Chinese Science Bulletin Editorial Board Honorary Editor General: ZHOU Guangzhao (Zhou Guang Zhao) Editor General: ZHU Zuoyan Editor-in-Chief: XIA Jianbai Executive Editors: Physics CHEN Dongmin CHEN Hong LONG Guilu Earth Sciences GUO Zhengtang NIU Yaoling ZHENG YongfeiChemistry BAO Xinhe JIANG Guibin XI Zhenfeng Technological Sciences LU Gaoqing TAO Wenquan XU NingshengLife Sciences WU JiaRui WU WeiHua ZHAN QiMin Information Sciences ZHANG Ping ZHENG Nanning ZHOU Zhihua Associate Editors: Physics, Mechanics and Astronomy

CHEN Nuofu CHEN Yanfeng DING Dajun FAN Jing FANG Jing GONG Xingao GUO Guangcan HOU Meiying HU An HU Gengkai HUANG Tao JI Peiwen JIANG Zuimin JIN Changqing KANG Yilan LI Ruxin LI Yadong LI Youquan LI Zhiyuan LIAO Xinhao LING Guocan LIU Jixing LÜ Rong LU Wei LUO Qingming MA Yuanliang PENG Lianmao QIAO Guojun SHAO Zhifeng SUN Yang TONG Dianmin WANG Jiyang WANG Pengye WANG Wei WANG Weihua WANG Yupeng WEI Baowen WEI Jianyan WU Dejin WU Ying WU Yueliang XIAO Min XU Furong XU Hongxing XUE Ping ZENG Heping ZHANG Hailan ZHANG Tiancai ZHANG Tianjue ZHANG Weiyi ZHANG Zhigang ZHAO Zhentang ZHENG Dongning ZHU Jinsong Chemistry

BO Zhishan BU Xianhe CAI Zongwei CHE Shunai CHEN Jingwen CHEN Zhongning DENG Feng ZHANG Deqing DU Walter GAN Liangbing GAO Changyou GAO Mingyuan GONG Liuzhu GUO Lianghong GUO Zijian HAN Buxing HAN Keli HAN Yanchun JIN Guoxin LI Xiangdong LI Yan LI Yanmei LIANG Wenping LIU Chunzhao LIU Yu LIU Zhengping LIU Zhenyu LIU Zhipan MA Dawei MA Huimin MAO Bingwei PANG Dai-Wen QI Limin SHAO Yuanhua SHEN Wenjie SU Chengyong SU Hongmei SU Zhongmin TANG Yi WAN Xinhua WANG Guanwu WANG Qiuquan WANG Ye WU Kai WU Lizhu XI Zhen XIA Xinghua YANG Bai YANG Jinlong TANG Yong YU Hanqing YU Jihong YU Shuhong ZHOU Bingsheng ZHU Benzhan ZHU Lizhong ZOU Hanfa

Life Sciences

CHEN Guoqiang CHEN Yan CHENG Zhukuan CHONG Kang FAN Silu FANG Shengguo FU Xiangdong FU Xiaolan GONG Zhizhong GUAN Youfei HAN Zeguang HU Zhihong HUANG Dafang HUANG Li LIN Jinxing LIU Zhihua MA Dalong PEI Duanqing QIN Song QIU Xiaobo QU Chunfeng QU Lianghu QU LiJia SHA Jiahao SHAO Rongguang SHI Suhua SUN Mengxiang TAN Renxiang TIAN Zhigang WAN Jianmin WAN Shiqiang WANG Gang WANG Jianwei WANG Yiping WU Mian WU Weiren XU Tao XU Xudong XUE Yongbiao YANG Hongquan YU Jialin ZHANG ChuanmaoZHANG Lixin ZHANG Xu ZHANG Xue ZHANG Xuemin ZHANG Yaping ZHONG Yang ZHOU Pingkun ZUO Jianru

Earth Sciences

CHAI Yuchen CHEN Fahu CHEN Hongbin DONG Yunshe FANG Xiaomin FENG Xueshang FU Suiyan GAO Shan GAO Shu GAO Xing GENG Ansong JIAN Zhimin JIANG Shaoyong LI Jiancheng LI Shenghua LI Xia LI Xianhua LI Yan LIU Kambiu LIU Libo LIU Xiaodong LÜ Houyuan LU Huayu LU Quanming LUO Yongming MA Jianzhong MENG Jin PAN Yongxin PENG Pingan SHAO Xuemei SHEN Yanan WANG Dongxiao WANG Huijun WANG Rucheng WANG Yang WU Dexing WU Fuyuan XIAO Jüle XIAO Wenjiao XIE Shucheng XU Yigang YE Kai YU Zicheng YUAN Xunlai ZHANG Hongfu ZHANG Lifei ZHANG Peizhen ZHANG Renhe ZHANG Xiaoye ZHANG Zhongjie ZHAO Dapeng ZHAO Guochun ZHENG Hongbo ZHENG Jianping ZHOU Guangsheng ZHOU Liping ZHOU Zhonghe ZHU Maoyan

Technological Sciences

CHEN Xianhui CHENG Huiming DING Han DING Jiandong HE Yaling HU Xijun JIN Hongguang LI Yanrong LI Yongxiang LU Zhaoping LUO Jianbin NAN Cewen NI Jinren PENG Xiaofeng SHI Jianlin SU Wanhua SUN Hongbo SUN Jun SUN Yuanzhang WANG Ruzhu (R.Z. Wang) WANG Zhonglin WEI Bingbo XU Chunxiang YANG Guowei YAO Kefu YU Aibing YU Dapeng YU Xiping ZHANG Di ZHANG Yinping ZHANG Zhefeng ZHENG Chuguang ZHU Xianfang

Information Sciences

FENG Dengguo GAO Jinghuai HAN Wenbao HONG Wei HU Dewen HU Zhanyi JIANG Xiaoyi LI Cuihua LI Shaoqian LIU Min LIU Yuliang LIU Zhiyong LU Jian LU Zuhong MEI Hong NI Lionel M. TAN Min WAN Mingxi WANG Feiyue WANG Huaimin WANG Jing WANG Jue WANG Xiaoyun XU Yangsheng YAO Xin YIN Qinye ZHANG Rong ZHANG Xing ZHANG Xuegong ZHAO Jizhong ZHU Ninghua ZHU Shihua ZHUANG Yueting ISSN 1001-6538 CHINESE SCIENCE BULLETIN CN 11-1785/N

邮发代号: 80-214 (英)

Supported by NSFC

Chinese Science Bulletin

© 2009 SCIENCE IN CHINA PRESS

Springer NEW

S &

VIE

WS

Monsoon and general circulation system

BERGER André

Institut d'Astronomie et de Géophysique G. Lemaître, Université Catholique de Louvain, Chemin du Cyclotron 2, 1348 Louvain-la- Neuve, Belgium

The instrumental period of meteorological record is too short to capture the full variability of the climate system and, in particular to visualize the type of climate that is predicted to occur over the next decades and centuries. It is therefore important to reconstruct past climates and understand past climatic variations. Among them, those related to the monsoon appear to be more and more im-portant for Society. The economy, culture and rhythms of life of a large part of humanity are critically influenced by the evolution and variability of the monsoon. In particu-lar, the Asian monsoon is one of the most spectacular oc-currences in the climate system, strongly influenced by the combination of the thermal contrast between the Eurasian continent and the Indo-Pacific Ocean and the presence the Tibetan Plateau. Climatologically, the mon-soon regions as the most convectively active areas, are situated in the Intertropical Convergence Zone and ac-count for the majority of global atmospheric heat and moisture transport[1]. The need to better understand the monsoon leads inevitably to the close inspection of its worldwide activity during the geological times to provide a long-term perspective from which any future change may be more effectively assessed.

It is in that framework that the substantial paper by Wang Pinxian in this issue[2] is more than welcome. It is an authoritative review, and probably also the first in which the monsoon issues are reviewed in a global scale through a so long geological history. Wang’s main points are fundamental, and hence important. First, it is to em-phasize the role of the tropics as an extremely important source of changes in the global climate system. Second, it is to regard the monsoons as a global feature through which the tropical forcing exerts its climatic role world-wide. I totally agree with these arguments, and share

Wang’s concern that the climate and paleoclimate com-munity has, since too long, viewed the monsoons as lo-cal phenomena, without trying enough to relate their behavior together. But the problem being extremely complex, let me take the opportunity to put it in its broad context.

The reader will find an interesting brief introduction to monsoon, its origin, history and localization in http://en.wikipedia.org/wiki/Monsoon. Etymologically, monsoon comes from the Arabic word mawsim, which means season, explaining why the original definition includes only major wind systems that change direction seasonally. The most spectacular occurrence of such seasonally varying circulation is undoubtedly in South-ern and Eastern Asia. It is the associated heavy rainfall there, which has allowed by extension the other regions of the world to qualify as monsoon regions. With time, the term monsoon has been broadened to include almost all of the phenomena associated with the seasonal weather cycle within tropical and subtropical land re-gions of the Earth, showing immediately the relevance of Wang’s argumentation to deal with monsoon as a global tropical system.

Giving stronger emphasis to the tropical forcing does not mean that we can isolate the tropical climate from the rest of the world, as it is part of the Earth’s climate system. The response of the climate system to the energy that we receive from the Sun (both at the seasonal and geological time scales) involves all latitudes and seasons and depends upon the feedback mechanisms inherent in the climate system itself. Among them, the water vapor Received February 11, 2009; accepted February 12, 2009 doi: 10.1007/s11434-009-0170-y email: andre.berger@uclouvain.be

www.scichina.com | csb.scichina.com | www.springerlink.com Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1111-1112

feedback is particularly efficient in the intertropical zone which, in addition, receives half of the energy sent by the Sun. It therefore can certainly not be ignored in the discussion. The other feedback is the albedo-temperature one, which is significant in the high latitudes, covered by snow and ice. The amplitude of its seasonal cycle is so large―mainly in the Northern Hemisphere― that it can not either be ignored, especially in problems where seasonality is important, like in the atmospheric general circulation. An example of the complex, rather unex-pected, is the possible relationship between high latitude ice sheets and precipitation over East China[3]. The sur-face albedo in the polar latitudes influences indeed the magnitude and seasonal behavior of the latitudinal gra-dient in net radiation which, with the rotation of the Earth, drives the three-cell regime observed in the mean meridional circulation[1]. The Hadley cells tend therefore to fluctuate in intensity and locations with the seasons, being strongest in the winter hemisphere, and so does the ITCZ, a band of organized convection reflecting the convergence of surface air from the two hemispheres inside the Hadley circulations. This ITCZ, which mi-grates north and south annually with the Sun and in-volves the monsoon particularly over southeastern Asia and northern Australia during the solstices, is therefore strongly linked to the high latitudes and not only to the tropics. Finally, this movement of the ITCZ is also asso-ciated to the largest seasonal pressure variations which are found over the Asian continent where a strong anti-cyclone develops over Siberia during winter and a low-pressure system forms during summer north of the Indian subcontinent. This implies that the longitudinal gradient of pressure (and temperature) between the Asian continent and the Pacific and Indian oceans can-not be ignored either.

As far as the astronomical theory is concerned, I share Wang’s view[2] that the astronomical signal in paleocli-mate data is very helpful for better understanding the

physical mechanisms driving the historical behavior of the monsoon. Looking for the astronomical frequencies in proxy records is certainly the first knowledgeable step. Their list is available in Berger (1978)[4] who was the first to calculate them, in Berger et al. (2005)[5] where more details can be found about the origin of the 100 ka cycle and in Berger et al.[6] where precession harmonics are shown to exist in equatorial insolation. The next step is then to analyze the relationship between the insolation forcing and the climate system response. Here, we must also be very careful in choosing the insolation values. As shown in Berger and Pestiaux (1984)[7], the total energy received during a season is only a function of obliquity whereas the length of the seasons is only a function of precession. Combination of both signals characterizes the behavior of the average seasonal insolation and of the daily insolation in general. This is important to con-sider because the time behavior of these different insola-tion parameters differs also significantly from one to the others[8].

Beside the global scope, one of the highlights of Wang’s review[2] is the attempt to trace back to the deeper geological past. With the pre-Quaternary records of climate, it is effectively possible to address the long-term orbital cycles, and to examine the monsoon behaviors at the time when the North or both poles were ice-free. Although the data limitation currently precludes firm conclusions, it is worthy noting Wang’s idea on if the long eccentricity signals in the climate system were related to long-term monsoon cycles.

There would be a long way to go for the science community to understand the interactions between the tropical forcing and high-latitude forcing. Towards this end, I totally agree with Wang’s argumentation about paying more attention to the importance of the tropical forcing in modulating the Earth’s climate system. It is rightly this kind of efforts that are beguiling mixture of coherence and debate to keep us all on our toes.

