Summer pressure pattern is the case, when the subtropical anticyclone over the NW-Pacific expands over Japan, while NE-Asia is still under low pressure. Meteorologists in Japan call it as the “south-high-north-low” pattern. The appearance of this summer pressure pattern was studied for the years 1907-1966.
Mean annual number of days with this pattern reaches 25.2 days. Its number varies, however, from 9 days in 1910 and 1954 to 45-47 days in 1946 and 1947 as shown in Table 1. Its secular change does not show any clear tendency or cyclic variation as given in Fig. 1. The occurrence frequency increases sharply from July 15, coming to maximum on July 25-29. After a pause during August 4-6, it appears again very frequently from August 8 to 13. The daily statistics and the five-day totals of appearance are given in Tables 3 and 4 respectively.
The five-day composite maps of anomaly of 500mb topography, sea-level pressure and air temperature anomaly were made for the respective patterns: (1) summer, (2) frontal and (3) typhoon pressure patterns. The definition and the calendar of these pressure patterns were described in Meteorologische Rundschau Vol. 21, 1968. The composite maps are given in Figs. 2-10. Comparison of the maps with each other came to the following conclusion: (1) Plus anomaly of air temperature appears over Japan in the case of summer pressure pattern. The ridge line of the NW-Pacific anticyclone in this case expands to west, south of Japan. Accordingly, the warm southerly flow prevails over and arround Japan. (2) Minus anomaly of air temperature prevails in the case of summer pressure pattern in the region, where the low pressure area expands from the continent. The border line of the region coincides with the 1010mb isobar. (3) When the minus anomaly region of the 500mb topography occupies the Sea of Okhotsk from the continent, sea-level pressure becomes lower and also minus anomaly of air temperature in the same region. (4) If its minus anomaly region occupies over Japan, it is the case of the frontal pressure pattern: Minus air temperature anomaly prevails over Japan. (5) In the case of typhoon pressure pattern, the low pressure appears in the regions from East China Sea, via Korean Peninsula, to the Sea of Okhotsk and the Aleutian. The regions with minus anomaly of air temperature coincide roughly with them. (6) The areal mean of air temperature anomaly around Japan, 35°N×130, 135 and 140°E, reaches +1.36°C in the case of summer pressure pattern. On the other hand, the mean around northern Japan, 45°N×130, 135 and 140°E, does -2.13°C in the case of frontal pressure pattern.

View full abstract

The southern Sanriku coast is a typical rias coast in Japan with serrate shoreline ranging from Miyako Bay to Ojika Peninsular and poorly developed coastal terraces, is in striking contrast to the northern Sanriku coast with broad raised coastal terraces and nearly straight shoreline. The rias coast is considered to be the submerged maturely dissected margins of Kitakami mountains mainly consisting of pre-Tertiary rocks, and the coastal terraces along the embayments of the rias coast prove the uplifts of the area. The geomorphological history of the area seems to be a more complex one.
The development of coastal landform of the area are divided into two stages; the stage of drowned embayments and the stage of coastal terraces.
Judging from the Senganda. Formation (uppermost Pliocene) in the western coast of Kesennuma Bay and the piedmont steps or lower peneplain (pre-Pleistocene) in the northern Sanriku coast, the forming of rias coast outline as submergence coast is dated back to Plio-Pleistocene. It could be assumed that the submergence had been brought by a tectonic movement. At the following stage the coastal terraces have been formed inside the main embayments, and the stage is regarded as the stage of modification of rias coast, when the rias coast was modified by a series of submergence and emergence occurred alternatively, which is proved by the terraces surfaces and terrace deposits. Glacio-eustatic change of sea level seem to have been the principal factor in the geomorphic evolution during the stage, especially during the latter half of the stage.
The coastal terraces can be classified into five levels; from the upper to the lower, Takada, Marumori, Sakari, Ofunato and Kadonohama Terraces at Ofunato region, and Toyomane, Omoe, Namiita and Lower Terraces at Miyako region. The levels of coastal terraces at both regions may be correlated respectively.
Broad raised terraces developing along the northern Sanriku coast from Miyako to Omoto are designated as Settai, Taro and Furusato Terraces from the upper to the lower. These terraces are characterized by the higher elevations and the southward inclinations of their surfaces. This region appears to have uplifted by crustal movement more than the southern rias coast region where has been less influenced during the terrace forming stage.
At Miyako region, higher terraces are well developed, but lower ones poorly, and the deposits can be observed at a few locations. At Ofunato region, on the contrary, the terrace surfaces are narrow in general, but the deposits of the lower terraces are observable at many locations. Hence, the process of terrace forming is well recognized at this region. Especially the deposits of Sakari and Ofunato Terraces are composed of fluvial beds in lower part and marine beds atop. Dissecting valleys, which had been cut during the regression of sea, were buried beneath these deposits. There are no critical evidence on the chronology of these ter-races, but according to the height of Ofunato Terrace surface and the red weathered crusts and soils covering Ofunato Terrace which is the lowest terrace among the terraces covered by the red weathered crusts and soils, it may be assumed that Ofunato Terrace had been built during the time of Riss-Würm interglacial.
Recent drowned valleys, corresponding to final stage of the history of Sanriku rias coast, are found inside the main bays. Some of them are obscured by subsequent deposition or remain as gorges without deposition. These valleys which were cut under the condition of low sea level during the last glacial, have been drowned by the post-glacial transgression.

