This paper reviews the recent geological-research results on climatic changes and initiation of monsoon climate caused by uplifting of the Himalayas and Tibetan Plateau. All lines of evidence obtained from the Siwalik Group of foreland basin as well as deep sea drilling cores from the Indian Ocean indicate that monsoon was set up in the Late Miocene about 10 to 8Ma. Abrupt increase of sedimentation rate at 10Ma and appearance of flood splay deposits at 9 Ma suggest seasonal flood has begun around 10 to 9Ma. The first appearance of detrital grains of kyanite and gneiss at 9.2 Ma implies metamorphic rocks in the core of the Himalayas exposed on earth surface and rapid denudation has started by that time. Faunal and floral changes at 9.5 Ma and 7.4 Ma demonstrate that vegetation and climate have gradually changed from wet subtropical evergreen forest to dry grassland. Synchronous changes of σ13C and σ18O value of pedigenic carbonate also support the interpretation. The first appearance of diatomaceous mudstone in the core collected from the upwelling area off Oman suggests upwelling triggered by southwest summer monsoon was set up at about 10Ma. Abrupt increase of Globigerina bulloides in 8.5-7.4 Ma and radioralia Colloshaera in 8 Ma indicates southwest monsoon was strengthened and upwelling was active around 8 Ma. Intraplate deformation recorded in deep sea fan sediments has started in between 7.5 and 8 Ma. This fact also indirectly supports rapid uplifting of the Himalayas and Tibetan Plateau. Initiation of north-south normal faulting in the High Himalayas and Tibetan Plateau at 10-5 Ma is another evidence that the Himalayas and Tibetan Plateau may have attained at least 4, 000 m by the end of Late Miocene. In opposition to those opinions, Chinese paleobotanists insist that elevation of the Tibetan Plateau and the Himalayas were lower than 3, 000 m at the end of Late Miocene. They estimate very rapid uplifting after 2 Ma on the basis of fossil leaves and pollen of evergreen oaks from the Late Pliocene beds in the north slope of the Himalayas.
Paleoenvironmental changes during the Quaternary in Tarim Basin or Taklimakan Desert are reviewed according mainly to the results made by present authors. Pollen analyses for a borehole core, taken from the dried-up lake of Lop Nur, made clear that the vegetation of Tarim Basin in the early Pleistocene was the forest-steppe, suggesting comparatively wet condition. Since the middle Pleistocene, desert steppe or desert environments have been dominant. From the geomorphological data, vast amount of clastics were transported into the desert by floods, and eolian activities lead the formation of Draa from 18, 000 to 13, 000 years ago. Fluviation was activated 10, 000 to 9, 000, 6, 000 to 5, 000, and around 2, 000 years ago. Holocene environment is extremely arid, especially in the last 1, 500 years.
The Himalaya is a fold-and-thrust belt in the northern margin of the Indian continental plate, and a product of the collision between India and Asia since the Eocene. The Himalayan orogen consists of several tectonostratigraphic units bounded by diachronous thrusts younging from north to south; namely, the Kangmar Thrust in southern Tibet, the Main Central Thrust, the Main Boundary Thrust, and the Main Frontal Thrust. These thrust faults are splays off from a major mid-crustal decollement named the Main Detachment Fault or the Main Himalayan Thrust. In he Nepal Himalaya, there is also an out-of-sequence thrust, which cuts across the Main Central Thrust, within the Lesser Himalaya. Abundant geological and geomorphological evidence has suggested that the Himalayan uplift began in the Tethys (or Tibetan) Himalaya, which is situated to the north of the present crest line of the Himalaya (the Higher Himalaya); the uplift then shifted southward, and was finally accelerated in the Higher Himalaya since the Pliocene. Such a migration of the uplift through time is attributed to the southward younging of the activity of above thrusts. Nine zircon fission-track ages from the crystalline rocks of the Higher Himalaya and the Main Central Thrust zone in the Kali Gandaki and Modi Khola valleys, central Nepal, show a linear increase from 1.2Ma to 2.3Ma with increasing elevation. An average denudation rate of 0.9mm/y during this period is thus calculated. Assuming a closing temperature of 260°C for fission-tracks in zircon and a paleo-geothermal gradient of 35°C/ km, averagedenudation rates since the Late Pliocene are over 3mm/y. The denudation rates increase as approach the present, and rearch >6mm/y after 1.2Ma. The trend of linear increase of fission-track ages with increasing elevation is not disturbed by the Main Central Thrust; therefore suggesting the cessation of Main Central Thrust activity in the Late Pli ocene and a flexural uplifting since then, not a differential movement along the Main C entral Thrust. Three zircon fission-track ages from granites of the Kathmadu Nappe located south of the out-of-sequence thrust are about 9Ma, and yield an average denudation rate of 0.8mm/y for the period 9-0Ma. The rapid uplift of the Higher Himalaya in central Nepal since the Late Pliocene may have been due to elastic doming of the Lesser Himalayan rocks beneath the Higher and Lesser Himalayas. The doming has resulted from the fault-bend folding of the rocks along the ramp of the out-of-sequence thrust at depth caused by the northward slippage of the Indian crust along the Main Detachment Fault.
