Quaternary (geological) maps, on both small and large scale, published in various countries (see references) are reviewed compared with the recently published Quaternay Maps of Japan, which comprise I. Landforms, Geology and Tectonics (1:1, 000, 000, three sheets) and II. Prehistoric Remains and Paleogeography (1:4, 000, 000, one sheet with four maps). These maps were compiled by an editorial committee of the Japan Association for Quaternary Research, and published by the University of Tokyo Press in July 1987. Several general problems concerning Quaternary maps are discussed: (1) arrangement of legends and of time-space diagrams, (2) presentation of overlapping and underlying layers and topographic surfaces, (3) expression of denudational slopes formed during the Quaternary, (4) expression of submarine features, and (5) paleogeographic maps.
The recent progress of stratigraphical studies on the Japanese Quaternary has led to the establishment of standard geologic sections (Pliocene-Holocene) of representative basins in Japan and the correlation of each one with the others. In the course of compiling the Quaternary Maps of Japan, the geologic sections were chronostratigraphically subdivided into 5 units, i. e. Pliocene, Lower Pleistocene, Middle Pleistocene, Upper Pleistocene, and, Holocene, based on the definitions approved and/or recommended internationally. Unsolved problems, however, still remain in correlating marine sequences with non-marine sequences, and especially in correlating the Pliocene-Pleistocene boundary in the former with that of the latter. The Kazusa Group, a marine sequence of the Boso Peninsula in Kanto basin, is regarded as an expected hypostratotype section of the Pliocene-Pleistocene boundary. Precise correlations of the Kazusa Group with the Osaka Group of nonmarine-marine origin and the Yamato Group of nonmarine origin, which are well studied bio- and magneto-stratigraphically, are of prime importance. The relation between marine and nonmarine sequences are examinable within the Kanto and Niigata basins, because of the coexistence of both sequences in these basins. Bio- and magneto-stratigraphical and radiometric studies on the Pliocene and Pleistocene in the Kanto and Niigata basins are expected to be completed in the near future.
The purpose of this paper is to review recent progress and problems in the study of terraces and lowlands in Japan, as revealed during the preparation of Quaternary Maps of Japan (1:1, 000, 000, 1987), published in celebration of the 30th anniversary of Japan Association for Quaternary Research. Special attention is paid to a comparison of recent research with that undertaken about 30 years ago when the Association was established. Mapping of terraces and lowlands has progressed rapidly by means of interpreting aerial photographs, ca. 1:40, 000 or 1:20, 000 (or larger) in scale, which now cover all of the Japanese islands. Distribution of terraces and lowlands shows marked regional and local variation more than expected, mainly reflecting the different nature of Quaternary vertical movements. Extensive terraces have developed in the areas which are underlain by thick Neogene and lower Pleistcene sediments, in other words, the areas that changed from subsidence to uplift some time during the Quaternary. At the same time, there are still great differences in the number and accuracy of the papers which discussed these landforms. With the development of techniques for detailed mapping of terraces, the reconsideration of their origin and significance becomes necessary in some cases; for example, “low relief erosion surfaces” of Noto Peninsula are now considered to be remnants of a flight of marine terraces, and it requires a considerable change of interpretation of paleogeography since the middle Pleistocene. Small remnants of marine terraces have been found along the complicated shoreline of ria coasts such as Sanriku, eastern Shikoku, and others, indicating that the formation of the original configuration of the ria type shoreline should be dated back to the middle or late Pleistcene. One of the important advances in the study of terraces and lowlands over these 30 years is the increased information on radiometric dates of late Quaternary terraces that became available in the 1970s with the use of 14C, U-series and fission-track methods as well as the recently developed ESR method. Dating of terraces in the coastal areas made it possible to establish the sequence of late Quaternary sea level fluctuations and their worldwide correlation. Higher sea levels of ca. 120, 100, 80 and 60ka have been reconstructed in South Kanto, Kikai Island of the northern Ryukyus (with younger high sea level of ca. 40ka), and Hateruma Island of the south Ryukyus, where only three older high sea levels were recognized because of the low uplift rate of this island. Morphostratigraphic correlation of the main terrace, associated with major seal level rise, to the main stage of the last interglacial of ca. 120-125ka has now been confirmed in the areas mentioned above and in some other areas in Japan by different methods, including tephrochronological method. The ages of higher sea levels of the middle Pleistcene are also identified in some areas. Numerous radiocarbon dates of Holocene deposits underlain by coastal lowlands have been obtained since the 1960s and used as essential data for the reconstruction of various relative sea level curves since the maximum lowering of the sea level of ca. 15, 000yBP. However, separation of the tectonic factor from relative sea level curves is still difficult, and determination of the height of paleo sea level is still problematical. Simultaneity of depositional flurial terraces in central Japan, confirmed by widespread dated tephra, strongly suggests that they represent climatic terraces which were formed in the cold period of ca. 2.5-3.0 and 5.0-5.5ka. The correlation of terraces of the inland area with those of the coastal areas is not yet well established, however.