1 Peixioto J P, Oort A H, Physics of Climate. New York: American In-

stitute of Physics, 1993. 520 2 Wang P X. Global monsoon in a geological perspective. Chin Sci

Bull, 2009, 54(7): 1113―1136 3 Yin Q, Berger A, Driesschaert E, et al. The Eurasian ice sheet rein-

forces the East Asian summer monsoon during the interglacial 500000 years ago. Clim Past, 2008, 4: 79―90

4 Berger A. Long-term variations of daily insolation and Quaternary climatic changes. J Atmos Sci, 1978, 35(12): 2362―2367[DOI]

5 Berger A, Mélice J L, Loutre M F. On the origin of the 100-kyr cy-

cles in the astronomical forcing. Paleoceanography, 2005, 20, PA4019, DOI: 10.1029/2005PA001173[DOI]

6 Berger A, Loutre M F, Mélice J L, et al. Equatorial insolation: from precession harmonics to eccentricity frequencies. Clim Past, 2, 2006. 131―136

7 Berger A, Pestiaux P. Accuracy and stability of the Quaternary insolation. In: Milankovitch and Climate. Berger A, Imbrie J, Hays J, et al. NATO ASI Series C, Vol. 126, Dordrecht: D. Reidel Publishing Company, 1984

8 Berger A, Loutre M F, Tricot Ch. Insolation and Earth's orbital peri-ods. J Geophys Res, 1993, 98(D6): 10341―10362[DOI]

1112 BERGER André Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1111-1112

Chinese Science Bulletin

© 2009 SCIENCE IN CHINA PRESS

Springer

www.scichina.com | csb.scichina.com | www.springerlink.com Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

FRO

NTI

ERS

G

EO

LOG

Y

Global monsoon in a geological perspective

WANG PinXian State Key Laboratory of Marine Geology, Tongji University, Shanghai 200092, China

Monsoon is now considered as a global system rather than regional phenomena only. For over 300 years, monsoon has been viewed as a gigantic land-sea breeze, but now satellite and conventional observations support an alternative hypothesis which considers monsoon as a manifestation of sea-sonal migration of the intertropical convergence zone (ITCZ) and, hence, a climate system of the global scale. As a low-latitude climate system, monsoon exists over all continents but Antarctica, and through all the geological history at least since the Phenorozoic. The time is ripe for systematical studies of monsoon variations in space and time.

As evidenced by the geological records, the global monsoon is controlled by the Wilson cycle on the tectonic time scale (106―108 a). A “Mega-continent” produces “Mega-monsoon”, and its breakdown leads to weakening of the monsoon intensity. On the time scales of 104―105 a, the global monsoon displays the precessional cycles of ~20 ka and eccentricity cycles of 100- and 400-ka, i.e. the orbital cycles. On the time scales of 103 a and below, the global monsoon intensity is modulated by solar cy-cles and other factors. The cyclicity of global monsoon represents one of the fundamental factors re-sponsible for variations in the Earth surface system as well as for the environmental changes of the human society. The 400-ka long eccentricity cycles of the global monsoon is likened to “heartbeat” of the Earth system, and the precession cycle of the global monsoon was responsible for the collapse of several Asian and African ancient cultures at ~4000 years ago, whereas the Solar cycles led to the de-mise of the Maya civilization about a thousand years ago. Therefore, paleoclimatology should be fo-cused not only on the high-latitude processes centered at ice cap variations, but also on the low-latitude processes such as monsoons, as the latter are much more common in the geological his-tory compared to the glaciations.

monsoon, ITCZ, low-latitude processes, orbital forcing, Wilson cycle

Although scientific attention to monsoons can be traced back to nearly 350 years ago, only recently have mon-soons been recognized as a global system. The “global monsoon” system embraces monsoons in all continents but Antarctica, and this new concept denotes a new stage of monsoon studies on a global scale. The recent devel-opment of the “global monsoon” concept is driven by the application of remote sensing and other new tech-niques on one hand, and by the urgent need to improve our understanding of the global hydrological cycle on the other. As a major part of climate disasters are related to hydrological cycles, the monsoon variations have at-tracted attention of climatologists in all tropical and sub-

tropical regions. In result, the classical definition of monsoon has been challenged, and the concept of “global monsoon” is introduced as a global-scale sea-sonal overturning of the atmosphere and the associated seasonal contrast in precipitation[1].

However, the new trend in modern climatology has not yet been responded to by the paleo-community. The paleo-monsoon is still being studied as regional phe-nomena. According to the prevailing notion in paleocli- Received December 15, 2008; accepted January 15, 2009 doi: 10.1007/s11434-009-0169-4 email: pxwang@online.sh.cn; pxwang@tongji.edu.cn Supported by National Basic Research Program of China (Grant No. 2007CB815902)

1114 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

matology, the climate changes should originate from the high-latitude ice-cap, followed by the low-latitude proc-esses, hence paleo-monsoons are studied just as a sup-plement to the ice-sheet history. But the conventional wisdom is challenged by the new discoveries based on new techniques. Why the low-latitude processes have to follow the high-latitude ones, if the tropical climate change was found to precede the change in boreal ice-sheet, and prominent climate changes were discov-ered in the “Hot-House” world before the formation of polar ice-sheets? Why the low-latitude processes, such as monsoon and ENSO, could not directly respond to the orbital forcing, without transmission through the ice-sheet? It is believed, the “global monsoon” and low-latitude processes in general will become the new focal points in paleoclimatology, calling for approaches to paleo-monsoon in a global perspective. Paleoclima-tology is not only to study polar ice-sheets with the temperature changes, but also the low-latitude processes including monsoons and the associated hydrological cycles with floods and droughts[2].

The present paper is an attempt to review the mon-soon variations in space and time, on the basis of avail-able data. It begins with monsoons in the modern world, followed by the Holocene, Pleistocene and Cenozoic, focusing on the global monsoon response to the Earth’s orbital forcing. Then we will discuss monsoon history on the tectonic time scale driven by plate tectonics, and in the last part some suggestions are made for studies of long-term changes in the global monsoon.

1 Global monsoon concept and its geo-logical implications

1.1 Global monsoon concept

The monsoon has been known since the ancient times as the seasonal reversal in the direction of the near-surface wind. As early as in 1686, Halley interpreted the mon-soon as a gigantic land-sea breeze, driven by differential solar heating between sea and land, and the seasonal reversal in wind direction is central to this model. Now with the development in technology, both satellite and conventional observations support a new concept, which considers the monsoon as a manifestation of seasonal migration of the intertropical convergence zone (ITCZ)[3]. Since the ITCZ is a circum-global feature, the monsoon is also a global system and occurs in all conti-nents but Antarctica.

Also known as the “climatic equator”, the ITCZ is a divide of atmospheric circulations of the northern and southern hemispheres (Figure 1B). There are two kinds of ITCZ in the modern world: In the Atlantic and the eastern Pacific, the ITCZ is located always north of the equator with only minor seasonal shifts, the trade winds of the two hemispheres converge here and produce only limited amount of convective clouds; in the western Pa-cific and Indian Ocean the ITCZ migrates seasonally far away from the equator, and the trades of the two hemi-spheres meet here to produce intensive tropical weather systems such as typhoon and convective clouds, forming the “monsoon trough” (Figure 1A)[4].

As seen, there are two different hypotheses of mon-soon: sea-land differential heating, or ITCZ shifting. Although the two concepts are not necessarily mutually exclusive, there seems to be no room for their compro-mise: The classical concept of monsoon requires at least 120˚ degree reversal in the direction of prevailing winds near-surface between winter and summer, which is ab-sent in America; hence there should be no monsoon in America, and it makes no sense to talk on a “global monsoon”. The opposite concept believes that monsoon could exist without land, and the monsoon is interpreted as an ITCZ substantially away (>10°) from the equator, but not as a result of heating contrast between land and sea[7]. The latter view had sparked hot debate among climatologists, although in reality monsoon is controlled by both ITCZ shift and sea-land heating contrast. The new concept of “global monsoon” and its relationship with ITCZ, however, provides deeper insights into the mechanism of monsoon and also helps the pa-leo-monsoon studies to proceed from description to ex-ploration. As the “climate equator”, the ITCZ is of cen-tral importance to geology when studies on climate evo-lution extend from polar ice-sheets to low latitude pro- cesses.

The ITCZ and the western Pacific warm pool, with the maximal solar heating and SST, are the most inten-sive supplier of vapor and energy from the ocean to the atmosphere[8]. The ascending flow from here makes up the upward branch of the monsoon circulation, i.e. the meridianal Hedley cell, which bring about aridity to the region of its descending flow and the maximal precipita-tion to the region of its ascending flow[9]. The global- scale seasonal overturning of the atmosphere throughout the tropics and subtropics is called “global monsoon” and delineates the large-scale shifts of the precipitation

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1115

FRO

NTI

ERS

G

EO

LOG

Y

Figure 1 Geographic distribution of the modern monsoon systems. A. The Asia-Africa-Australia monsoon region (redrawn from ref. [5]); B. Seasonal shift of ITCZ for boreal summer and winter; C. Six monsoon regions according to annual precipitation range (grey shade): a, North American; b, South American; c, North African; d, South African; e, Asian; f, Australian-Indonesian (redrawn from ref. [6]). zones[1]. Therefore, a combination of wind and rain, i.e. the seasonal reversal of wind direction and the seasonal alteration of dry vs humid climate, can be used to meas-ure monsoon strength. This will be the basic concept of monsoon adopted in this paper. 1.2 Modern global monsoon system Several regional monsoon systems are embedded within the global monsoon, therefore the monsoon is global and regional at the same time. A quantitative criterion has been solicited to measure the monsoon strength and to delineate the monsoon regions, and a number of mon-soon indices have been proposed, such as “the dynami-cal normalized seasonality”[10] or a definition based on wind and “Brightness Temperature”[11] proposed by Chinese authors. Although no single index has been unanimously accepted, all of the proposals are based on two features: the wind or/and rainfall. Trenberth et al. (2000)[1] delineated 6 monsoon regions in the world based on overturning in divergent circulation, i.e., wind,

and distinguished the global-scale overturning from the relatively shallow overturning of regional scale. Using the data from 1979 to 1993 they found that the global and regional factors make up 60% and 20%, respectively, of the annual cycle variance in monsoon circulation. On the other hand, Wang & Ding (2006, 2008)[6,12] used rainfall, namely the precipitation range for the period of 1948―2003 (Figure 1C), to define the global monsoon domain with similar 6 monsoon regions ( North Ameri-can, South American, North African, South African, Asian and Australian-Indonesian). They also found an overall weakening trend of the global land monsoon precipitation over the last 56 years. The scheme with six monsoon regions, three for each of the Hemispheres, is widely accepted in the modern climatology (ref. [13] as an example), and paleo-monsoon records have been re-ported from all the 6 regions in recent years, revealing their common features of the global monsoon system as well as regional specialties. The latter can be illustrated

1116 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

with the absence of cross-equator flow in the North American monsoon[14], the limitation of northward ex-tent in the North African monsoon[15], the enhanced ocean influence on the Australian monsoon[13], amongst others. If the seasonal range in precipitation is compared, the most remarkable seasonal contrast occurs in South Asia and South America, followed by African and then North America[16]. Because the geological records of rainfall is more common than those of wind, a precipita-tion-based definition of monsoon is easier to be adopted to paleo-monsoon studies, not to mention its social rele-vance dealing with flood/drought disasters. This defini-tion will be followed in the present study.

Among the six monsoon regions of the modern world, the Asian and North African monsoons are most exten-sively discussed, in contrast with the American mon-soons. Although the modern climatology community has recognized the North and South American monsoons since the last decade (such as in refs. [17,18]), and a number of international projects have been set up for

their studies, in the paleoclimatology the American monsoons are mostly reported as shifts of ITCZ. For example, the humidity changes in the Amazon basin can be interpreted either as the ITCZ migrations, or varia-tions of the South American monsoon (Figure 2). With the southward migration of the ITCZ in February―May, the South American summer monsoon brings precipita-tion to the Amazon basin. If the southward displacement of the ITCZ is impeded by the warm SST in the tropical Atlantic and the northward low-level atmospheric flow, the South American monsoon wanes and causes droughts in northeastern Brazil[19].

As mentioned earlier, the monsoon can be viewed, from a global aspect, as seasonal migrations of the ITCZ (Figure 1B). Accordingly, some meteorologists do no see essential difference between the monsoon and trade wind, and the monsoon is distinguished by its cross- equatorial flow which enables seasonal expansion of the tropical wet centres out of the equatorial zone[11]; an example is the definition of monsoon requiring the ITCZ

Figure 2 The South American monsoon and ITCZ. Solid lines denote the austral summer (January) and winter (July) positions of ITCZ, dashed line indicates the average position of the South Atlantic Convergence Zone (SACZ); grey shad shows the humid Amazon and Atlantic rainforests (based on refs. [21, 22, 23]).

