The phenomenal annual variation in temperature of the shallow groundwater in Egawa Region, Tokushima Prefecture, has attracted hydrological interests since the 1930's. Water temperature in Egawa Spring (ES in Fig. 1) reaches its maximum in early winter and minimum in early summer. Approximate mean annual extremes of the temperature are about 23°C and 8°C, respectively, for winter and summer.
The authors assumed that this phenomenal temperature variation would be caused by ground-water flow in the wide area around Egawa Spring. Hence, observations on temperature and level of groundwater were made along with the direct measurement of groundwater flow.
The shallow aquifer in the region is non-artesian and consists of medum or large size pebbles (Fig. 2), The permeability of the aquifer has been assumed to be on the order of 10°cm, sec^-1. Its exact thickness has not been determined.
The distribution of groundwater temperature in the region shows a clear arrangement of high and low zones. In winter, the zone of high temperature is situated in the upper reaches of the region and the low temperature zone in the lower reaches. Reverse distribution of the temperature is ob-served in summer (Fig. 3), Each temperature zone moves to the lower reaches (eastward), chang-ing its temperature. The movement of the temperature zone is not inconsistent with the distri-bution of groundwater level. The annual variation of temperature almost disappears near the east end of the study area, with a time-lag of about one year. Groundwarter temperature outside the banks of the Yoshino River was estimated from ground temperature at the depth of lm.
The movement of the temperature zone was traced at one- to three-month intervals during 1984 and 1985 (Fig. 4). The movement reflects the groundwater flow, which has been assumed to be 4_??_8m • day^-1 (average: about 5m • day^-1). Since the distance between high and low temperature zones corresponds to the displacement over a half year, the speed of groundwater flow is obtained by di-viding the distance by 180 days. This eatimate gives 3_??_9 m • day^-1 as the groundwater flow.
The electric conductance of water and dissolved oxygen were investigated to trace the water flow. Although decisive results were not obtained, the saturation level of dissolved oxygen in the upper reaches was almost 100% and it decreased toward the downstream.
Direct measurement of groundwater speed was made using salt water as a tracer. Salt water was injected through a tube in a specially drilled well; then the flow was detected by the change in elec-tlic conductance. The calibrated speed of groundwater was 3.12m bull;day^-1 for Station V 1 and 5.14m •day^-1 for Station. V2, and these values were within the range of the estimated ones from the temperture field. Measured flow direction is shown in Fig. G, which seems in accord with the direc-tion of the temperature zone and the distribution of the groundwater level.
The stretch of the phenomenal groundwater temperature taken from three regions is plotted against the permeability of the aquifer. Although the data are not complete and the representation is rather rough, a relationship between stretch and permeability may be expected.
From the above observations, the authors conclude that this phenomenal temperature field is brought about by the wide movement of groundwater in Egawa Region. Groundwater comes from the Yoshino River as submerging water, and moves along the Egawa River at a speed of about 5m per day.

View full abstract

In some years after 1969, 336 wider-area municipal spheres (WAMSs) called hoihishichoson-ken were established as comprehensive planning regions within most prefectures except for those com-prising the areas surrounding metropolises. The purpose of establishing WAMSs was to promote inter-municipal administration, to improve the infrastructure, and to minimize regional disparities in the country as a whole. Similarly, 21 wider-area administrative spheres surrounding metropolis-es (daitoshi-shuhen-hoihi-gyoseihen) (WASSs) were established at the periphery of metropolitan areas in order to solve their common problems. Then, based on a revision of the Medical Care Act in 1985, secondary medical care regions (niji-iryohen) were set up, and the number of beds they required was determined in the regional medical care planning of each prefecture. Although a standard population in secondary medical care regions was not determined, WAMSs were expected to have a standard population of more than 100, 000 people and to correspond to the city regions of their central cities in order that they can operate actively.
In this paper the writer tries to analyze urban tributary areas and daily city regions of central cities according to the analysis of connectivity between central cities and their surrounding munici-palities by using commuting data among municipalities (1985 Population Census of Japan, Vol. 6, Part 1). Instead of the concept of a city region the writer uses the concepts of a “daily city region” corresponding to the commuting areas of a central city and an “urban tributary area” consisting of a wider area. As shown in another paper (Morikawa, 1990), a municipality qualifies as a central city if the workers of at least 3, 000 persons are engaged in sales and service job of the city.
The writer examined the spatial relationships between WAMSs and urban tributary areas and then observed the characteristics of each WAMS. The main results can be summarized as follows:
1. Figures 1 to 3 show the boundaries of urban tributary areas and WAMSs. In Hokkaido there are a number of isolated municipalities with an outcommuting ratio of less than 15 percent, and central cities without commuting areas exist within a group of small-sized cities (see Fig. 1). Out-side Hokkaido, isolated municipalities are distributed largely in mountainous regions and distant islands. In contrast, the commuting areas of three metropolises shown in Figures 4 and 5 expand beyond the boundaries of their own prefectures. Thus, most of the cities in prefectures neighboring the main urban agglomerations, such as those in Saitama (near Tokyo), Nara (near Osaka), etc., connect more strongly with a metropolitan area than with the neighboring cities within their own prefectures. Thus, WAMSs in neighboring prefectures are established without consideration of such commuting areas of metropolises.
2. A total population of 48, 510, 000 people (40.1%) resides in these commuting areas. In their peripheries the phenomenon of step-wise commuting areas is observed; for example, over 5 percent of the workers in Chiba City commute to Tokyo's 23 wards, those in the surrounding municipalities of Chiba City commute to Chiba City, and those in the area surrounding Mobara commute to Mobara City, which is included in the commuting area of Chiba City. Such a phenomenon demon-strates a remarkable regional disparity in working conditions.
3. There are some WAMSs which are abnormal in terms of the standard population and spatial structure: those with two central cities and a large population that can be divided into two WAMSs; those with too large population and a central city; and those with a total population and a central
city of smaller size. The principles for establishing WAMSs are suited to prefectures with decentra-lized regional urban systems that consist of relatively larger cities smaller than their prefectural capitals.

View full abstract