Present-day denudation of the Himalaya-Tibetan plateau occurs mainly at its southernmargin (High Himalayan front) by headward erosion of small tributaries of Ganges, Brahmaputra, and Indus. At this erosion front, large unroofing more than severalkilometers, which may exceed annealing depths of fission-tracks in apatite and zircon, couldoccur in a short period of time. This would result in dramatically accelerated erosion rates, as have been deduced from cooling age determinations of rocks from near the erosion frontand have often been misinterpreted. Observations of present-day river load of Ganges andBrahmaputra and the sediment volume of the Bengal deep -sea fan and the Ganga basin haveindicated that denudation rate of the Himalayas has been fairly uniform (1.0±0.3mm/yr) since Early Miocene time up to the present. This suggests that the topographic relief of theplateau's front has essentially been constant during the past 17-20Myr. Simple evolutionarymodels of the Himalaya-Tibetan plateau, considering both denudation and crustal thickeningdue to plate convergence, suggest that the plateau had grown as high as the present height (-5km) only 15-30Myr after the onset of collision, without reaching the dynamicequilibrium between denudation and uplift. This implies that the height of the plateau iscontrolled by crustal strength, which is likely to be equal to the gravity-induced stress (-1.4kbar) due to -5km relief. Plate convergence since Early Miocene time has been accommodatednot by crustal thickening but by expansion of plateau's width andsubsequently bylateral (east-ward) escape of the Tibetan plateau.
Map reading and field observation of the glaciers around Mt. Kongur, West Kunkun, Xinjiang Provice, Western China revealed that the present ELA (Equilibrium Line Altitude) of the glaciers is much lower in the side of Tarim Basin (about 4, 200-4, 300m) than in the Tibetan Plateau side (about 4, 750-5, 200m). The Last Glacial ELA reconstructed by the AAR method indicates a similar steep gradient of ELA (about 4, 050-4, 200m in Tarim side, about 4, 600-4, 950m in Tibetan side). ELA of the penulitimate glaciation (just before the Last Interglacial) shows a similar situation (about 3, 500-3, 700m in Tarim side, about 4, 400-4, 700m in Tibetan side). Depression of ELA was much larger in the penultimate glaciation (about 350-700m) than in the Last glaciation (about 100-120m). Depression of ELA was larger in Tarim side than in Tibetan side, especially in the penultimate glaciation. The age control of the moraines was attempted by the carbon and AMS datings for the Little Ice Age and Late Glacial moraines. Measurement of the magnetic susceptibility of loess covering the outwash terrace surface which connects topographically to the moraines upstream was used for the age determination of older moraines. The result supports the idea that the Tibetan Plateau was not covered by a single and uniform large ice cap even in the penultimate glaciation.
In the Himalayas-Tibet, glaciers varied much due to uplift of the area and variation of the monsoon as well as general climatic change since Quaternary ice ages. The variation of glacier mass balance is described qualitatively on the basis of present characteristics of summer-accumulation type glaciers. Variation of such glaciers is sensitive to that of summer temperature. Transition of annual variation patterns of the glacier mass balance is shown for the period from the last ice age to the present. Increase of ablation due to decrease of the surface albedo is considered to be effective in shrinking of the glaciers during this period. It is pointed out that development of methods to reconstruct the paleo-water-cycle under changing influence of the monsoon is important to clarify the history of areal characteristics of glacier accumulation.
