There are 61 stations in Japan which have successive rainfall records for more than 60 years up to 1961. For these stations, residual mass-curves of their annual rainfalls were constructed respectively based on the 30 years normal values of the period 1921-50.
_??_
y_i: cummulative rainfall index for i year
R_i: annual rainfall of i year
R: amount of normal rainfall (mean of 1921-50)
C: constant=100_??_=summary cummulative index of the period from the starting year of the observation to 1920.
Three different patterns of the mass-curves were distinguished and they had their own distribution areas:
1) The front Japan (Pacific Side) type—The curves showed a continuous upward trend up to 1920. This meant that most years of the period 1880-1920 had the above normal rainfall. Average states followed successively for about 30 years. Since 1948 thee recent remarkably rainy period has continued until the present, having about 10% more rainfall annually than the normal year on the average Particularly it was rainier in the western Japan (Figs. 2, 3).
Winter rains (Oct.-March) were more variable than the summer ones (April-Sept.) a Especially the magnitudes of the recent increases were remarkable (Fig. 9).
2) The rear Japan (Japan Sea Side) type—In general, the changing patterns were somewhat similar to those of the front Japan (Fig. 4). However, the magnitude of year to year variation was much smaller and was only about ±2-3%. Therefore, it was rather difficult to distinguish the existence of the rainy period from that of dry ones (Fig. 5) . The increasing tendency of the recent years was not distinctive. The same changing patterns were recognized for their secular changes of summer and winter rains too. Concerning rainfall, succession of the average year or average summer and winter was characteristic of the region (Fig. 10).
3) Hokkaido type—The changing pattern was just the opposite to the front Japan type. Dry periods were clearly recongnized during the period 1880-1920 and for the latest decade (Fig. 6).

View full abstract

The Sado Island with 857 km² in area is situated in the Japan Sea, about 35 km off Honshû. This islandd consists of two regions, Ôsado and Kosado. The former region which lies in the northern half is larger than the latter and also higher in altitude, and considered to be a tilted block with steeper foreslopes on the east side. This area is composed mainly of volcanic rocks such as andesite, liperite, basalt, and Miocen green tuff. Surrounding this island are coastal terraces developed widely. The author surveyed these coastal terraces and their deposits in the Ôsado region, in order to clarify their distribution, altitude, and nature and to consider the geomorphic history. The main results are summarized as follows:
1) The coastal terraces are divided into six plains owing to their altitude and continuity. They are as follows: the first terrace (160-220 m in altitude), second (80-140 m), third (60-120 m), fourth (35-70 m), fifth (25-40 m), and the sixth (5-8 m).
2) These terraces are well developed in the western coastal region in general. On the east side, however, they are no more than a fragmentary distribution. Among the terraces of the Ôsado region, the third and fourth terraces show a broad and cotinuous distribution. The sixth terrace is also developed all around the island in spite of the limited width.
3) Considering from the thickness and facies of the terrace deposits, the nature of the terrace plains can be classified into four groups. A a In the western coastal region, every terrace plain is of a marine origing formed by striking abrasion of the open sea. B: On the coast along the Mano Bay, the terraces are thought to be of a marine origin formed by the cut and built action of waves in the inland bay. C: In the Kuninaka Plain, the second and third terraces are elevated fans built by the rivers from the northern mountains. On the contrary, the fourth terrace is a dissected coastal plain which consists of thicker neritic sediments deposited in the inland bay. D: In the eastern coastal region, marine terraces are hardly recognized, and most of the flat plains are very small elevated fans consisting of poorly sorted boulders transported by the torrential streams down from the steep slopes.
4) The sixth terrace continues to the alluvial plain of the Kuninaka area, composed of marine alluvial deposits containing the fauna which indicates warmer water temperautre than the present sea in the vicinity and having about 70 m in thickness. Therefore, the formation of the sixth terrace plain is to be correlated with the transgression of early alluvial age which is recognized at many places in Japan. As for the fourth terrace which has a wide distribution and thicker neritic sediments, it may be considered that it was built by the transgression preceeding the regression of the last glacial age just before the alluvial trans-gression.
5) The differences in the character of terrace plains which were mentioned in 3) seem to haue been influenced chiefly by the asymmetry of topography and the condition whether open-sea-facing or not. Rock control is recognized where stacks are found on the terrace plains in the volcanic rock areas of Ôsado. Concerning the formation of the terrace as a whole, however, it is of the secondary importance, since topographic differences such as those of slopes seem to be the primary differentiating causes.
6) On every marine terrace plain, the elevations of the ancient strand lines increase in an accelerated manner towards the middle part of both sides of the Ôsado coasts, and this tendency is more evident on the older terraces. This means the upwarping movement with the axis at the middle part of the island has continued since the formation of the first terrace plain.

View full abstract

It is well known that the existence of the city temperature was confirmed in various cities in Japan from the results of previous researches. However, temperature distribution is affected by many factors such as topographical situation, size and functional character of a city, and so on. So, selection of the city where these factors are as simple as possible becomes necessary for analysis of the detailed distribution of temperature in and around the city. Recently a special project for the investigation of city climate was undertaken by the Research Group of City Climate in Japan and an intensive observation was carried out in several cities from this point of view.
The purpose of this paper is to analyze, by using the data from the above-mentioned project, the detailed distribution of temperature in and around any urban area located in the central part of a plain, which is thought to be not affected by surrounding topographical releafs without much influence of season and weather conditions.
Kumagaya City is located near the center of the Kantô Plain and had a population of about 50, 000 in the urban area in 1956. There were neither houses and buildings constructed with concrete or stone, nor factories generating a large amount of heat. The urban area of Kumagaya City is represented by the lines of 10 percent building coverage as illustrated in Fig. 2. This area is shown with a darker tone in Figs. 2, 3, 4, 5.
Observations were repeated in the ealy morning, daytime and night in each of four seasons from Septmber in 1956 to March in 1957. Intensive net-work of temperature obtained by mobile observation is represented on Fig. 1, using 2 sets of sensible thermistor thermometer mounted on the automobiles, Accuracy of the readings of thermometer is 0.2°C, and the representativeness of observed temperature is about 100m in the vicinity of the observation point in this study. Then, the results of observation are sufficiently able to clarify the purpose of this paper after time correction.
Typical examples of the temperature distribution in Kumagaya City are shown in Figs. 3 and 4. Distribution of temperature in calm condition is drawn in Fig. 4 by a composite map method, as data were not available for actual calm condition. The results always show the existence of city temperature though the value of it seem to vary with the difference of season and the variation of weather condition. Above all, the effect of wind is apparently perceived on the temperature distribuion in Fig. 3.
Then, the effect of wind is analyzed in detail. Temperature distribution in calm condition has a relatively close correlation with the map for the distribution of building coverage in Fig. 2. The area of the highest temperature situates at the busiest part of the city in calm weather. In windy condition, however, location of the area of the highest temperature removes leeward in the urban area, as shown in Fig. 4. (Arrow indicates wind condition and dots represent observation points. The top of arrow shows the position of the center of the highest temperature area for corresponding condition.) Moreover, wind effect is vividly perceived in Fig. 5 that shows the departure of temperature distribution in windy condion from that of calm weather. Relatively cool air from the surrounding country flows into the urban area through area of low-rate building coverage.
Finally, the relationship between the city temperature, defined by the difference of maximum temperature observed in the urban area from minimum temperature that represents the surrounding area, and weather factors expressed by the records of the Kumagaya Local Meteorological Observatory is represented in the linear empirical formula by least square method.
D=4.21-0.08N-0.12n-0.52V+0.01e-0.04T (night)

View full abstract