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1117

FRO

NTI

ERS

G

EO

LOG

Y

migration more than 10° away from the Equator[7]. De-spite of the absence of unanimous agreement on this concept among climatologists, it is useful in paleocli-matology. Given the limitation in resolution and com-pleteness of the geological records, it is not always pos-sible to discriminate monsoon and trades whereas the migration of ITCZ is recognizable. Therefore, the ITCZ lies in the core of geological reconstruction of the tropi-cal climate processes and provides a substantial criterion in paleo-monsoon studies. Certainly, the close tie be-tween the monsoon and ITCZ does not mean their equality, and the range of ITCZ migration should not be used to delineate the monsoon regions. In the global monsoon system there are subtropical monsoons beyond the ITCZ ranges, along with the tropical monsoons di-rectly controlled by the ITCZ shifts. An example is the East Asian monsoon which includes both the Pacific zone and the Meiyu-Baiu Frontal zone, corresponding to tropical and subtropical monsoons, respectively; and a similar pattern exists in South America where the South Atlantic Convergence Zone (SACZ) belongs to sub-tropical monsoon (Figure 2), similar to the Meiyu-Baiu Frontal zone in East Asia[20].

1.3 Geological implications of global monsoon

The concept of global monsoon in the atmospheric sci-ences has inspired new thoughts into paleoclimatology. Since the monsoon represents seasonal migrations of ITCZ, and the migrations are predetermined by the Earth’s rotation and its tilted axis, the monsoon should be a perpetual feature existing throughout the geological history; only a regional monsoon system has its history of establishing and fading. Since the location and inten-sity of the monsoons depends on the differential heating between land and sea, all changes in sea-land distribu-tion, plateau uplift, and sea surface temperature (SST) can impact the monsoon system. Some authors believe that the land-sea contrast is not a necessary condition for the monsoon, and the seasonal changes in SST along would be sufficient to cause the ITCZ migrations, re-sulting in the “aquaplanet monsoon” concept[7,24], a diametrically opposite view to the classical concept of monsoon origin. Objectively, this new hypothesis em-phasized perpetuity of the monsoon in the Earth system, but in the reality the SST itself depends on the sea-land distribution. The monsoon variations not only respond to the seasonal migration of the solar insolation, but also to changes in the underlying surface features such as SST

and topography. Therefore, the monsoon is sensitive to both the orbital parameters of the Earth and the tectonic deformations in the geological aspect.

The geological evolution of the global monsoon em-braces a wide spectrum of time scales, ranging from the Wilson cycles of the plate tectonics to the centennial and decadal cycles of the solar activities. As the recent stud-ies show, not only the regional monsoons are changing in their intensity, but also the global monsoon displays its variations. The second half of the 20th century, for instance, has witnessed an overall weakening of the global monsoon[12], whereas previous studies reported an intensification of the regional monsoon over the past 400 years (such as ref. [25]). The discrepancy in results underscores the necessity of systematical paleo-mon- soon studies on a global scale to explore the monsoon variations in space and time, and this is the way to un-veil the forcing mechanism of the monsoon variations and to find the trend of the monsoon evolution.

Over the geological history, the “Ice-House” regime with large-size polar ice-cap lasted much shorter than the “Hot-House” regime without distinctive ice sheet, and the bipolar ice caped Earth since the last 2―3 Ma is unique for the 600 Ma of the Phanerozoic at the least. In the absence of large ice cap, the low-latitude processes must be the primary driving force for the changing Earth’s climate system, and the global monsoon as the major component of the low-latitude processes should be the priority in the Phanerozoic paleo-climate studies. Paleoclimatology has been concentrating on the devel-opment of the polar ice-sheets and the formation of oceanic deep water, now it has to devote attention also to the migrations of the ITCZ, the climate equator, and to explore the mechanism of changes in the low-latitude processes, such as the monsoon-ENSO linkage. Both belonging to low-latitude processes, the modern obser-vations have shown some negative connection between the monsoon and El Niño: the monsoon weakens when El Niño is strong, and vice versa[26]. This demonstrates the relationship between two atmospheric circulations: the longitudinal Hadley Cell responsible for monsoon, and the latitudinal Walker Cell effecting the El Niño. The relative intensity between the two cells changes in the geological history and influences the global climate. The transition from the warm Pliocene to the Pleistocene ice ages was possibly associated with the intensification of the latitudinal Walker Cell at the cost of the weaken-

1118 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

ing longitudinal Hadley Cell[27]. Up to now the tropical paleoclimatology is poorly understood, and the geologi-cal evolution of the global monsoon holds the key to this new field of climate studies.

2 Global monsoon and orbital cycles

2.1 Global monsoon in Holocene

The Holocene was the starting point in paleo-monsoon studies and has yielded the richest paleo-monsoon re-cords. High levels of lakes across parts of Africa, Arabia, and India between 10 ka and 5 ka BP were incompatible with the glacial cycles, and this has led Kutzbach (1981)[28] to calculate the global average (not 65°N) so-lar radiation for July 9ka ago and to find the preces-sional cylicity in the orbital forcing of monsoon. After-wards, both the computer modeling and geological ob-servation have supported this hypothesis[13] and ap-proved that the precessional cycle of monsoon variations is a global feature with opposite phases between the Northern and Southern Hemispheres.

With the large amount of high-resolution climate re-cords appearing over the recent years, the general trend of the Holocene global monsoon has been becoming clearer. The reported sequences, no matter for the Indian monsoon based on the stalagmite δ 18O from Qunf Cave in southern Oman (Figure 3A)[29] and Globigerina bul-loides % from the Arabian Sea[30], the African monsoon based on terrigenous detritus % from ODP 658 in the tropical Atlantic off Western Africa (Figure 3C)[31], the North American monsoon based on Ti% in laminated deposits from the Carioca Basin off Venezuela (Figure 3B)[32], or the East Asian monsoon based on stalagmite δ 18O from Dongge Cave, south China (Figure 3D)[33], all display the same feature of gradual weakening of the summer monsoon in the northern Hemisphere since 8000 a BP. The late Holocene reduction in monsoon precipitation exerted a significant impact on the early human societies, and the major droughts around 4000 a BP were responsible to the collapse or migration of civi-lizations in Egypt, Mesopotamia and India[34].

Meanwhile, the solar radiation variations caused by the precessional cycles are opposite in phase between the two Hemispheres, thus the austral monsoons should be weak in the early and middle Holocene and strengthen in the late Holocene. Indeed, the pollen spec-trum shows that the Amazon rainforest has expanded

Figure 3 The common trend of the Holocene monsoons in the Northern Hemisphere. A. South Asian monsoon: the stalagmite δ 18O from Qunf Cave in Southern Oman[29]; B. North African mon-soon: terrigenous detritus % from ODP 658, tropical Atlantic off Western Africa[31]; C. North American monsoon: Ti% in laminated deposits from the Carioca Basin off Venezuela[32]; D. East Asian monsoon: stalagmite δ 18O from Dongge Cave, South China[33].

remarkably in the last 3 ka[35]; and sediment record of Lake Titicaca near 17°S in Peru indicates driest condi-tions between 5.5―8 ka, the beginning of increased humidity after 5 ka and a high lake level at 3 ka[36], im-plying intensification of the summer monsoon in the late Holocene. As the monsoon can be considered as the seasonal shifts of ITCZ, the precessional cycle of the global monsoon is also manifested as North-South mi-grations of the ITCZ. During the early Holocene, as the summer insolation enhanced in the Northern Hemi-sphere, the ITCZ was displaced northward, and the bo-real summer monsoon intensified. This was followed by a decrease in summer insolation and southern migration of the ITCZ, resulting in weakened summer monsoon in the Northern Hemisphere. This trend can be illustrated with central America where the shift of the ITCZ is ex-

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1119

FRO

NTI

ERS

G

EO

LOG

Y

pressed by the changes in the monsoon precipitation[32], as well as by the fluctuations of the SST in the Eastern Pacific[37].

The above discussions have shown the global nature of the orbital forcing of the Holocene monsoon; but the monsoon system is complicated by the regional differ-ences. The precessional cycle is the common feature shared by the various monsoon systems of the Earth, or more precisely, of the Hemispheres. Meanwhile, the specific sea-land settings of the regions may bring about different monsoon response to the same orbital forcing. This can be shown with the two monsoon systems in the southern Hemisphere, the South American and Austra-lian monsoons. The late Holoecne intensification of the South America monsoon was mentioned above, but the Australian monsoon does not show the same trend, and the difference is rooted in the SST. As shown by the numerical modeling, the Australian monsoon was not weak during the early Holocene because of the positive feedback of the SST[38]. Also due to the positive feed-back of the sea, the African summer monsoon exhibits the most conspicuous response to the Holocene insola-tion change among the six regional monsoons, which can be illustrated with a striking case in the Sahel. A humid climate was prevailing there in the early Holo-cene, and about 7 ka the Chad Lake had an area about 340 000 km2, or only 8% less than the present-day Cas-pian Sea[40], but shrunk to the present area of 6 000 km2. The extreme aridity in the modern Sahel can not be ex-plained by the global factor of orbital cycle only, but is attributed to the oceanic forcing, as the abnormally high Atlantic SST off Africa causes deep convection to mi-grate over the ocean and reduces monsoon precipitation on land[41].

The monsoon variations is further complicated by the influence of the high-latitude climate, and this is clearly seen in the monsoon changes at the sub-orbital scales. The Dansgaard/Oeschger (D/O) events of millennial scale have been detected in the East Asian[42–44] and West African monsoons[45,46] and are well correlated with the Greenland ice-core records. Again, the regional monsoon systems differ in their suborbital variations because of the specific features in the sea-land patterns and other background settings. According to Weldeab et al. (2007)[46], for example, the West African monsoon displayed greater millennial-scale variations in the pe-nultimate interglacial, but the East Asian monsoon var-ied more remarkably during the last glacial (MIS 2-3).

In the last deglciation, the SST in the Gulf of Guinea rose with the increasing insolation, but the intensifica-tion of the African monsoon lagged behind the SST by about 7 ka, and the authors ascribed the lag to the high-latitudes influence[46].

Apart from the orbital forcing, the monsoon also re-sponds to insolation changes caused by the cycles of the solar activities[47]. For example, the 200-a Suess cycles of the North American monsoon precipitation were rec-ognized in the lake deposit sequence over the past 2600 years from northwest Yucatan, Mexico[48]; and the cy-clic droughts are believed to have led to the collapse of the Maya culture at AD 750―1050[49].

2.2 Global monsoon in Pleistocene

(i) Monsoon and glaciation. The long sequences of deep-sea sediments and loess-paleosol deposits have enabled systematical studies of the monsoon evolution over the Pleistocene, and the pioneering work was based on the African monsoon record from the Mediterra-nean[50] and the Indian monsoon records from the Ara-bian Sea[51]. Thanks to the remarkable progress over the last two decades, now the history of the African-Asian Paleo-monsoon has traced back to the Miocene, and the link between the monsoon and glaciation in the Pleisto-cene has become a hot spot in researches. As the Asian paleo-monsoon was synthesized a few years ago by the SCOR-IMAGES Working Group, it is no need here to duplicate the discussion on Asian monsoon in glacial cycles. Nevertheless, one point in the synthesis is wor-thy of a special attention: “monsoon variability should not be considered only in the context of glacial-inter- glacial variability as is commonly the case, but also in the broader context of tropical and subtropical variabil-ity involving ocean-atmosphere interaction (SST and moisture flux)”[52]. In fact, the closest relationship be-tween monsoon and glacial cycles was observed in the loess-paleosol sequences: the loess accumulation during glacials suggests enhanced dust transport by the winter monsoon, the paleosol formation during interglacials indicates increased vapor supply by the summer mon-soon. However, a simplistic approach to the mon-soon-glaciation linkage would be misleading if the stages of intensified winter vs summer monsoons is simply correlated to the glacials and intergalcials with-out phase comparison. The strong winter monsoon is a specific feature of the modern East Asian monsoon, and there is no question that the winter monsoon intensified

1120 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

with the development of the boreal ice-sheet in the Pleistocene; but there is no reason to conclude that the summer monsoon strength grows in the interglacials. Its weakening since the 8―9 ka, as discussed above, shows that the summer monsoon as a low-latitude process ba-sically follows the precessional forcing but not the ice- sheet changes.

(ii) Precession cycles of monsoon. The opposite trends observed in the Holocene from two Hemispheres convincingly support the precession forcing of the monsoon, and this should apply also to the Pleistocene. As discovered in the 1980s, large lakes repeatedly ap-peared in the arid regions of Africa and then dried up, following the 20-ka precession cycles of monsoon pre-cipitation[53]. The strengthened African summer mon-soon may cause floods of the Nile River and provide the oligotrophic Mediterranean with rich nutrients which in turn stimulate diatom blooms and deposition of sapropel layers on the sea bottom. In result, the formation of sap-ropel occurs every 20-ka in the Mediterranean in re-sponse to the precession regardless of the glacial cy-cles[50]. Similar precession-controlled variations of 20-ka cyclicity in productivity were also reported from the Arabian Sea driven by the Indian monsoon[54]. Since the same pattern could be traced further to the equatorial Pacific, the variations in productivity are attributed to the low-latitude processes in the ocean, but not con-trolled by the ice-sheet changes[55].