Loess deposits spread widely in central and East Asia, where the main area of loess deposition is on the Chinese Loess Plateau. Climatic conditions in central and east Asia allowed a substantial loess accumulation since 2, 400ka., while in South and Southeast Asia, the presence of loess is recognized locally in a periphery of Thar Desert and on the terraces of South India and Thailand. A weakened monsoon during glacial phases seems to diminish a loess accumulation in these regions. Most areas of East Asia such as the Japanese Islands, Korea and Taiwan were affected the depositionof eolian dust originated in the drier area of Asian continent, especially during the Last Glacial time. A depositional rate of loess and eolian dust in the glacial period was higher than in the interglacial. Fine silt production in deserts and glacierized areas seems to have increased in glacial period. On the Chinese Loess Plateau and western Japan, a depositional rate of loess and dust in MIS 2 was 1.3 times higher than in MIS 4. On the contrary, in northern Japan, a dust deposition rate of MIS 2 was 2.5 times higher than in MIS 4. The difference of rates between the western and the northern Japan seems to reflect a paleoenvironmental change of the source areas of eolian dust: a desert condition might have increased more in the northern Asia from which the eolian dust mainly came to the northern Japan, because of a drier condition in winter by a decrease of snowfall. ESR intensity signal of fine quartz (finer than 20μm) during MIS 2 in the Japanese Islands varies in different localities. Under the assumption that the values of ESR intensity signal of eolian fine quartz reflect those of the quartz at the source area of eolian dust, the source area and trajectory of eolian dust are tentatively reconstructed. The northern Japan where the ESR intensity signal is high (10-12), the eolian dust seems to be supplied from Siberia and Mongolia where the Precambrian rocks with a high ESR intensity signal are widely exposed. The central and southern Japan where the ESR intensity signal is medium (5.8-8.7), the eolian dust seems to be supplied from central Asia where the Paleozoic-Mesozoic rocks with a medium ESR intensity signal are widely exposed. The southernmost islands of Japan where the ESR intensity signal increases again, seem to have an eolian dust supply from South China and India where the Precambrian rocks are widely exposed. On the basis of ESR intensity signals, three major courses of eolian dust transport in MIS 2 are proposed: the winter monsoon in the northern Japan, the summer subtropical jet in the central and southern Japan, and the winter subtropical jet in the southernmost islands of Japan. The uplift of the Tibetan Plateau and Himalayan Range seems to enhance a production and supply of Asian loess, by the intensification of Siberian high and Asian monsoon system. Glacier extension on the Tibetan Plateau and Himalayan Range, which is related to the uplift of these areas, also enhanced the production of loess material in the source area. The detail study which correlates the glaciation with the loess deposition in central Asia is much more required.
We attempted to clarify changes of magnetic susceptibility and clay mineralogy of the deep-sea sediments in ODP site 974B during the last 4.5 Ma. These sediments are composed mainly of calcareous nannoplankton ooze with organic matters and eolian dusts. Sequential changes of illite crystallinity, expressing as full width of half maximun of 10 A peaks by X-ray diffraction patterns, indicate changes of dry-wet climatic conditions in a supply area. Due to the atmospheric circulation pattern influenced by the Himalayan -Tibetan Plateau, the tropical easterly jets have lead northern Africa to dry conditions. So, it is possible to settle the uplift age of the Himalayan-Tibetan Plateau from the beginning age of dry climatic condition in northern Africa. As a results, we could clarify two evidences as follows: 1) Illite concentrations and magnetic susceptibility indicate concentrations of eolia n dust carrying from northern Africa have increased from 3.5 Ma ago. In other words, the recent atmospheric circulation, that is carrying eolian dust into the Mediterranean Sea, has been confirmed from 3.5 Ma ago. 2) Based on oscillations of illite crystallinity during the 4.5 Ma, the northern Africa region became dry condition 0.8 Ma ago. We thought that this event of 0.8 Ma was caused by the uplift of the Himalayan-Tibetan Plateau.
