The variation of discharge and water quality of streams with time at intervals of three hours were measured in the representative six rivers in the Kasumigaura lake basin. The mean value and the coefficient of variation are shown in Table 1. With regard to the specific load of some pollutional materials, e. g. Cl, the standard deviation of variation with time is large as the mean value becomes higher (Fig. 2-6). The other materials, e. g. NH₄-N, the mean value in the Seimei River is high. but the variation is relatively constant (Fig. 2-7).
As for the respective rivers, variations with time of individual values of flowing discharge and load divided by each mean value are shown in Fig. 6. In the Sakura River, which has the largest basin area of 349 km², all of the variations of load are relatively small (Fig. 6-1). On the other hand, in the heavily polluted Sanno River the amplitude of variation is large, especially in PO₄, COD, NH₄, and NO₂, but the pattern of the variation is comparativery simple, which means that the phases of variation wave of different pollutional loads are coincided due to the centralization of main pollutional sources most by in the mid-stream (Fig. 6-6). Moreover, in the Sonobe River the pattern of variation is considerably complicated. The reason for this is that in the basin of this river various pollutional sources, e. g. factory, fecal matter treatment, and pig farm, which have fairly different water qualities are scattered separately along the stream (Fig. 6-2). Thus, the pattern of variation with time in a river considerably differs according to the conditions of pollutional sources in the basin.
For the study of water pollution in a river, the consideration of specific load is significant as already described in papers. But in the computation of total load flowing into the Lake Kasumigaura, the load of individual river is needed, and the standard deviation of the total load, ε, is of the form:ε=√ε₁²+ε₂²+......+ε_n², where ε₁, ε₂, ε_n, are the standard deviations of load of individual river. Therefore, since the standard deviation in almost all of pollutional materials is of high value in the Sakura River, the average error of total load is influenced strongly (Fig. 9).

View full abstract

There has been a growing concern to the eutrophication of the lake Kasumigaura, the second largest lake in Japan, since the lake is an indispensable water resource for the Kanto area. To determine the cause of water quality degradation of the lake, it is important to reveal the effects of the drainage from various sources including industrial and sewage waste water both on the discharge and water quality of streams. Therefore, the investigation of discharge and water quality of streams which flow into the lake has been carried out since 1973. Individual streams are distinctive in catchment area, population density and point source etc.(Table 1).
The specific pollutional load represents the load of pollutants which flow into the lake from a unit area of stream basin. The specific pollutional loads in such streams as Sanno (No.13), Sonobe (No.5) and Hanamuro (No.7) were much larger than those of streams in rural area (Figs. 5-7).
Total loads which flow into the lake through all 12 researched rivers are shown in Figs. 9-11. These researched rivers cover the basin area of 900 km². The load belonging to a face source is estimated as the load that enters into the lake through the river in which any large point source does not exist. The difference between the total load and the face load is large and it is due to the load from large point sources (L. P. S). In other words, the value of total load is raised due to large point sources, especially in T-P.
The annual total load and the load of L. P. S are shown in Table 3 in which the percentages of L. P. S in the total load are large.
The load of nutrients input to the Kasumigaura lake basin by precipitation and irrigation water from the lake is calculated. The concentration of rain water is 0.8-1.1 ppm as T-N (Table 4) and this load attains to 910-1250 ton/year in the researched area basin (900 km²). T-N load of irrigation water is 80-140 ton/year in this area. Inputs due to these two loads are almost equal to the value of output in stream water named as the face load (Fig. 12). Some part of the face load may be occupied by the passing load added to the basin by precipitation and irrigation.
The concentration of nutrients increases in such a process of water flow as rain→face outflow (in rural area) →final outflow to lake (Fig. 17). The concentration of final outflow water is almost two times higher than the concentration of face outflow water (Figs. 14-16). The effects of large point sources (L. P. S) on the increase of concentration of water flowing into the lake are significant. Reversely the concentration of mountain water is lower than that of rain water (Fig. 17). Even though the load is large, this kind of low concentration water from mountains has no effect on the rapid increase of lake water concentration. Not only the loading but also its concentration must be discussed about water pollution.
Thus, we consider that the main cause of eutrophication of Kasumigaura lake is due to the load of L. P. S, and this load must be eliminated first for the purification of the lake.

View full abstract

Runoff characteristics of each individual paddy field must be noticed for the runoff analysis of paddy field area based on the parametric system synthesis.
In this article, a stochastic method is offered for evaluation of the aforementioned characteristics.
First, the mechanism of runoff through an outlet is noticed and the coefficient of discharge is experimentally examined with paddy fields in the model basin. In addition, such factors as the elevation of outlet sill, the surface unevenness of field, the ponding depth (depth of flooding water), the height of border, etc. are taken up to be the important parameters for the runoff and their statistical characteristics are investigated.
Secondly, a computer simulation method of rainfall-runoff relation is mentioned. Parameters concerning runoff are determined for each paddy field by using random numbers submitted to measured distribution functions in its program. The discharge from each field and the depth of flooding water are calculated by using the coefficient of discharge and the equation of continuity, in which the rate of ponding area and of residual rainfall on dried area are considered especially.
The application of this method to the runoff analysis in the model basin is practiced. The computation of an unsteady flow in the drainage canal is carried out by the simplified characteristic method. Comparing with the observed hydrograph, this simulation shows good applicability of this method for the aforementioned aunoff analysis considerably.
Further, effects of boundary conditions such as the initial depth of flooding water, the surface unevenness, etc. on the resulting runoff and also effects of various initial values in this pseudo-random number procedure are examined. The numerical results show that the effects of these parameters and initial values on both peak and total runoff are insignificant.