The recent appearance of high-resolution climate re-cords from ice-core bubbles and stalagmite isotopes, which show clear precession cyclicity of the monsoon variations, has brought some fresh air to the paleocli-mate studies, showing clear precession cyclicity of the monsoon variations. δ 18Oair from ice core controlled by both ice-volume and vegetation biomass exhibits distinct precessional periods (Figure 4B). When the ice-volume component is removed from δ 18Oair, the resulting “Dole effect”[56] is indicative of the global vegetation size, and its prevailing 20-ka cycles suggest that the variations in vegetation biomass is driven by the boreal summer monsoon[57]. Furthermore, the precession period also dominates in the methane concentration record in ice- core, being influenced by insolation forcing of north- tropical (monsoonal) wetlands[58]. All the above listed signals of the global monsoon follow insolation varia-tions at the precession period in the Northern Hemi-sphere, and it is easily explained by the present domi-

nance of continents in the Northern and of ocean in the Southern Hemisphere[59].

Up to now, terrestrial paleo-monsoon records of the late Pleistocene with a highest time resolution are from stalagmites in South China, including Hulu Cave in Jiangsu, Gongge Cave in Guizhou and Sanbao Cave in Hubei[33,42,44,61,62]. Looking at the stalagmite δ 18O se-quences over the 220 ka, the most striking feature is the 20-ka precession cycles (Figure 4A)[44]. Because the stalagmite has recorded the δ 18O of rain water, the East Asian monsoon history generated by the Chinese sta-lagmites is of a particular value by revealing “the rhythm of the rain”[63]. Stalagmites are certainly not re-stricted to East Asia, and the same precession cycles are observed in stalagmites on the eastern coast of the Mediterranean for the African monsoon[64], and in Oman for the Indian monsoon[29]. The same is found for the South American monsoon recorded in stalagmites from Brazil, although the phase is opposite to the Northern Hemisphere monsoon[21]. Of particular interest are the two monsoon systems from North and South Africa. When the North African monsoon index based on dia-tom from the tropical Atlantic (Figure 4C) is compared with the South African rainfall estimated from sediment properties in the Pretoria Saltpan (Figure 4E), precession signal is found to prevail in both records, but the phase of the 20-ka cycles is opposite. This is again a strong evidence for the precession forcing of the monsoon, in-dicating that the monsoon systems are driven by low- latitude forcing of the two Hemispheres (Figure 4D).

(iii) Debate on phase lag of monsoon variations. Looking back on the development of paleoclimate

studies, the deep-sea records have enjoyed high priority over the past 30―40 years. Foraminiferal δ18O sequences from various oceans are well correlated based on glacial cycles, and age models are generated after their astro-nomical tuning to obliquity, and the high-latitude pro- cesses are hence found at the core of climate cyclicity. Now the newly appeared high-resolution records from ice-core and stalagmite challenge the conventional ap-proach. Unlike the marine sediments, the ice-core bub-ble contains “fossil air”, and the stalagmite δ 18O pre-serves signal of “fossil rain”, offering a more direct re-cord of the monsoon, as compared with marine archives which can be affected by the “Great Conveyer Belt” and regional circulation of the ocean. Consequently, it be-comes a question whether the Quaternary chronology

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1121

FRO

NTI

ERS

G

EO

LOG

Y

Figure 4 Precession periods of the monsoon in the Northern and Southern Hemispheres over the last 200-ka. A. Stalagmite δ 18O from Sanbao and Hulu caves, South China (solid line), dash line shows summer insolation at 65°N; B. δ 18Oair from Vostok ice-core; C. North African monsoon index based on diatom counting at Core RC24-07, tropical Eastern Atlantic (solid line), dashed line shows July insolation at 20°N; D. Summer insolation at low-latitudes in the Northern (solid line, July at 20°N) and Southern (dashed line, January 30°S) Hemispheres; E. Pre-cipitation-based South African monsoon variations estimated from sediment properties in the Pretoria Saltpan, South Africa (solid line), dashed line indicates January insolation at 30°S. Note the opposite phase between E and A-C (A and B simplified from ref. [44]; C,D,E from ref. [60]).

should be based on the monsoonal precession cycles as shown in ice-core bubble and stalagmite isotope, or on the obliquity cycles of ice-sheet fluctuations. Proposals are being made to reconsider the orbital forcing of the climate changes and the “tuning” of chronology. On the millennial time scale, “ Chinese interstadials” have been introduced to indicate strong summer monsoon events in stalagmite[44,62]. On the orbital time scale, a meth-ane-based time scale for Vostok ice was proposed on the basis of tuning to July 30°N insolation[65] (Figure 5(b)). At the same time, a debate has been launched concern-

ing the orbital forcing of the monsoon variations. Systematical investigations of deep-sea records of the

monsoon started from the Indian Ocean[51]. Analyzing a 350-ka-long sequence of the Indian monsoon records in the early 1990s, Clemens and Prell found that the prox-ies of monsoon-induced upwelling are sensitive to pre-cession and obliquity forcing, and that latent heat export from the southern subtropical Indian Ocean are domi-nant driving mechanisms at the obliquity band according to the phase relationship[66]. More than ten years later, they built up the summer-monsoon stack combining 5

1122 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

Figure 5 Various monsoon records over the past 350-ka. (a) Multi-proxy South Asian summer monsoon stack (solid line), dashed line de-noted Summer monsoon factor[67]; (b) comparison of Vostok CH4 concentration (solid line) and July 30°N insolation (dashed line)[65].

proxies (Figure 5(a)) and the summer monsoon factor using principle components analysis to evaluate the late Pleistocene variations of the Indian monsoon. They con-cluded that approximately 26% of the total variance re-sides in the obliquity band and only 18% in the preces-sion band, suggesting the main orbital forcing of the Indian summer monsoon is obliquity rather than preces-sion[67]. The conclusion is supported also by a study on loess sequence where the obliquity signal is found to be much stronger than that of precession in the summer monsoon records based on magnetic susceptibility[68]. Recently Ruddiman (2006)[69] questioned this view of monsoon forcing as being at odds not only with the newly appearing monsoon records from stalagmites and ice-cores, but also with the original hypothesis by Kutz-bach (1981)[28] on the precessional forcing of monsoon variations. He argues that the methane concentration (Figure 5(b)) and Dole effect from ice-core and the oxy-gen isotope of the low-latitude stalagmites, all show a clean precession period which is the summer monsoon signal; whereas the obliquity-dominating “monsoon proxies” in the Indian Ocean, even though they occur in a monsoon region, are associated with other process such as winter monsoon or offshore transport of particu-late and nutrient material from the continental shelf[69].

In their reply Clemens and Prell (2007)[70] claim that the focus of debate lies in the phase difference. Between

the precession-driven radiation and summer monsoon response, Ruddiman (2006)[69] suggests no phase lag for the ice-core and stalagmite records, while Clemens and Prell assumed a phase lag as long as 8 ka (or 120°). Meanwhile, Kutzbach et al. (2007)[71] used a fully cou-pled general circulation ocean-atmosphere model to study the response of northern and southern hemisphere summer monsoons to orbital forcing over the past 284-ka. In result they found a lag only about 30°, or al-most no lag, at the precession band, practically denying the phase lag interpretation for the Indian Ocean. The debate seems to have occurred between the marine and terrestrial records, but in essence it is on the basic ques-tion about orbital forcing of monsoon: is it precession, or obliquity plus precession? In other words, whether the summer monsoon variations are forced by preces-sion-driven low-latitude insolation, or by obliq-uity-driven insolation controlling the high-latitude proc-esses? The debate, as believed, will continue, but it cov-ers two sets of questions: how to recognize the summer monsoon proxies, and how to discriminate global and regional signals in monsoon records. Up to now, there is no unanimously agreement on proxies of monsoon in-tensity[52]. Not only marine proxies, the stalagmite-based monsoon proxies are also disputed. On the other hand, monsoon variability is a mixing of both global and re-gional components. There is global monsoon signal in

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1123

FRO

NTI

ERS

G

EO

LOG

Y

the records, such as the methane concentration and Dole effect signal in ice-cores, but superimposed there is also a regional signal, such as the precipitation changes caused by regional sea-land thermal contrast in the re-gion[72] or the influence by the high-latitude processes. A convincing conclusion will be reached only after sorting out the key questions and after accumulation of more geological and modeling data.

2.3 Global monsoon in Cenozoic (i) Long eccentricity: “heartbeat” of Earth system. Be-cause a millennial if not centennial time resolution is required to recognize precession cycles, it is a difficult exercise for obtaining these short orbital cycles from pre-Quaternary sequences. More applicable are thus longer orbital cycles, namely the 100-ka and 400-ka cy-cles of eccentricity. Eccentricity enters into the climate system mainly through modulating the amplitude of climate precession, as the difference between aphelion and perihelion depends on eccentricity, and precession forcing disappears if eccentricity equals zero. The Earth is currently experiencing its eccentricity minimum, but the climate precession forcing is still strong as we have seen from the monsoon changes in the Holocene. We can image, therefore, the monsoon variations must be much more significant during the eccentricity maximum. Unlike precession, eccentricity forcing of the monsoon has no difference for the Northern and Southern Hemi-spheres; and of particular importance is long eccentricity of 400-ka which is the most stable orbital parameter throughout the geological history[73,74]. The long eccen-tricity cycle far exceeds the glacial cycle in time length and modulates the long-term monsoon variations. By controlling the weathering rate and other processes, the long-term monsoon cycles lead to periodic changes in the oceanic carbon reservoir. Since the residence time of carbon in the oceanic reservoir is much longer than 100-ka[75,76], the 400-ka period of the monsoon is best manifested in the inorganic δ 13C and in the carbonate reservation, representing the major rhythm of the mon-soon at the orbital time scales. The most conspicuous signal of the rhythm is the oceanic carbon isotope maximum, or δ 13Cmax

[77–79], normally occurring at the eccentricity minimum.

Some 30 years ago, high-resolution marine δ 18O and δ 13C sequences were available only for the late Pleisto-cene. Because of the limited length of time sequences, the longest cycle discerned was the 100 ka short eccen-

tricity, but as Berger noticed at that time, astronomically the most important eccentricity term has a periodicity of 413-ka, and “if longer cores can be investigated by tech-niques allowing high resolution spectrum, I would ex-pext…the main period of 413000 a becoming strongly displayed”[80]. Now the high-resolution deep sea records extended back to the entire Cenozoic thanks to the pro-gress in technology. An ubiquitous feature common to all these long sequences, as predicted, is the presence of a 400-ka long eccentricity period, most distinctly ex-pressed in the δ 13C records. Typical 400-ka long eccen-tricity cycles were reported from the Oligocene of the tropical Pacific (Figure 6A) and likened to be the Earth’s “heartbeat”[81] showing the rhythms of the global climate changes. The 400-ka period is obvious also in the δ 13C sequence of Late Oligocene to Early Miocene from the tropical Atlantic (Figure 6B)[82]. For the Middle Mio-cene, the long eccentricity cycles were noticed in δ 13C records many years ago[83,84], but in a less distinct form due to the resolution limitation; clear 400-ka cycles were observed only recently in the new dataset of benthic δ 13C of 13.0―16.5 Ma from the South China Sea and the tropical Eastern Pacific (Figure 6C)[85]. Now, the long eccentricity cycles have been extensively reported from various δ 18O and δ 13C records of various time intervals of the Cenozoic, including the transition from Paleocene to Eocene[86], Early Oligocene (30.5―34.0 Ma)[87,88] , Middle Oligocene[89], and Late Oligocene to Early Miocene (20.5―25.5 Ma)[90,91] or Middle Mio-cene (16―24 Ma)[92], displaying a basic rhythm in the Cenozoic hydrological and oceanic carbon cycles. As will be shown later, this basic rhythm is nothing but a long cyclicity in the global monsoon variability.

(ii) Long eccentricity in Pliocene monsoon records. One of major findings of the ODP cruise to the South

China Sea is the change of the long-term cyclicity in the oceanic carbon reservoir over the last 5 Ma. Originally discovered in the southern South China Sea, this cyclic-ity was found to be common to the global ocean, imply-ing changes of the oceanic carbon reservoir on the 105 a time scale[77–79]. The best record of the long-term varia-tions in the oceanic carbon reservoir was observed in the Mediterranean region (Figure 7B), where the interna-tional stratotypes from southern Italy and Sicily are classical sequences for astronomical stratigraphy[93], and at its basis lie orbital cycles of the African monsoon.

1124 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

Figure 6 The 400-ka long eccentricity period in oceanic δ 13C records of the Cenozoic. A. Oligocene benthic δ 13C at ODP 1218(8°53′N, 135°22′W), tropical Pacific, and long eccentricity cycles from 21 to 34 Ma[81]; B. Late Oligocene-Early Miocene benthic δ 13C at ODP 926(3°43′N, 42°54′W)and ODP 929 (5°58′N, 43°44′W), tropical Atlantic, and long eccentricity cycles from 18 to 26.5 Ma[82]; C. Middle Miocene benthic δ 13C at ODP1146(19°27′N, 116°16′E), South China Sea, and ODP1237(16°S, 76°23′W), tropical Eastern Pacific, and long eccentricity cycles from 13 to 16.5 Ma[85]. Note different scales between the three sequences.