The deep sea sediments off Oman contain records of the history of the Southwest Monsoon, as seen from lithology, paleontology, isotopes, geochemistry and magnetostratigraphy. Glacial-interglacial fluctuations in upwelling related to the Southwest Monsoon can be traced back 250ka. In contrast, from 450 to 250ka, carbon isotope differences indicating upwelling strength did not show such distinct cyclicity. Upwelling was strong during interglacial stages and had less fluctuation, and was weak during glacial stages with large fluctuation. This upwelling and climate can be traced back to the late Miocene. The strongest upwelling is estimated to have occurred in the Pliocene-Pleistocene time, based on calcareous nannofossil assemblages, isotopic indicators, and organic carbon content. During glacial stages several factors suggest that the Arabian Peninsula was humid, including a very high calculated rate of sedimentation based on oxygen isotope stratigraphy, several to ten times that of interglacial stages, and an increase in the absolute flux of fluvial sediments and variability of lithofacies. The Arabian Peninsula was separated by high mountains along its southwest margin from east Africa, which was arid during glacial stages.
Recent atmospheric circulations over the Asian Continent have been influenced by the Himalayan Range and the Tibetan Plateau as a wind barrier. Based on recent meteorological data, the existence of the Himalaya-Tibet within the troposphere has caused meandering of the Westerlies, intense activity of the Asian Monsoon and appearance of the Easterlies. Meandering of the Westerlies has led to increase precipitation in the East Asia region including northeastern China and Japan, and has formed the soils with poorly crystalized illite minerals in the Chinese Loess Plateau and the Japanese Islands. In nothern India and Nepal including the frontal basin of the Himalayan Range such as the Kathmandu Basin, intense activity of the Asian Monsoon has led abruptly to increase precipitation named “onset of monsoon”. Also, appearance of the Easterlies from the Himalaya-Tibet to northern Africa within the upper part of the troposphere has led to dry and spread the arid area in the northern Africa. There are many sedimentary records indicating Late Quaternary climatic changes around the Himalayan-Tibetan Plateau. We expected to determine the formation age of recent atmospheric circulation. This age indicates when the Himalayan-Tibetan Plateau was abruptly uplifted and elevated. In other words, formation ages of more humid climate in northeastern China, the Japanese Islands and the Kathmandu Basin, and the driest climate in the northern Africa are useful to determine age of uplift and elevation of the Himalayan-Tibetan Plateau. On the base of above-mentioned evidences and hypothesis, we attempted to investigate magnetic susceptibility of the loess-paleosol sequence in the Luochuan area of th e northeastern China, and illite crystallinity of loess-paleosol sequence in the Kathmandu Basi n and deep-sea sediments of the Japan Sea and the Mediterrannean Sea. Because characteristic changes of magnetic susceptibility and illite crystallinity indicate dry -wet oscillations as useful detectors. As results, we clarified three evidences as follows: 1) Large amplitudes of magnetic susceptibility oscillations by large pricipitation in interglacial periods appeared in the Luochuan loess -paleosol sequence during last 600ka. 2) Large amplitudes of illite crystallinity oscillations by large precipitations caused decomposition of illite minerals appeared in the deep-sea sediments of the Sea of Japan during the last 800ka and in the loess-paleosol sequence of the Kathmandu Basi n during the last 1.1Ma. 3) Amplitudes of illite crystallinity oscillation during the last 800k a were smaller than those amplitudes since 800ka ago. Investigation mentioned above indicates that critical uplift and elevations of the Himalayan-Tibetan Plateau achieved to induce activity of westerlies, easterlies and the Asian Monsoon, and suggests that determined ages of critical elevation caused atmospheric circulation changes range from 0.6 million years to 1.1 million years ago.