View full abstract

In the study of underground water in a wide area, there are many cases we must deal with the internal boundary domain of the neighboring rivers and lakes.
As the water level in the internal boundary is often previously given, the calculation of underground water variation is not governed by equation (1), therefore, the mathematical simulation of underground water in the wide area becomes sometimes complicated.
We tried to set up a method so as to be able to calculate the whole underground water variation in the domain under consideration by using only one equation-the equation (1), and applied this method to the lake Yogo area as an example.
In this example, the following assumptions were made:
(1) The lake is much more permeable than the surrounding area (Fig. 6 and Fig. 7).
(2) In the lake, the ratio of rainfall entering the aquifer is 0 (Fig. 9) and the evapotranspiration is also 0.
(3) In the lake, there occurs a difference between the given lake water level YOGOH (N) and the lake water level {SH (I, J, N)} calculated by equation (1) at the time N.
Then, we regard this difference as rainfall (difference of (+)) or evapotranspiration (difference of (-)) and make some correction by adding or subtracting the difference in the formula calculating the underground water level at the time (N+1).
Thus, we obtained good results for practical use as shown in Fig. 3 and Fig. 4.

View full abstract

Time-series analysis was employed in analyzing the water temperature of the Sodekawa warming canal in Iwate Prefecture. An irrigation period record (from March 1 to September 1, 1969) of hourly water temperature obtained at two points along the warming canal was used for the analysis. The distance between these two points (the inlet and outlet of the warming canal) is 1, 345 meters.
The sequence of the water temperature was first separated into the seasonal deterministic component, the periodic deterministic component, and the stochastic component. Then, three mathematic models were applied for these components, i. e., (1) quadratic equation model for the seasonal deterministic component, (2) Fourier series equation model for the periodic deterministic component, and (3) the first-order autoregressive model for the stochastic component.
A simple model consisting of the quadratic equation model linked with the Fourier series model and the first-order autoregressive model could account for 99. 0% to 99. 3% of the total variance of observations.
Based on the fitted model, we have provided a useful forecast function for the water temperature of 1 hour to 3 hours hence and an accurate regression equation for the water temperature measurements at the inlet and outlet of the warming canal.

View full abstract

One of the reasons why the road in rural settlement. remains still unimproved, lies in that we have no indexes which set the goal of road improvement. And the reason why we do not have such indexes comes from the conceptual uncertainty of the rural settlement.Namely, we could not find appropriate indexes concerning the state of settlement.
Therefore, in this report, we define the physical settlement as the homestead-road-system. From this definition, we also define the rural settlement-road as the road which combines every homestead in the region of rural settlement.
Through an actual survey of two towns, Chuuz-cho and Kooga-cho, in Shiga Prefecture, we could find feasible the following indexes for road improvement:
1. As an index for settlement state, the road length per homestead is useful, that is,
l=L/(f_h+f_c) ≅L/f_h,
where l: road length per homestead
L: length of rural settlement-road
f_h: number of homesteads
f_c: number of community facilities.
2. As indexes for road improvement, road-length rate and road-contact rate are useful, that is,
m_i= (l_i/L) ×100
c_i= (f_i/f_h) ×100
where m_i: road length rate of road width level i
l_i: road length of road width level i
c_i: rate of homesteads in contact with road of road width level i
f_i: number of homesteads in contact with road of road width level i

View full abstract

The backsand phenomenon in a steep slope river was studied experimentally and the following facts were found:
1) The backsand mechanism is classified in to 3 types according to the flow discharge and bed load. Namely, it changes from the stepped type to the transition and undular types with increase of the bed load. And the state of undular type backsand corresponds to that of antidune in the regime criteria for alluvial streams.
2) The bed load over the reservoir sedimentation is varied with the progress of flood, and approaches a definite value q_Bf after lapse of a long time. The slope of sedimentation also approches a definite value I∞ at the same time. The ratios, q_Bf/q₀ and I∞/i (q₀: total bed load in the upstream of river or supplied bed load, i: initial bed slope of river) are related to q₀/q_B (q_B: bed load transportation capacity of river for initial bed slope). The volume of backsand is affected by the type of backsand mechanism largely.
3) Assuming the rule of formation of sedimentation delta based on the test results, the movement of the beginning point of deposition and the front of delta with the lapse of time were estimated theoretically.

View full abstract

1) To evaluale the recent development of greenhouse food and flower production, the author has investigated the engineering problems of soil sickness and considered how these injuries can be avoided by the use of certain techniques.
2) Methods to avoid soil sickness are various as are the commercial practices of the growers. The causes of soil sickness are as get unknown and moreover it cannot be complettly avoided.
3) On the basis of case studies, a choice of several methods to reduce soil sichness can be made.
4) Countermeasure costs have been analysed and the complex commercial conditions of the growers reduced.
5) From the results of the analysis, it is found that many methods of soil improvement (e. g. under-drainage, subsoil improvement etc.) can be effectively used.
6) Before there techniques can be used innovations of greenhouse production have to be made in association with land rearrangement.

View full abstract