The morphology of the continental shelf around the Japanese Islands was examined. The depth of the shelf edge measured at 455 points is 140m on an average and almost always shallower than 200m. The marginal terrace, 250 to 500m in depth, stretches along the continental shelf off the southwest Japan coast facing the Japan Sea and off the Sanriku coast in northeast Japan. The continental shelf is generally divided into three segments: inner shelf (terrace shallower than about 60m), middle shelf (gentle slope between 60 and 100m), and outer shelf (terrace deeper than about 100m). The features of saddles and caldrons in the Tugaru and Tusima Straits indicate an abrupt change in the currents through these straits after the latest glacial age. Such a change is consistent with the paleo-climatic change in the Japan Sea estimated with the results of core analysis.
Studies on late Quaternary deep-sea tephras are briefly reviewed in oceans and seas of the world and around the Japanese islands. A great number of modern research has been carried out on deep-sea tephra layers especially in the eastern Mediterranean Sea, Atlantic Ocean, Caribbean Sea and Pacific Ocean. Researches focus on the magnitude and type of individual explosive volcanism, the mechanism of deposition of tephra, correlation of macroscopic tephra layers and biostratigraphic dating of tephra, and resulted in disclosing several explosive volcanic eruptions of extraordinarily great magnitude such as the eruption of Campanian Y 5 ash in the Mediterranean, Los Chocoyos ash around Central America, Wairakei ash around New Zealand and Toba ash in the Bengal Bay and Indian Ocean. In deep-sea sediments around the Japanese islands there are numbers of tephra layers, which are characterized and identified by petrographic and chemical methods. In the Sea of Japan six tephra layers have been identified, three of which are alkalic and are the products of major plinian eruptions of two volcanoes situated on and near the Korean Peninsula: Baegdusan and Ulreung-do. The other three are subalkalic and are corelated with large-scale eruptions from caldera volcanoes in Kyushu: Kikai, Aira and Aso. They are recognized not only in the Sea of Japan but also in the East China Sea and northwest Pacific Ocean, together with widespread tephras derived from Ata and Kikai calderas. All of them are co-ignimbrite ash fall deposit, and well dated by radiometric and stratigraphic methods. Palaeooceanography of late Quaternary around Japan are successfully studied with the aid of such widespread and time-marker tephras. There remain several tephras unidentified around Japan. In addition detailed volcanological and interdisciplinary studies of deep-sea tephras will be one of the problems to be studied in near future.
Paleoenvironmental changes in the Japan Sea provide a clue for estimating the absolute value of eustatic sea level change during the last glacial age. Oxygen isotopic curves of planktonic foraminiferal tests in six piston cores from the Japan Sea show an abrupt change at the time just before deposition of AT volcanic ash. This phenomenon was caused by a supply of fresh water into the Japan Sea from the Huang Ho River in China, when sea level had dropped more than 95m, if we can assume that tectonic movement around Cheju Island located at south of the Korea Peninsula was negligible over the last twenty-several thousand years. The sea level at the last glacial maximum period is inferred to be 127±30m shallower than the present level. This value was estimated from the oxygen isotopic curves of benthic foraminiferal tests in two piston cores from the Pacific Ocean, by extrapolating the sea level difference (95m) between the age of the AT ash layer and the present to that between the last glacial maximum period and the present.
The Quaternary age of the Japanese islands is characterized by renewed, intense crustal deformation, whose mode and rate are quite different from those of the preceding late Tertiary. In view of these characteristics, the Quaternary period of the Japanese islands can be labelled as follows: a) an age of increasing topographic relief, b) an age of land emergence, c) an age of block faulting, d) an age of mm/year tectonics, e) an age of compression tectonics, and f) age of plate collision. The crustal movements contain a) intermittent fault movement due to brittle fracture of the crust, and b) elastic and c) non-elastic secular deformations. An archipelago-wide regional emergence in the Quaternary is recognized, which is probably due to regional non-elastic deformation. The basic concepts on crustal movement which are generally accepted are those of“the uniform mode and rate of movement through the Quaternary”and“the uniform mode of the repetition of intermittent (seismic) crustal events.”Recent studies suggest a change in mode and rate of deformatiom in the middle of the Quaternary, which may be partly explained by collisions of the Izu block with Honshu and of northeast Japan with southwest Japan since the middle Quaternary.