The 400-ka cycles in the Pliocene deposits from

southern Italy consist of alternating low-productivity carbonates and high-productivity sapropels or dark lay-ers, and the sapropel layers are formed under the condi-tions of intensified monsoon precipitation[95]. As men-tioned above, a strong summer monsoon in North Africa leads to the Nile floods, and the nutrient-rich river water input causes diatom blooms in the oligotrophic Mediter-ranean and deposition of sapropels[96,97] or dark layers[98] there. Both the summer monsoon and the sapropel for-mation are controlled by the precession cycles[50]. At the eccentricity maximum the climate precession has the largest amplitude of variations, the enhanced monsoon results in accumulation of sapropel clusters and in lighter δ 13C, and the carbonate % declines due to the diatom blooms and enhanced supply of terrigenous de-tritus; conversely, at the eccentricity minimum the am-plitude of precession reduces, monsoon weakens, supply of nutrients declines, δ 13C becomes heavier and cocco-lithophores flourish resulting in higher carbonate % (Figure 7B). As the Mediterranean is prevented from the

influence of the “Great Conveyer Belt” by its shallow sill depth, and Africa is a continent striding the equator (see Figure 1B), the African monsoon is responding di-rectly to the orbital forcing with a minimal impact of outside processes. Therefore, the long eccentricity in the Pliocene δ 13C in Italy is better expressed than in other regions, representing a classical example of the low- latitude response to orbital forcing.

Similar 400-ka cyclicity is observed in the Pliocene δ 13C sequences of other oceans and in long-range mon-soon records. This can be illustrated with the Indian monsoon eolian dust records in the Arabian Sea (Figure 7C)[31], or with the African monsoon dust records in tropical Atlantic off West Africa (Figure 7D)[94], all showing monsoon enhanced at the 400-ka eccentricity maximum and weakened at its minimum. The intensity of chemical weathering estimated from the element ratio in the Pliocene of the northern South China Sea shows the same long eccentricity control of the East Asian summer monsoon precipitation[99].

(iii) Long-term cycles in Pleistocene oceanic carbon

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1125

FRO

NTI

ERS

G

EO

LOG

Y

Figure 7 Long eccentricity cycles and the global monsoon over the last 5 Ma. A. Eccentricity cycles (100-ka and 400-ka); B. Planktonic δ13C from Rossello composite section, Sicily, and ODP 964, Mediterranean (after Lourens, 19941)); C. Eolian dust% in ODP 721/722 (16°38′N, 59°50′E), Arabian Sea; D. Dust flux (g m−2 a−1)[94] at ODP 659(18°05′N, 21°02′W), eastern equatorial Atlantic off Africa.

reservoir. The above discussion shows that the 400-ka cycles are pervasive in all Cenozoic records, from Pa-leocene to Pliocene, of the oceanic carbon reservoir: δ 13C becomes lighter and displays larger amplitude of fluctuations at eccentricity maximum, compared to the δ 13C heavier values (“δ 13Cmax”) at its minimum. As seen from the Pliocene, the long cycles are associated with low-latitude processes. Large ice-sheet did not exist in the early Cenozoic, and the monsoon variations were then central to the global climate changes. As “heart-beat” of the Earth system, the long-eccentricity pacing of the climate runs through the entire Cenozoic Era[100].

However, this long eccentricity cycles becomes ob-scured in the Pleistocene after 1.6 Ma, and δ 13Cmax oc-curred at 1.0 Ma (MIS 25-27) and 0.5 Ma (MIS 13), no more corresponding to the eccentricity minimum (Figure 8C). The same is observed in the deep-sea carbonate %

records (Figure 8B) of all the oceans, implying that the long-term changes in the oceanic carbon reservoir do not follow the 400-ka eccentricity cycles in the Pleisto-cene[79]; rather, a quasi-cyclicity of 0.5 Ma appears in the last million years. As the long-eccentricity signal be-comes obscured also in other records of the Pleistocene monsoon, such as in dust records of the Indian and Afri-can monsoons in the Indian (Figure 7C) and Atlantic Oceans (Figure 7D) or in magnetic susceptibility record of the East Asian monsoon in the Chinese Loess Pla-teau[101], the turn at 1.6 Ma indicates a mode shift in the long-term changes of the monsoon climate.

It remains unclear what caused the quasi-cyclicity in the Pleistocene oceanic carbon reservoir. Most probably the growth of the boreal ice-sheet has reorganized the previous way of energy exchange and damped the ec-centricity forcing, disturbing the “heartbeat” and leading

1) Lourens L J. Astronomical forcing of Mediterranean climate during the last 5.3 million years, Ph.D. Thesis, Utrecht University, Netherlands, 1994,

247

1126 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

Figure 8 Three major stages in the oceanic carbon reservoir since 1.6 Ma. A. eccentricity; B. Subtropical South Atlantic Susceptibility (SUSAS) stack[109], negatively correlated with CaCO3%; C. planktonic δ 13C of ODP Site 1143(modified from ref. [79]).

to a kind of “arrhythmia” of the global climate sys-tem[100]. It must not be a coincidence that the turn in long-term cyclicity occurred at 1.6 Ma, as it was the time when the permanent El Niño-like conditions ended and the east-west asymmetry in upper ocean established in the equatorial Pacific[102,103]. As low-latitude phe-nomena, the monsoon and El Niño are closely linked in some complicated way, the former represents the longi-tudinal migration of the ITCZ, and the latter is related to the latitudinal asymmetry in the equatorial upper ocean. The reorganization of the ocean structure at 1.6 Ma, however, was not restricted to the upper ocean and might have an Antarctic origin. As shown by fossil dia-toms, the nutrient-rich Southern Ocean water penetrated into the low-latitude ocean during 3.2―1.6 Ma[104,105], giving rise not only to the asymmetry in the equatorial upper ocean, but also to reorganization of the deep-water structure, such as the divide of the deep-water from the intermediate water in the Atlantic since 1.6 Ma[106]. It is of interest to note that the dumping of the long eccen-tricity forcing of the oceanic carbon reservoir is not spe-cific to the Pleistocene. At the end of the Monterey Carbon shift event of 17―14 Ma in the Miocene, the final long cycle of δ 13C lasted longer than 500-ka, and both the limiting δ 13Cmax events did not correspond to the eccentricity minimum[107,108]. The case is very simi-lar to the 1.6 Ma turn and is presumably associated to the rapid growth of the Antarctic ice-sheet. Therefore, the key to understand the long cycles of the oceanic carbon reservoir is to unveil the meaning of δ 13Cmax. Because Earth is now passing again through eccentricity

minimum and a new δ 13Cmax episode, it is crucial to understand the role played by the long cycles of the oceanic carbon reservoir in the global climate system.

3 Global monsoon and Wilson cycle

3.1 Tectonic factor of monsoon evolution

The tectonic forcing of the monsoon climate has been a long-lasting topic in paleoclimatology, largely focusing on the hypothetical link between the Tibetan uplift and Asian monsoon[110]. As the first controlling factor inside the Earth system, sea-land distribution pattern deter-mines the intensity and location of the monsoon. Young (1987)[111] and Firmeyer (1998)[112] discussed how the distribution of continents and orography influences the monsoon: Low-latitude positions of the continent favor monsoon development, and a north-south mountain bar-rier on it enhances the cross-equator flow. Webster (1981)[113] hypothesized if all the continents were gath-ered around the northern pole, it would be a continental cap north of 14°N with a nearly complete aridity inland, and with monsoonal precipitation in summer along the coastal region. Of course, all the inferences need to be checked with the geological records.

In reality, there were geological times when all the continents on the Earth assembled into one major conti-nent, the last of which was the “Pangaea”. This Mega- continent generated Mega-monsoon[114,115], the most in-tensive monsoon system in the geological history. The Wilson cycle consists of alternation of sea-floor spread-ing and continent collision, accompanied by the aggre-

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1127

FRO

NTI

ERS

G

EO

LOG

Y

gation and dispersal of continental crust and the corre-sponding waxing and waning of the global monsoon. Since a Wilson cycle takes hundreds of million years, the tectonic-induced monsoon cycles may also last for 108 years, much longer than the orbital-induced mon-soon cycles. As continent collision leads to mountains formation and plateau uplift, the monsoon system is further enhanced by the topographic features. Moreover, the monsoon climate also denotes an un-equal distribu-tion of the hydrological cycle in space and time, i.e. a seasonally and geographically focused precipitation, the enhanced global monsoon means also enhanced aridity in non-monsoon region[116], thus a “mega-drought” co-occurs with “mega-monsoon”.

Tectonics also exerts its influence on geographic dis-tribution of the ITCZ and monsoon. In the modern earth system, the ITCZ exhibits large-scale migrations in the Indian and West Pacific oceans, giving rise to the Af-rica-Asian monsoon region; whereas in the Atlantic and Eastern Pacific the ITCZ is more stable and biased to the north of the equator (Figure 1A). As shown by numerical modeling, the northern position of the ITCZ can be mainly ascribed to the configurations and directions of the coastlines (Figure 1B)[117].

Thus, tectonic deformations shaped the sea-land dis-

tribution and continental topography which in turn con-trols the monsoon intensity and distribution. Today, Eurasia is the largest continent in the world and also the largest monsoon region; India has the most intensive monsoon system due to its position north of the equator, with the Indian Ocean to its south and Tibet to its west. However, the picture of the modern monsoon was com-pletely different ~200 Ma ago, during the time of megacontinent “Pangaea”.

3.2 Megamonsoon of the Megacontinent

From the late Permian to Early Jurassic (~180―250 Ma) all continents assembled into two major landmasses, Laurasia and Gondwanaland, joined near the equator into the super-continent Pangaea, culminating at the early Triassic. The modeling results show a Mega- monsoon of global scale with a reversal in wind direc-tion between summer and winter monsoon circulations (Figure 9 C,D) and large-scale migration of the ITCZ over Pangaea. Precipitation was concentrated near the Tethys coast and almost absent inside the continent (Figure 9 A,B)[114]. The mega-continent is distinguished by extreme continentality and a wide annual range of temperature (50℃) hinterland[118].

The idea about a continental-wise mega-monsoon

Figure 9 Modeling megamonsoon of the Pangaea. A. Schematic diagram illustrating monsoonal circulation in northern summer. Arrows show surface winds, stippling indicates heavy, seasonal rains. B. Modeled precipitation rate (mm/d) on Pangaea for summer. C―D. Modeled surface winds on Pangaea for winter (C) and summer (D), note the seasonal reversal of the wind direction. The grey bar shows the poleward limit of summer monsoon over land and the Intertropical Convergence Zone over ocean (after ref. [119]; data from refs. [114, 120]).

1128 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

first appeared in 1973 when Robinson drew the hypo-thetical position of the ITCZ on the late Permian and early Triassic paleogeographic maps with a 40 degree range of seasonal migration and a mega-scale monsoon system[121]. Although the cross-equator megacontinent provided tectonic settings for development of the mega-monsoon, the monsoon circulation could penetrate into the hinterland only with the presence of highland[122]. The Appalachians in America and Variscan range in Europe reached then a mean altitude of 4500 m and 2000 ― 3000 m[123], respectively. Such altitudes are inferior to Tibet today, but the ranges were of great importance for the monsoon development. The Triassic was distinguished by the most intensive monsoon and also by extreme aridity, and the Pangaea was the time of maximal accumulation of evaporates[124].

Judging from the extensive geological evidence for the “megamonsoon”, the Pangaean monsoon was de-veloped in North America but not in Africa or Asia, be-cause North America was in the low-latitude region of Pangaea, whereas North and South China were outside the main body of the megacontinent. In the southwestern United States, wind-deposited sands accumulated to a thickness of 2500 m during the 160 million years when Pangaea straddled the Equator. These strata are the thickest and most widespread aeolian dune deposits known from the entire global sedimentary record[125], accompanied by loess accumulation in a semiarid re-gion[126].

The Late Triassic Chinle Basin, western Colorado Plateau, was then situated between 5°―15°N, at the margin of the tropical monsoon region (“CL” in Figure 10A). The lacustrine deposits interbedded with eolian dusts provide archives of monsoon evolution in this cli-matically sensitive region[127]. The micro-lamination in the lacustrine deposits of the Newark super-group, northeast US, recorded the alternation of dry/humid conditions and related lake level fluctuations in the Pangaea from Late Triassic to Early Jurassic[128]. De-tailed studies on the 6700-m-long section have revealed a full range of periods of monsoon climate change over the past 33 Ma: varves with 0.2―0.3-mm-thin couplets of alternating light (dry winter) and dark (rainy summer) layers implying seasonal contrast of monsoon climate (Figure 10B); thicker sediment rhythms representing the 20 ka precession cycles (4 m on average), 100 ka (20―25 m) and 400 ka (90 m) eccentricity cycles (Figure 10

C,D)[129]. Under the heavy summer rain, wind-blown sand in dunes at Pangaea’s western margin collapsed into slumps, as recorded in slumped masses in the Early Jurassic Navajo Sandstones in the SW United States[125].

