of deformation of subgrade and subbase courses were assumed to change whth the vertical normal stvesses at the surface of each layer. The third was the finite element method, in which the moduli of deformation of each layer were assumed to change with stress invariant of each element. The results of the finite element method were closer to those from the quasi-linear method than those from the linear method. The modulus of deformation of the crushed-stone layer was not the same when the subgrade soil under it was different, but that of the cement stabilized soil was almost the same for all three kinds of subgrade. The modulus of deformation of the cement stabilized soil layer was changed with load intensity or with stress invariant, and the results from the compression test and bending test for the test.pieces showed quite a different value. But, the results from the test pieces were similar to those of their lower and upper limits, obtained by the finite element analysis.

View full abstract

It is our purpose to study saturated-unsaturated seepage through fill-type dams by using a numerical approach and to give some guidelines for their design. The numerical procedure is based on the iterative Galerkin-type finite element method which was developed by S. P. Neuman (1973). This first paper gives an outline of his approach in a two-dimensional flow system and shows some problems on its application. Numerical examples are compared with experimental, analytical and saturated finite element solutions for one-dimensional unsaturated flow, steady and transient seepage with the following being concluded:
(1) Given appropriate moisture properties of unsaturated soils, this approach gives a rational solution to the problem of flow through porous materials.
(2) It is suitable especially for studying unsteady seepage in which unsaturated flow may play an important role.
(3) It cannot be applied to the flow through extremely dry soil.
(4) To make the moisture properties of unsaturated soils clear is important for further development of saturated-unsaturated seepage analysis.
Using the approach given by Neuman, in the our related. paper, we will discuss the characteristics of saturated-unsaturated flow through zoned embankment dams under steady conditions and rapid drawdown in a reservoir.

View full abstract