Activities of Asian monsoon and westerlies during the last 2.4Ma were determined through the investigation of sedimentary features in the Chinese loess-paleosol sequences and the deep-sea sediments of the Sea of Japan. It was already known that sequential changes of magnetic susceptibility in the loess-paleosol sequences of the Chinese Loess Plateau were closely concerned with the glacial-interglacial cycles. Also, sequential changes of illite crystallinity, expressing as full width of half maximum of 10 A peaks by X-ray diffraction patterns in late Quaternary fine-grained sediments at Site 795A of Ocean Drilling Program in northeastern part of the Sea of Japan, were well correlated to standard records of oxygen isotope changes by SPECMAP. The samples of this research were derived from the Luochuan of northwestern China and from deep-sea sediments in ODP site 795A. This research is dealing with the changes of magnetic susceptibility and/or clay mineralogy of the loess-paleosol sequences during the last 2.4 Ma. The results are summarized as follows: 1) Illite is one of the main detrital minerals in the loess-paleosol sequences, and illite concentrations in loess are higher than those concentrations found in paleosol. Therefore, illite crystallinity and concentrations are useful indicators to detect concentrations of eolian dust deriving from the Chinese Loess Plateau. 2) During interglacial periods, wet atmospheres drifted from the Eastern China Sea into the inland area of the Asian Continent. This increase of moisture has promoted the hydration and decomposition of illite crystals in paleosols. 3) According to the investigation of sedimentary features, it is obvious that there was little supply of eolian dust origin minerals transported from the arid area of western China to the Chinese Loess Plateau. This fact cannot always support the formation process of the loess-paleosol sequences, as it was interpreted by Kukla etal.(1988). 4) The large amplitudes of Magnetic susceptibility in the loess-paleosol sequences during the last 0.6 Ma and there was a high degree of illite crystallinity in the deep-sea sediments during the last 0.8 Ma appeared. These facts suggest that changes of atmospheric circulations were caused by the uplift of the Himalayan/Tibetan Plateau about 0.6-0.8 Ma ago.
Interannual variations of summertime precipitation over arid and semi-arid regions in China and Mongolia are investigated. The correlation between variations in summer precipitation in these regions and all-India monthly rainfall, and atmospheric circulation are then examined. The interannual variation of summer precipitation of Taklimakan Desert is mainly related to the windward mid-latitude circulation. A clear negative correlation between the interannual variation of precipitation of this region and all-India monsoon rainfall is seen in June and July. This relationship is caused by a rather local atmospheric circulation change seen over Central Asia. Water vapor transport around the arid region of the Interior of the Eurasian Continent was then investigated using the re-analysis objective analysis data provided by the European Centre for Medium range Weather Forecast (ECMWF). Mongolia and northern China receive water vapor from the northwest through the vertically integrated water vapor transporting fields of the mean summer. One water source for these regions is located in Western Siberia. In the lower troposphere, most of the water vapor is transported to the Taklimakan Desert from the northwest along the eastern periphery of Tianshan Mountains in the mean summer state. By compositing the water vapor transport pattern of heavy precipitation, we found that the southerly water vapor routes pass over the Tibetan Plateau and along the eastern periphery of the Plateau in the lower troposphere. This water vapor flux pattern is related with the southwestward extending trough located to the north of this region and the ridge located in Central Asia. Because of this pattern, more water vapor comes from the south in a wet year (e.g., 1981) than normal years.
Effects of mountains on climate are investigated by an atmospheric general circulation model coupled with a 50-m depth slab ocean model. A model simulation with the present-day orography reproduces most of the principal features of the observed climate. Another simulation without mountains (NM) is done and is compared with the one with mountains (M). Orography-originated stationary waves are separated from those induced by the existence of land-sea distribution. In northern winter, about two thirds of the stationary wave is explained by orography in this study. The wintertime subtropical jet in the upper troposphere in NM is more zonal and weaker than in M. The summertime subtropical jet in NM does not shift northward much and is located over China and Japan. The Asian summer monsoon in NM is substantially weaker than in M. Surface monsoon westerlies are weak, a rain belt stays south and the precipitation over India and China is less in the model without orography. On the other hand, the precipitation in the continental interior is large and the ground is wet in NM due to large moisture transport from the ocean. The global mean sea surface temperature in NM is warmer than in M. This occurs mainly from increased solar radiation into the ocean due to decreased lower tropospheric clouds over the subtropical eastern Pacific. Land surface temperature in NM is warmer than in M due to lapse-rate effect. However, the summertime surface temperature in Europe, Russia and Canada is cooler in NM than in M because ground surface in NM is wetter.