In the Late Jurassic, the “megamonsoon” collapsed with the break-up of Pangaea; but the distribution pat-tern of precipitation remains monsoonal in Late Jurassic, replaced by a predominating planetary zonal in the Cre-taceous[130]. As a typical case of the “Hot-House” Earth system, the Cretaceous climate has repeatedly been a subject of numerical modeling. Although monsoon cli-mate is no more a feature of the Cretaceous, there are also works on the Cretaceous monsoon. For example, Buch (1997)[131] reported that the Cretaceous circum- global current flows westward in Tethys based on a nu-merical simulation, and local surface currents reverse during the Eurasian monsoon months only at the south-ern margin of the Eurasian continent.

Obviously, there is little comparison by scale between the Cretaceous monsoon and the Pangaean megamon-soon. 3.3 Establishment of the Late Cenozoic Monsoon System Only fragmentary reports have been published on the Paleogene monsoon, mostly about monsoon precipita-tion inferred from fossil flora. The earliest record is the early Paleocene tropical rainforest fossils from Colorado, only 1.4 Ma after the Cretaceous/Tertiary boundary. The intensive precipitation is interpreted as of monsoon ori-gin resulted from the relief effects of the Laramide orogeny[132]. Although the local orographic effect only has minor influence on the global monsoon intensity, it may be strong enough to change the regional pattern of monsoon[133]. Another example is from south-central Australia where an Eocene flora was discovered compa-rable with the modern monsoon forest in Northern Aus-tralia, raising a question about the “Eocene monsoon forest in central Australia[134]. Anyway, the Pangaea monsoon records have been reported mainly from North America, the low-latitude region in the Mesozoic; and the African-Asian monsoon region has appeared only in the late Cenozoic.

To answer the question when the modern monsoon system established, we can use the changes of climate patterns in the geological records. The latitudinal distri-bution of climate zones is indicative of planetary at-mospheric circulation, but it is disturbed by the estab-

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1129

FRO

NTI

ERS

G

EO

LOG

Y

Figure 10 Late Triassic monsoon records in North America. A. Locations of the Colorado Plateau with the Chinle Formation (CL), and a chain of rifted basins on the East with the Newark Supergroup (NW). B. Photograph of microlaminated mudstone showing organic- rich/carbonate-rich couplets as annual varves. C. Lake-level fluctuations in a section of the Newark lake sediments, showing 20 ka, 100 ka, and 400 ka cycles. D. Average spectral estimates of sediment cycles in the Newark Basin with the modern precession spectrum (after ref. [119]).

lishment of a monsoon system. Parrish et al.(1982)[135] reconstructed the lithology-based rainfall maps for 7 time intervals from Triassic to Miocene and found that the monsoon system collapsed with the break-up of the supercontinent, and reappeared after the collision be-tween Indian and Asia in the middle Miocene. Following the same principle, Zhou (1984)[136] hypothesized on the basis of lithology in China that the East Asian monsoon started at the Paleogene-Neogene turn, a hypothesis supported by later studies[137,138]. The discovery of the Miocene loess-paleosol succession at Qin’an[139], the analyses of deep-sea sequence in the South China Sea[76,140], the results of numerical modeling[141], and the comparison of the successive phytogeographic patterns of the Cenozoic in China[142], all indicate that the East Asian monsoon system was established no later than the earliest Miocene, most probably in the Late Oligocene.

However, the known history of all other monsoon

systems is shorter than that of the East Asian monsoon. The Indian monsoon became reportedly prominent only since 8 Ma[143,144], and the African monsoon has an even shorter history[145]. The difference is probably related to the insufficient data available, because in general the literature on pre-Quaternary monsoon history is sparse, and this explains why the history of the East Asian monsoon was restricted to the last 2 Ma or 8 Ma just a few years ago. Another question is the different use of monsoon proxies, such as the appearance of C4 plant. Ten years ago, C-4 plant appearance 6―8 Ma ago was thought to mark the establishment of monsoons[146], but later it becomes clear that the C-4 plant development depends on the atmospheric CO2 concentration rather than monsoon climate[147].

As in the earlier geological past, the establishment of the late Cenozoic monsoon system was driven by two tectonic factors: sea-land distribution and plateau uplift,

1130 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

with more attention paid to the latter because of the evi-dent climate impact of the Tibetan Plateau[148]. GCM experiments indicate that strong monsoons can be in-duced by solar forcing only when the elevation of Ti-bet-Himalaya is at least half that of today[110,149]; ac-cordingly, the onset of the Indian and East Asian mon-soons was assigned to ~8 Ma when uplift of the Tibetan Plateau intensified, and the monsoons have enhanced since then with the further plateau uplift [150]. For many years it has remained a prevailing cognition that mon-soons were established with the plateau uplift and pro-gressively intensified during the late Cenozoic. This hypothesis, however, is challenged by the recent discov-eries which have changed not only the time of its onset, but also the trend of development since then. As shown by the new deep-sea geochemical data from the South China Sea, the influence of the East Asian summer monsoon has decreased continuously since the Early Miocene instead of strengthening[151], and this has been shown to be a general tendency of the global chemical weathering[152]. Since the intensification of the African monsoon was ascribed to the uplift of the East African Plateau[145], the above discussion should by no means be understood as a disapproval of the climate role played by the plateau uplift. What we are arguing is that there is no simple linear relationship between the plateau uplift and the monsoon climate.

Changes in sea-land distribution have provided the background for the Asian monsoon to establish. The closure of the Tethys, emergence of the Turgay Sea, and the collision of India with Asia have led to the formation of the Late Cenozoic assembly of continents including Europe, Asia and Africa, formulating the nucleus of a new generation of super-continent[153]. To show the sig-nificance of sea-land distribution for monsoon develop-ment, it suffices to mention that the retreat of the Paratethys in Central Asia along might cause the strengthening of the Asian monsoon and the aridification of the inner land, playing a role as important as the Tibet uplift[154]. Another climate impact of the tectonic changes in sea-land distribution is through gateways of currents. The northward displacement of the Australian Plate switched the source of flow through Indonesia from warm South Pacific to relatively cold North Pacific waters, resulting in decreasing temperature in Indian Ocean, reducing precipitation over East Africa and a weakening of the Indian monsoon in general[155].

Thus, the Wilson cycles of landmass assembly and

supercontinent break-up run through the Earth history, and the monsoon evolution over the past 300 Ma took place in the framework of the last Wilson cycle. Of course, there are more complicated and detailed proc-esses of changes in sea-land distribution and plateau uplift within each of the Wilson cycle. It is a challenge for the Earth science community to test the tectonic hy-pothesis of monsoon development in space and time by integrating the geological records and by numerical modeling.

4 Concluding remarks

4.1 Monsoon evolution and Earth system

To sum up, the global monsoon evolution runs through the 600-Ma Phanerozoic at the least. On the tectonic time scale of 106―108 years, it is responding to the landmass assembly and break-up in the Wilson cycle, with “mega-monsoon” over the mega-continent and its collapse when the mega-continent breaks up. On the orbital time scale of 104―105 years, the global monsoon is controlled by the orbital forcing, displaying 20-ka precession cycles, 100-ka and especially 400-ka eccentricity cycles. On the 103 years and shorter time scales, the solar and other cycles predominate the monsoon variations. A systematical study of the global monsoon evolution is new to paleoclimatology. In fact, all new concepts in the Earth sciences have been recognized as local at the beginning and then propagated into global. The “global tectonic” and the “Great Oceanic Conveyer Belt” have taken decades of data accumulation before their proposal. The same applies to the atmospheric science. Only with remote sensing techniques does it become possible to have a global vision, to observe precipitation over land and sea on a global scale and to describe the large-scale overturning of the cross-equatorial circulations. Similarly, only with the accumulation of regional paleo-monsoons records will the integration into the global paleo-monsoon be possible. The significance of the global monsoon studies goes far beyond the monsoon itself, as it raises a question about the tropical forcing of climate change.

Although climate disasters have been known from the legends since the ancient time, the modern science for a long period did not accept the concept of fundamental changes in the global climate. By the end of the 19th century, leading climatologists still believed that climate

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1131

FRO

NTI

ERS

G

EO

LOG

Y

was nearly constant, and despite of the fluctuations in the measured figures, the mean values remained sta-ble[156]. Large-scale changes in climate were discovered in the 19th century, but in two different aspects: Histo-rians and archeologists saw deluges, while geologists saw glaciations. The focus in modern climatology lies on hydrological cycles, on dry vs humid variations, whereas paleoclimatologists emphasize glacial cycles, changes between warm and cold, and the difference has been maintained until the present day. In contrast with modern climatology which is mainly engaged in low- latitude processes, paleoclimatology in general remains ice-sheet-centered and the majority of the paleo com-munity insists that the global climate change is driven by changes in the ice-sheets through the “Great Oceanic Conveyer Belt”. Studies on the global monsoon in a geological perspective, as discussed in this paper, are targeted to demonstrate the global climate impact of the low-latitude processes. As we believe, the meridianal migrations and latitudinal changes of ITCZ, the “cli-matic equator”, should be considered as the basic ele-ments in climate evolution, and the tropical and ice-sheet forcings are closely connected but relatively independent two major factors in the climate system.

The long-lasting prevalence of high-latitude domi-nance in paleoclimatological studies has its deep root in history and technology. If modern climatology is first of all to deal with climate disasters, such as floods, drought, typhoon or storms which are predominantly of tropical origin, paleoclimatology was born through debates on the reality of ice ages. The quantitative approach to pa-leoclimatology was initiated for reconstruction of the Quaternary ice age and focused on temperature changes and high latitude processes. In the technical aspect, the long absence of effective proxies for low-latitude SST has led to a misconception about the inertia of tropical climate in glacial cycles[157]. It remains a challenge to measure paleo-salinity which is badly needed for inves-tigating hydrological cycles. Up to now, paleoclimatol-ogy is far more advanced in estimating paleo-tempera- ture than paleo-salinity, and in reconstructing paleo-ice- sheet than paleo-wind-field. With additional limitations in sampling and analyzing techniques, high-resolution paleoclimate reconstruction was for a long time applica-ble only to the late Quaternary. However, the recent nu-merous discoveries of significant climate changes in the tropics and in the pre-Quaternary “Hot-House” Earth have greatly broadened the scope of paleoclimatology. It

has been recognized that orbital cycles are not restricted to the ice sheet, and changes in orbital parameters can directly drive tropical climate changes. Monsoon can respond to the orbital forcing directly, not necessarily be driven intermediately by high-latitude processes. More-over, the characteristic orbital cycles of equatorial inso-lation, ranging from semi-precession to eccentricity[158], are helpful in recognizing processes of low-latitude ori-gin. As the “Hot-House” regime has lasted much longer than the “Ice-House” over the geological history, the hot-house climate featured by the monsoon cycles should be more representative than the glacial climate in geology records.

The studies of global monsoon and other low-latitude processes are expected to unveil some basic mechanisms in the Earth system evolution. An example is the above mentioned precession and eccentricity: how these orbital cycles set rhythms of the Earth system variations through controlling the global monsoon intensity? Thanks to the application of magnetic susceptibility, X-ray fluorescence and other core scanning techniques, more and more high-resolution climate sequences have been published for the pre-Quaternary, displaying ubiq-uity of the orbital periods in the records. Examples in-clude precession and eccentricity cycles in the Late Cre-taceous magnetic susceptibility from the South Atlan-tic[159], or the 400-ka period in Fe-content and magnetic susceptibility of the Paleocene ocean[160]. Because of its stability and ubiquity in the geological history, the 400-ka period of long eccentricity has been taken as “the tuning fork of geological time” to tune the ages of geo-logical boundaries[85,92] , and to be used as an unit in geochronology[89]. Unfortunately, the Pleistocene Earth is unique in the entire Phanerozoic to have both of its poles ice-capped, and the 400-a rhythm is significantly disturbed[78]. According to Pierrehumbert[161], “climate is a pas des deux between carbon dioxide and water, with important guest appearances by surface ice cover.” The Quaternary is the time of “guest appearances”. When we applaud the guest for its outstanding performance, we should not ignore or overlook the lead roles.

4.2 Systematical studies of global monsoon

The monsoon studies have been highly prioritized in China, and a wealth of the modern and past monsoon data has been pooled in China. Therefore, the monsoon study could be an ideal choice for the Earth system sci-ence in the country, if the global monsoon evolution is to

1132 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

be approached both in space and time on a cross-disci- plinary basis. For further researches in China, the fol-lowing recommendations can be made:

(i) Studying East-Asian monsoon in global frame-work. One of drawbacks in the current paleo-monsoon studies in China is the lack of precision in concepts. Quite often monsoons are discussed without distin-guishing winter vs summer, tropical vs subtropical, East Asian vs NW Pacific[162]. To study the East Asian mon-soon in the framework of the global monsoon, it requires a clear idea about the subject in the global system, a comparison with other monsoon systems, and a dis-crimination of global vs regional changes. Moreover, the low-latitude processes such as ENSO and trades are closely linked but not identical with the monsoon. The Asian monsoon, for example, is highly sensitive to ENSO, and the close connection has been noticed in the paleo-records, but the study of the ENSO evolution in the geological history is just in the infancy (such as ref. [163]). Therefore, it would be premature to discuss the relationship between ENSO and monsoon in the geo-logical history. Even so, there is no doubt that the geo-logical perspective of low-latitude processes is a new growing point in paleoclimatology.

