In July 1986, the authors discovered palsas (peaty permafrost mounds) first in Japan. They are located in the bog on the south of Hiragatake (43° 37'N, 142°54'E; 1, 720m a. s. 1.) in the central part of the Daisetsuzan Mountains, central Hokkaido.
In this paper, the authors describe the size, form and internal structure of the palsas in the Daisetsuzan Mountainss on the basis of field investigation carried out in late September 1986 and discuss the palsa development. The areal changes of palsas during the period of 1955-82 are also illustrated, on the basis of air photograph interpretation, in order to relate them to the recent climatic changes.
The palsa bog is located on a broad pass on the andesitic lava plateau and occupies an area of 10.3 ha. The bog is generally covered with peat layer of about im thick. According to the general rate of peat accumulation (0.3mm/y) on the Daisetsuzan Mountains, the initiation of the peat accumulation is estimated ca. 4, 000-3, 000 y. B. P. By the effect of blowing-out of the strong westerly wind in winter, the snow depth on the bog is less than several decimeters, and the ground surface partly exposes. In summer, the westerly wind brings frequently low clouds or fogs onto the pass so that the amount of insolation over the bog is much lesss than in its surroundings. The annual mean air temperature at Hakuun hat (2, 000m a. s. 1.), about 5km north of the palsa bog, was -3.8°C in 1985. These environmental conditions seem to be favorable for the formation of palsas.
Fifteen palsas were identified on air photographs, and one small palsa was found by field investigation (Photo 1, Fig. 2). These palsas, whose plane figure is generally circular or elliptic, are characterized by flat upper surface and relatively steep side-slopes, and are often accompanied with ponds just around them. They range from 4 to 80m in diameter and from 0.2 to lm in height. Most of the palsas are mineral-cored palsas represented by the following three palsas and some of them are apparently aggradation palsas.
previous studies on palsas is proposed also in this bog. According to the concept, Palsa A, Palsa B and Palsa C are thought to be in an initial stage, in a stage soon after the mature stage, and in a collapsing stage of development, respectively. The thermokarst depression with standing water shown in Photo 8 occurred after the disappearance of Palsa D. This corresponds to the last stage of palsa development.
Judging from the description on these three palsas, a peat cover of more than 60cm thick is required for the growth of palsas at present. According to the general rate of peat accumulation (ca. 0.3mm/y) in the Daisetsuzan Mountains, about 2, 000 years are necessary for the peat accumulation of 60cm thick. If the thermal condition for growing palsas is constant, the palsa formation in the studied bog seems to be younger than 2, 000-1, 000 y. B. P.
Air photographs taken during the period of 1955-82 revealed that the total area of palsas in the study area was reduced by about 36% during this period, although a few of them seem to be growing at present. Among these palsas, Palsa D disappeared completely and replaced by a thermokarst depression with water during a decade. The expansion by one and a half times of the thermokarst depression on the east of Palsa C indicates that the palsa decayed abruptly during a decade. According to the record of accumulated air temperature and precipitation in summer during the period of 1951-82 at Asahikawa about 45km WNW of the study area, high air temperature and/or precipitation were recorded when the conspicuous reduction in total area of palsas occurred. Therefore, when the development process of palsas is discussed, not only the collapse in the cyclic development of palsa itself but also the decay of palsas in relation to the change of regional climatic conditions must be considered.

View full abstract

In advanced countries, especially in the United States, a new population change is appearing which deviates from the tendency existing till about the late 1960s. It is a product not only of a change in demographic structure but also of a change in socioeconomic structure due to a high level of technology development. Previously, urbanization stimulated the economic growth of a region; urban growth, high density and heterogeneity of urban cores were basic features of urbanization. Theories and policies of regional development, such as growth center policy and polarization theory, have been considered in light of such a tendency. Therefore, the population turnaround in the United States was a great surprise to most social scientists. As a result since the 1970s a great many studies about the features of population change and its causes have been generated.
The aim of this paper is to clarify the features of population change through the findings of previous research and to reveal the problems with these studies. The principal findings are summarized as follows:
Due to the remarkable population changes which occurred in the United States since about 1970, the situation has already entered a stage of population deconcentration. In a reversal of the net out-migration phenomenon of the 1960s the population increase in the 1970s was recognized not only in adjacent but also in nonadjacent nonmetropolitan counties. Additionally, counterurbanization in the United States is characterized not only by such metropolitan-to-nonmetropolitan migration, but also by regional differentiation of population change due to the revolutionary development of transportation and communication technology. In the South and the West, the population increase occurred in large cities as well as in nonmetropolitan areas, whereas in the Northeast and the North Central the population decreased in the majority of metropolitan areas. In contrast, many West European countries still remain in the phase of decentralization and the main stream of migration continues to occur from central cities to their suburban areas. This tendency is occurring in Japan as well.
Various factors for the generation of counterurbanization in the United States have been considered. Though the technical development of transportation and communication has undoubtedly played a significant role as a foundamental factor of population change, the change of life style for American people cannot be ignored. Due to the rather significant development of the service sector in the 1970s the role of industrial dispersal into nonmetropolitan areas was not given under consideration. According to the analysis of questionnaire surveys, many inmigrants into non- metropolitan areas were people of relatively high social status but it was pointed out from the analysis of statistical data that they were often retirees who require the quality of life.
Recognizing that the phenomenon of counterurbanization may proceed in a similar way in other advanced countries, social scientists have been greatly interested in its trend. It is important that they confirm whether population dipersal is a temporary phenomenon or long-standing trend. Although the population growth in nonadjacent nonmetropolitan areas is a significant factor in understanding counterurbanization, the debate over whether this is a reversal or a continuance of the previous urbanization process is not yet resolved.
As shown in Figure 4, metropolitan-nonmetropolitan migration and the movement of plants and service establishments should be more fully examined in detail within the national, regional and daily urban systems. Additionally, an analysis of migration between metropolitan and nonmetropolitan areas, and between large-scale and small-scale metropolitan areas of the hierarchical structure of the national urban system should be conducted.

View full abstract