Japanese mountains have been elevated at the rate of 0.5-2.0mm/year in the global climatic flactuations during the mid-late Quaternary, and the geomorphic processes acted on the mountain slopes have also been alternated with climatic changes as well as uplift of the mountains. Changes of the morphogenetic environment of the late Quaternary in Japanese mountains are considered by means of palaeoglacier curve (Emiliani, 1978) compared to snow line and timberline (Lower limit of periglaciation) fluctuation curve in the mountainsand superposed the growth rates of the mountains on it (Fig. 1). Judging from fig. 1 the rapidly uplifted high mountains have been successively placed in periglacial environment since the middle Pleistocene, and lee side mountains of the snow bearing winter westeries were glaciated in the last glaciation and probably penultimate glaciation. Snowy, wind ward mountains near Japan Sea coast were possibly glaciated not only in the last two glaciations but also earlier ones. Evidences of penultimate glaciations are found in some mountains mainly in the snowy Japan Sea slope. Intermediate mountains of 1000-2500m (lower than 1500m in Hokkaido) in hight were situated in the periglacial environment during the last glaciation. Vast block slopes were developed under permafrost conditions in the less snow-covered mountain slopes especialy in northen and eastern Hokkaido at the time. On the contray, snowy mountains were dissected by snow-patch erosion and snow avalanche, and debris derived from slopes were transported by rapid streams fed by snow meltwater to the piedmont and alluvial fans developed there. Low mountains less than 1000m in hight have not yet been placed in periglacial environment. Low relief erosion surfaces or initial surfaces of mountains are most widely distributed in these low mountains and secondly in the block covered intermediate mountains and secondly in the block covered intermediate mountains due to feeble fluvial activity. Morphogenetic environments in Japanese mountains during the last glaciation are compiled as Fig. 3.
The lowest snowline altitude of the Last Glacial time was determined by using snowline data from previous studies and the author's observations which cover China, North and South Korea, the USSR, and Japan. The results are shown as a map of snowline altitude distribution over the whole of Eastern Asia, east of the 100°E longitude (Figure 1). Snowline gradient is steep along the margins of the Qinghai Plateau and along the Japan Sea coast of the Japanese islands. The snowline becomes higher toward the inland of the Asian continent, and also toward the inner part of the Pacific Ocean, both of which are occupied by a subtropical high. A relatively low snowline appears along the eastern margin of the Asian continent. This snowline“trough”stretches from Taiwan (Formosa) to southern Hokkaido through the northern part of the Northern Japanese Alps, where the snowline is 200-1, 000m lower than in the surrounding areas. The existence of this“trough”suggests that a relatively high level of winter precipitation was maintained during the Last Glacial as well as at present.
A floral change in Japan during the Quaternary period based on both macroscopic and microscopic fossils is given. Much attention is paid to the past distribution of the same or closely related plant species found in temperate mixed coniferous and deciduous broad-leaved forests between continental eastern Asia and Japan. An attempt is made to illustrate the relative importance of the constituents by sectorial division of the circle for two periods having a defined span of time: one covers the lowermost or the lower part of the Shimosueyoshi period and the other corresponds to the ages of the Aira-Tn (AT) ash during the Tachikawa period.
This article discusses some problems on the Quaternary mammal fauna of the Japanese islands which were raised in the course of the Quaternary Map Project of the Japan Association for Quaternary Research. The characteristics of the present Japanese mammal fauna are enumerated as follows: a rich number of species per area, a high degree of endemism, and a large number of indigeneous forest dwellers in the fauna. Among the many endemic species, the following are important in holding the key to the origin of fauna: Glirulus (Rodentia)-Urotrichus and Dymecodon (Insectivora) of the Honshu block, and Pentalagus (Lagomorpha)-Diplothrix and Tokudaia (Rodentia)-Mayailurus (Carnivora) of the Ryukyu Islands. The origin of the present mammal fauna of the Japanese Islands is assumed to date back to the Late Miocene or Pliocene time. It is also interesting that the Japanese fauna have been consistently composed of humid forest inhabitants since the time of its origin. Next the formation of the fauna is discussed. The first appearance of the extant species was possibly at the later time of the Middle Pleistocene. Therefore, the formation of the present mammal fauna probably began at the same time, 130-200 thousand years before the present. Subsequently, after the high sea level of the Last Interglacial, most of the pre-existing species were replaced by the extant species (with the exception of some peculiar forms) and disappeared before the advent of the final phase of the Last Glacial. Immigrants from the continent to the Honshu block and the Ryukyu Islands have scarcely been recognized since the early Late Pleistocene. Immigration from the south, which was formerly considered, viz of Siva-Malayan and Sino-Malayan forms, now seems unlikely, because it seems appropriate in some parts in explaining the faunal changes of the Japanese Quaternary mammals to adopt the vicariance hypothesis instead of the disperasl hypothesis.
Although archaeological studies always deal with the time-space attributes of artefacts, this by no means implies that the distribution phenomena are easily evaluated. This paper focuses on Late Palaeolithic society as seen in the site distribution pattern, with particular reference to the backed-blade industries of the Japanese archipelago. Research on the distribution system in backed-blade industries can be divided into three stages as follows: 1). 1949 to the first half of the 1960s: emergence of the distribution area of a typologically interrelated group of backed-blades. 2). The second half of the 1960s to the end of the decade: intensive analysis of one site or an industry with re-fitted stone materials, i. e., cores, flakes, and chips. 3). Up to the first half of the 1970s: archaeological sites and geologic sources of obsidian are correlated with geologic source identification by the Fission track method. Discussions about palaeolithic settlements and society have been continued since the second half of the 1970s. This paper discusses and summarizes some important articles concerning palaeolithic societies or group relationships, and also elucidates the four basic distribution levels (A1, A2, A3, and B) for chorological phenomena in the backed-blade industries.