(ii) Evolution of hydrological cycle in Earth system. The hydrologial cycle or dry/humid variations are the

focus of paleoclimate studies centred at the monsoon and tropical processes. It causes changes in chemical weathering intensity which in turn is closely tied to the carbon cycling and other biogeochemical processes[161]. In-depth investigations in humidity changes are needed for quantitative approaches to paleoclimatology, if the hydrological cycle becomes a priority. Some fundamen-tal questions should be addressed: Is there a link be-tween monsoon intensity and aridity? Were there differ-ent levels in aridity during the stages with weaker vs stronger monsoon intensity? Was the Cretaceous less dry than the Quaternary? Were the Cretaceous xerophytes comparable with those in the Quaternary in their resis-tance to aridity? Another question is about the non-marine petroleum in China. The saline stratified

oil-producing lakes occurred in China mostly in the Eo-cene, and become rare since the Miocene. Was this change associated with the turn from the planetary to monsoonal atmospheric circulation?

(iii) Compiling global monsoon archives since Pa-leozoic. The monsoon records since the Paleozoic have shown a close linkage between the monsoon and the Wilson cycle, but the geological data are fragmentary, and many publications are based on numerical simula-tion. A useful exercise is to compile all the data available and to run numerical modeling to trace the global mon-soon history back to the Paleozoic, including its inten-sity and geographic distributions. So far, valuable con-tributions have been made by American scientists on reconstructions of hydrological cycling and ice-sheet changes in the Late Paleozoic and comparisons with those in the Late Cenozoic[164]. Another example is the modeling of the early Paleogene, in which a paleocli-mate sensitivity test on the basis of paleogeography shows that large scale changes in geography and topog-raphy can change the monsoon pattern, leading to initia-tion of the monsoon in one area but to its closure in an-other[165]. Of course, such activities require a systemati-cal approach to cover at least the Phanerozoic.

(iv) Numerical modeling and prediction of future trend. The past climate researches are expected to throw light on prediction of its future trend. It has been a debate on how long the present interglacial will last. Al-though many publications have been devoted to the topic, most numerical simulations are based on high- latitude processes, without considering the low-latitude processes. Since the future changes in hydrological cy-cling are more relevant to the human society, prediction of the global monsoon on a long time scale is the urgent task of paleoclimatology. No scientific prediction of the future climate can be properly made without the mon-soon evolution taken into account.

Ding Yihui, Gu Guoxiong and Wang Bin are thanked for critical comments and constructive advice on an early version of this manuscript. Li Qiayu, Tian Jun and Li Jianru are thanked for their assistance in preparation and improvement of the manuscript.

1 Trenberth K E, Stepaniak D P, Caron J M. The global monsoon as

seen through the divergent atmospheric circulation. J Clim, 2000, 13: 3969―3993[DOI]

2 Wang P. Orbital forcing of the low-latitude processes (in Chinese). Quat Sci, Beijing, 2006, 26(5): 694―701

3 Gadgil S. The Indian monsoon and its variability. Ann Rev Earth

Planet Sci, 2003, 31: 429―467[DOI] 4 Chao W C. Multiple quasi equilibria of the ITCZ and the origin of

monsoon onset. J Atmos Sci, 2000, 57: 641―651[DOI] 5 Black D E. The rains may be a-comin’. Science, 2002, 297: 528―529[DOI] 6 Wang B., Ding Q H. The global monsoon: Major modes of annual

variations in the tropics. Dyn Atmos Ocean, 2008, (special issue 2):

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1133

FRO

NTI

ERS

G

EO

LOG

Y

165―183 7 Chao W C, Chen B D. The origin of monsoon. J Atmos Sci, 2001, 58:

3497―3507[DOI] 8 Pierrehumbert R T. Climate change and the tropical Pacific: The sleeping

dragon wakes. Proc Nat Acad Sci, 2000, 97(4): 1355―1358[DOI] 9 Webster P J, Magana VO, Palmer T N, et al. Monsoons: Processes,

predictability, and the prospects for prediction. J Geophys Res, 1998, 103(C7): 14451―14510[DOI]

10 Li J P, Zeng Q C. A unified monsoon index. Geophys Res Lett, 2002, 29(8): 1274, doi: 10.1029/2001GL013874[DOI]

11 Qian W H. Dry/wet alternation and global monsoon. Geophysl Res Lett, 27(22): 3679―3682

12 Wang B, Ding Q H. Changes in global monsoon precipitation over the past 56 years. Geophys Res Lett, 2006, 33: L06711, doi: 10.1029/ 2005GL025347[DOI]

13 Liu Z, Harrison S, Kutzbach J, et al. Global monsoons in the mid-Holocene and oceanic feedback. Clim Dyn, 2004, 22: 157―182[DOI]

14 Chou C, Neelin J D, Su H. Ocean-atmosphere-land feedbacks in an idealized monsoon. Quart J Royal Meteorol Soc, 2001, 127: 1869―

1891[DOI] 15 Chou C, Neelin, J D. Mechanisms Limiting the Northward Extent of

the Northern Summer Monsoons over North America, Asia, and Af-rica. J Clim, 2003, 16: 406―425[DOI]

16 Vera C, Higgins W, Amador J, et al. Toward a unified view of the American monsoon systems. J Clim, 2006, 19: 4977―5000[DOI]

17 Adams D K, Comrie A C. The North American Monsoon. Bull Am Meteorol Soc 1997, 78: 2197―2213[DOI]

18 Zhou J H, Lau, K M. Does a monsoon climate exist over South America? J Clim, 1998, 11: 1020―1040[DOI]

19 Chiang J C H, Koutavas A. Tropical flip-flop connections. Nature, 2004, 432: 684―685[DOI]

20 Ninomiya K. Similarity and difference between the South Atlantic Convergence Zone and the Beiu Frontal Zone simulated by an AGCM. J Meteorol Soc Jpn, 2007, 85(3): 277―299[DOI]

21 Wang X F, Auler A S, Edwards R L, et al. Wet periods in northeastern Brazil over the past 210 kyr linked to distant climate anomalies. Na-ture, 2004, 432: 740―743[DOI]

22 Marchant R, Hooghiemstra H. Rapid environmental change in African and South American tropics around 4000 years before present: a re-view. Earth-Sci Rev, 2004, 66:217―260[DOI]

23 Carvalho L M V, Jones C, Liebmann B. The South Atlantic conver-gence zone: Intensity, form, persistence, and relationships with in-traseasonal to interannual activity and extreme rainfall. J Clim, 2004, 17: 88―108[DOI]

24 Yano J I, McBride J L. An aquaplanet monsoon. J Atmos Sci, 1998. 55: 1373―1399[DOI]

25 Anderson D M, Overpeck J T, Gupta A K. Increase in the Asian Southwest monsoon during the past four centuries. Science, 2002, 297: 596―599[DOI]

26 Torrence C, Webster PJ. Interdecadal changes in the ENSO-monsoon system. J Clim, 1999, 12: 2679―2690[DOI]

27 Molnar P, Cane M A.. El Niño’s tropical climate and teleconnections as a blueprint for pre-Ice Age climates. Paleoceanography, 2002, 17(2), 1021, 10.1029/2001PA000663 [DOI]

28 Kutzbach J. Monsoon climate of the Early Holocene: Climate ex-periment with the Earth’s orbital parameters for 9000 years ago.

Science, 1981, 214: 59―61[DOI] 29 Fleitmann D, Burns S J, Mudelsee M, et al. Holocene forcing of the

Indian monsoon recorded in a stalagmite from Southern Oman. Sci-ence, 2003, 300: 1737―1739[DOI]

30 Gupta A K, Anderson D M, Overpeck J T. Abrupt changes in the Asian southwest monsoon during the Holocene and their links to the North Atlantic Ocean. Nature, 2003, 421: 354―357[DOI]

31 DeMenocal P B. Plio-Pleistocene African climate. Science, 1995, 270: 53―59[DOI]

32 Haug G H, Hughen K A, Sigman D M, et al. Southward migration of the Intertropical Convergence Zone through the Holocene. Science, 2001, 293: 1304―1308[DOI]

33 Wang Y J, Cheng H, Edwards R L, et al. The Holocene Asian mon-soon: Links to solar changes and North Atlantic climate. Science, 2005, 308: 854―857[DOI]

34 Gasse F. Hydrological changes in the African tropics since the Last Glacial Maximum. Quat Sci Rev, 2000, 19: 189―211[DOI]

35 Mayle F E, Burbridge R, Killeen T J. Millennial-scale dynamics of Southern Amazonian rain forests. Science, 2000, 290: 2291―2294[DOI]

36 Baker P A, Seltzer G O, Fritz S C.et al. The history of South American tropical precipitation for the past 25,000 years. Science, 2001, 291: 640―643[DOI]

37 Koutavas A, Lynch-Stieglitz J. Variability of the marine ITCZ over the eastern Pacific during the past 30,000 years: Regional perspective and global context. In: Diaz H F, Bradley R S, eds. The Hadley Circulation: Present, Past and Future. Boston: Kluwer Academic Press, 2004. 347―369

38 Liu Z, Otto-Bleisner B, Kutzbach J, et al. Coupled climate simulation of the evolution of global monsoons in the Holocene. J Clim, 2003, 16: 2472―2490[DOI]

39 Zeng N. Drought in the Sahel. Science, 2003, 302: 999―1000[DOI] 40 Lebranc M, Favreau G, Maley J, et al. Reconstruction of Megalake

Chad using Shuttle Radar Topographic Mission data. Palaeogeogr Palaeoclimatol Palaeoecol, 2006, 239: 16―27[DOI]

41 Giannini A, Saravanan R, Chang P. Oceanic forcing of Sahel rainfall on interannual to interdecadal time scale. Science, 2003, 302: 1027―1030[DOI]

42 Yuan D X, Cheng H, Edwards R L.et al. Timing, duration, and tran-sitions of the Last Interglacial Asian Monsoon. Science, 2004, 304: 575―578[DOI]

43 Kelly M J, Edwards R L, Cheng H, et al. High resolution characteri-zation of the Asian Monsoon between 146,000 and 99,000 years B.P. from Dongge Cave, China. Palaeogeogr Palaeoclimatol Palaeoecol, 2005, 236: 20―38[DOI]

44 Wang Y J, Cheng H, Edwards L, et al.. Millennial- and orbital-scale changes in the East Asian monsoon over the past 224,000 years. Na-ture, 2008, 451: 1090―1093[DOI]

45 Adegbie A T, Schneider R R, Röhl U, et al. Glacial millennial-scale fluctuations in central African precipitation recorded in terrigenous sediment supply and freshwater signals offshore Cameroon. Palaeo-geogr Palaeoclimatol Palaeoecol, 2003, 197: 323―333[DOI]

46 Weldeab S, Lea D W, Schneider R R, et al. 155,000 years of West African monsoon and ocean thermal evolution. Science, 2007, 316: 1303―1307[DOI]

47 Beer J, van Geel B. Holocene climate change and the evidence for

1134 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

solar and other forcing. In: Battarbee R W, Binney H A, eds., Natural Climate variability and Global Warming: A Holocene Perspective. Chichester: Wiley-Blackwell, 2008. 138―162

48 Hodell D A, Brenner M, Curtis J H, et al. Solar forcing of drought frequency in the Maya lowlands. Science, 2001, 292: 1367―1369[DOI]

49 Hodell D A, Brenner M, Curtis J H, et al. Climate change on the Yucatan Peninsula during the Little Ice Age. Quat Res, 2005, 63: 109―121[DOI]

50 Rossignol-Stick M. Africa monsoon, an immediate isotope response to orbital insolation. Nature, 1983, 303: 46―49[DOI]

51 Prell W L. Monsoonal climate of the Arabian Sea during the Late Quaternary: a response to changing solar radiation. In: Berger A L, Imbrie J, Hays J, et al., eds. Milankovitch and Climate. D. Riedel, Hingham, 1984. 349―366

52 Wang P X, Clemens S C, Beaufort L, et al. Evolution and variability of the Asian monsoon system: state of the art and outstanding issues. Quat Sci Rev, 2005, 24: 595―629[DOI]

53 Pokras E M, Mix A C. Earth’s precession cycle and Quaternary cli-mate change in tropical Africa. Nature, 1987, 326: 486―487[DOI]

54 Reichart G J, Lourens L J , Zachariasse W J. Temporal variability in the northern Arabian Sea Oxygen Minimum Zone (OMZ) during the last 225,000 years. Paleoceanography, 1998, 13: 607―621[DOI]