View full abstract

In Japan, the phenomenon of the concentration of population into cities is one of the most serious problems. Laws off migration by Ravenstein, E.G. have still gotten a footing in investigation into the national-wide migration with special reference to cities. Quantitative analyses have been applied to his findings since by many investigators. The present writer aims at understanding the characteristics and regularities of the recent migration by applying for the first time the following two statistical techiniques: coefficient of evenness derived from the Lorenz curve calculated on prefecture level (abridged as coefficient of evenness) and the coefficient of correlation (γ) between moving-in and moving-out population. The purpose of this study is to deal with this phenomenon from the viewpoint of population geography. The main object of the study is the moving-in and -out population in large cities, and, for comparison, the transference of population in all prefectures is of the secondary concern. The years of 1959 and 1966 were chosen for the calculation, and the attention was paid also to the changes in population during the period between 1959 and 1966.
The results are as follows:
i) In 1959, the coefficients of evenness of the moving-in population in the six large cities are respectively 0.493 for the ward areas of Tokyo, 0.388 for Osaka-city, 0.375 for Kyoto-city, 0.304 for Nagoya-city, 0.296 for Kobe-city and 0.291 for Yokohama-city. In 1966, the coefficients in the seven large cities are 0.497 for the ward area of Tokyo, 0.406 for Kyoto-city, 0.369 for Osaka-city, 0.349 for Nagoya-city, 0.304 for Yokohama-city, 0.298 for Kobe-city, and 0.219 for Kitakyushu-city. Since the coefficients of evenness of the moving-in population in large cities correspond to the population capacity of cities, the coefficients may safely become a parameter of measuring the levels of centrality in large cities. In case the coefficients of which cities the coefficients of evenness the moving-in population in the year of 1959 is above 0.300 and that of 1966 is above 0.340 in a given city, it may be fitly regarded as the national centre, and those in which the coefficients were below the values mentioned above the regional or local centre, the ward area of Tokyo, Osaka-city, Nagoya-city and Kyoto-city come under the category of the national centre, while Yokohama-city and Kobe-city fall into the category of the regional centre. In the case of Kitakyushu-city in 1966, it may be classified as the local centre, as the coefficient of evenness was extremely small. The differences in the coefficients of evenness of the moving-out population among the large cities are relatively small, as compared with those of the moving-in population, and the range of the former is less than one-third of the latter. It is simply because of the common tendencies of the large cities that were caused by the transfer of residential sites from the cities: the move-out of dwellers in the urban centre to the short-distance areas is striking in the large cities. By these reasons mentioned above, the coefficient of evenness of the moving-out population is not applicable for a measurement of centrality in large cities.
ii) The average distances of the moving-in population into the six large cities, in 1959, were 286 km in the ward area of Tokyo, 225 km in Kobe-city, 243 km in Nagoya-city, 223 km in Osaka-city, 207 km in Yokohama-city and 201 km in Kyoto-city, and the grand average distance of these cities was 261 km. In 1966, the average distances of the moving-in population into the seven large cities were 309 km in Kitakyushu-city, 292 km in the ward area of Tokyo, 271 km in Nagoya-city, 242 km in Kobe-city, 239 km in Osaka-city, 231 km in Yokohama-city and 214 km in Kyoto-city. The grand average distance of the six large cities, excluding Kitakyushu-city, was 269 km, longer by 8 km than that of 1959.

View full abstract