55 Beaufort L, de Garidel-Thoron T, Mix A C, et al. ENSO-like forcing on oceanic primary production during the late Pleistocene. Science, 2001, 293: 2440―2444[DOI]

56 Bender M, Sowers T, Labeyrie L. The Dole effect and its variation during the last 130,000 years as measured in the Vostok ice core. Global Biogeochem. Cycles, 1994, .8: 363―376[DOI]

57 Shackleton N J. The 100,000 year Ice-Age cycle identified and found to lag temperature, carbon dioxides and orbital eccentricity. Science, 2000, 289: 1897―1902[DOI]

58 Brook E J, Sowers T, Orchado J. Rapid variations of atmospheric methane concentrations during the past 110,000 years. Science, 1996, 273: 1087―1091[DOI]

59 Ruddiman W F, Raymo M E. A methane-based time scale for Vostok ice: climatic implications. Quat Sci Rev, 2003, 21: 141―155[DOI]

60 Partridge T C, demonical P B, Lorentz S A, et al. Orbital forcing of climate over South Africa: A 200,000-year rainfall record from the Pretoria Saltpan. Quat Sci Rev, 1997, 16: 1125―1133[DOI]

61 Wang Y J, Cheng H, Edwards R L, et al. A high-resolution abso-lute-dates late Pleistocene monsoon record from Hulu Cave, China. Science, 2001, 294: 2345―2348[DOI]

62 Cheng H, Edwards R L, Wang Y, et al. A penultimate glacial mon-soon record from Hulu Cave and two-phase glacial terminations. Geology, 2006, 34: 217―220[DOI]

63 Overpeck J, Cole J. The rhythm of the rain. Nature, 2008, 451: 1061―1062[DOI]

64 Bar-Matthews M, Ayalon A, Gilmour M, et al. Sea-land oxygen iso-topic relationships from planktonic foraminifera and speleothems in he Eastern Mediterranean region and their implication for paleorain-fall during nterglacial intervals. Geochim Cosmochim Acta, 2003, 67: 3181―3199

65 Ruddiman W F, Raymo M E. A methane-based time scale for Vostok ice. Quat Sci Rev, 2003. 22: 141―155[DOI]

66 Clemens S, Prell W L, Murray D W, et al. Forcing mechanisms of the Indian Ocean monsoon. Nature, 1991, 353: 720―725[DOI]

67 Clemens S C. Prell W L. A 350,000-year summer-monsoon multi-proxy stack from the Owen Ridge, Northern Arabian Sea. Mar Geol, 2003, 201: 35―51[DOI]

68 Sun Y B, Clemens S C, An Z S, et al. Astronomical timescale and palaeoclimatic implication of stacked 3.6-Ma monsoon records from the Chinese Loess Plateau. Quat Sci Rev, 2006, 25: 33―48[DOI]

69 Ruddiman W F. What is the timing of orbital-scale monsoon changes? Quat Sci Rev, 2006, 25:657―658[DOI]

70 Clemens S C, Prell W L. The timing of orbital-scale Indian monsoon changes. Quat Sci Rev, 2007, 26: 275―278[DOI]

71 Kutzbach J E, Liu X D, Liu Z Y, et al. Simulation of the evolutionary response of global summer monsoons to orbital forcing over the past 280,000 years. Clim Dyn, 2007, 30(6), doi: 10.1007/s00382-007- 0308-z[DOI]

72 Weijers J W H, Schefuss E, Schouten S et al. Coupled thermal and hy-drological evolution of tropical Africa over the last deglaciation. Science, 2007, 15: 1701―1704[DOI]

73 Berger A., Loutre M F, Laskar J. Stability of the astronomical fre-quencies over the Earth's history for paleoclimate studies. Science, 1992, 255: 560―566[DOI]

74 Matthews R K, Froelich C. Maximum flooding surfaces and sequence boundaries: comparisons between observations and orbital forcing in the Cretaceous and Jurassic (65―190 Ma). GeoArabia, Middle East Petrol Geosci, 2002, 7(3): 503―538

75 Kump L R. Interpreting carbon-isotope excursions: Strangelove ocean. Geology, 1991, 19: 299―302[DOI]

76 Katz M E, Wright J D, Miller K G, et al. Biological overprint of the geological carbon cycle. Mar Geol, 2005, 217: 323―338[DOI]

77 Wang P X, Tian J, Cheng X R, et al. Exploring cyclic changes of ocean carbon reservoir. Chin Sci Bull, 2003, 48(23): 2537―2548[DOI]

78 Wang P X, Tian J, Cheng X R, et al. Carbon reservoir changes pre-ceded major ice-sheet expansion at Mid-Brunhes Event. Geology, 2003, 31(3): 239―242[DOI]

79 Wang P X, Tian J, Cheng X R, et al. Major Pleistocene stages in a carbon perspective: The South China Sea record and its global comparison. Paleoceanography, 2004, 19, doi: 10.1029/2003PA000991[DOI]

80 Berger A. Support for the astronomical theory of climatic change. Nature, 1977, 269: 44―45[DOI]

81 Pälike H, Norris R D, Herrle J O, et al. The heartbeat of the Oligocene climate system. Science, 2006, 314: 1894―1998[DOI]

82 Pälike H, Frazier J, Zachos J C. Extended orbitally forced palaeocli-matic records from the equatorial Atlantic Ceara Rise. Quat Sci Rev, 2006, 25: 3138―3149[DOI]

83 Woodruff F, Savin S M. Mid-Miocene isotope stratigraphy in the deep sea: High- resolution correlations, paleoclimatic cycles, and sediment preserva-tion. Paleoceanography, 1991, 6: 755―806[DOI]

84 Flower B P, Kennett J P. The middle Miocene climatic transition: East Antarctic ice sheet development, deep ocean circulation and global carbon cycling. Palaeogeogr Palaeoclimatol Palaeoecol, 1994, 108: 537―555[DOI]

85 Holbourn A E, Kuhnt W, Schulz M, et al. Orbitally-paced climate evolution during the middle Miocene “Monterey” carbon-isotope excursion. Earth Planet Sci Lett , 2007, 261: 534―550[DOI]

WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136 1135

FRO

NTI

ERS

G

EO

LOG

Y

86 Cramer B S, Wright J D, Kent D V, et al. Orbital climate forcing of δ 13C excursions in the late Paleocene-early Eocene (chrons C24n-C25n). Paleo-ceanography, 2003, 18(4): 1097, doi: 10.1029/ 2003PA000909[DOI]

87 Salamy K A, Zachos J C. Latest Eocene-Early Oligocene climate change and Southern Ocean fertility: inference from sediment accu-mulation and stable isotope data. Palaeogeogr Palaeoclimatol Pa-laeoecol, 1999, 145: 61―77[DOI]

88 Zachos J C, Kump L R. Carbon cycle feedbacks and the initiation of Antarctic glaciation in the earliest Oligocene. Glob Planet Change, 2005, 47: 51―66[DOI]

89 Wade B S, Pälike H. Oligocene climate dynamics. Paleoceanography, 2004, 19, PA 4019, doi: 10.1029/2004PA001042[DOI]

90 Paul H A, Zachos J C, Flower B P, et al. Orbitally induced climate and geochemical variability across the Oligocene/Miocene boundary. Paleoceanography, 2000, 15(5): 471―485[DOI]

91 Zachos J S, Shackleton N J, Revenaugh J S, et al. Climate response to orbital forcing across the Oligocene-Miocene boundary. Sci-ence, 2001, 292: 274―278[DOI]

92 Billups K, Channell J E T, Zachos J. Late Oligocene to early Miocene geo-chronology and paleocanography from the subantarctic South Atlantic. Pa-leoceanography, 2002, 17(1), doi: 10.1029/2000PA000568[DOI]

93 Lourens L J, Hilgen FJ, Shackleton N J, et al. The Neogene Period. In: Gradstein F M, Ogg J G, Smith A G, eds., A Geologic Time Scale 2004. Cambridge: Cambridge University Press, 2005. 409―440

94 Tiedemann R M, Sarnthein M, Shackleton N J. Astronomic timescale for the Pliocene Atlantic δ 18O and dust flux records from Ocean Drilling Program Site 659. Paleoceanography, 1994, 9: 619―638[DOI]

95 Rossignol-Stick M, Nesteroff V, Olive P, et al. After the deluge: Mediterranean stagnation and sapropel formation. Nature, 1982, 295: 105―110[DOI]

96 Rohling E J. Review and new aspects concerning the formation of eastern Mediterranean sapropels. Mar Geol, 1994, 122: 1―28[DOI]

97 Sachs J P, Repeta D J. Oligotrophy and nitrogen fixation during Eastern Mediterranean sapropel events. Science, 1999, 286: 2485―

2488[DOI] 98 Van Os B J H, Lourens L J, Hilgen F J, et al. The formation of Pliocene

sapropels and carbonate cycles in the Mediterranean: diagenesis, di-lution, and productivity. Paleoceanography, 1994, 9: 601―617[DOI]

99 Wehausen R, Brumsack H J. Astronomical forcing of the East Asian monsoon mirrored by the composition of Pliocene South China Sea sediments. Earth Planet Sci Lett, 2002, 201: 621―636[DOI]

100 Wang P X. Feeling the Earth’s pulse from global monsoon records. Geophys Res Abst, EGU, 2007, 9, 05820

101 Xiong S F, Jiang W Y, Liu T S. Mega-pulses and megacycles in East Asian monsoon variations recorded in Chinese loess-red clay mag-netic susceptibility. Geophys Res Lett, 2006, 33, L18702, doi: 10.1029/2006GL027842[DOI]

102 Ravelo A C, Andreasen D H,.Lyle M, et al. Regional climate shifts caused by gradual global cooling in the Piocene epoch. Nature, 2004, 429: 263―267[DOI]

103 Lawrence K T, Liu Z H, Herbert T D. Evolution of the eastern tropical Pacific through Plio-Pleistocene glaciation. Science, 2006, 312: 79―83[DOI]

104 Lange C B, Berger W H, Lin H-L, et al. The early Matuyama Diatom Maximum off SW Africa, Benguela Current System (ODP Leg 175). Mar Geol, 1999, 161: 93―114[DOI]

105 Cortese G, Gersonde R. Plio/Pleistocene changes in the main biogenic silica carrier in the Southern Ocean, Atlantic Sector. Mar Geol, 2008, 252: 100―110[DOI]

106 Hodell D A, Venz-Curtis K A. Late Neogene History of Deepwater Ventilation in the Southern Ocean. Geochem Geophys Geosyst, 2006, 7, Q09001, doi: 10.1029/2005GC001211[DOI]

107 Shevenell A E, Kennett J P, Lea D W. Middle Miocene Southern Ocean cooling and Antarctic cryosphere expansion. Science, 2004, 305: 1766―1770[DOI]

108 Holbourn A, Kuhnt W, Schulz M, et al. Impacts of orbital forcing and atmospheric carbon dioxide on Miocene ice-sheet expansion. Nature, 2005, 438: 483―487[DOI]

109 Schmieder F, von Dobeneck T, Bleil U. The Mid-Pleistocene climate transition as documented in the deep South Atlantic Ocean: Initiation, interim state and terminal event. Earth Planet Sci Lett, 20001, 79: 539―549

110 Prell W L, Kutzbach J E. The impact of Tibet-Himalayan elevation on the sensitivity of the monsoon climate system to changes in solar ra-diation. In: Ruddiman W F, ed. Tectonic Uplift and Climate Change. New York: Plenum Press, 1997. 172―201

111 Young J A. Physics of monsoons: the current view. In: Fein J S, Stephens P L, eds. Monsoons. New York: John Wiley & Sons, 1987. 211―243

112 Dirmeyer, P A. Land-sea geometry and its effect on monsoon circu-lations, J Geophys Res, 1998, 103(D10): 11555―11572[DOI]

113 Webster P J. Monsoon. Sci Am, 1981. 70―80 114 Kutzbach J E, Gallimore R G. Pangean climates: megamonsoons of

the megacontinent. J Geophys Res, 1989, 9: 3341―3357[DOI] 115 Parrish J T. Climate of the supercontinent Pangea. Journal of Geology,

1993, 101: 215―233 116 Rodwell M J, Hoskins B J. Monsoons and the dynamics of deserts. Q J

R Meteorol Soc, 1996, 122: 1385―1404[DOI] 117 Philander S G H, Gu D, Halpern D, et al. Why the ITCZ is mostly

North of the Equator. J Clim, 1996, 9: 2958―2972[DOI] 118 Crowley T J, Hyde W T, Short D A. Seasonal cycle variations on the

supercontinent of Pangaea. Geology, 1989, 17: 457―460[DOI] 119 Wang P X, Li Q Y. Monsoon: Pre-Quaternary. In: Gornitz V, ed. En-

cyclopedia of Paleoclimatology and Ancient Environments. Heidel-berg: Springer, 2009. 583―589

120 Parrish J T, Peterson F. Wind directions predicted from global circu-lation models and wind directions determined from eolian sandstones of the western United States―A comparison. Sediment Geol, 1988, 56: 261―282[DOI]

121 Lamb H H. Climate: Present, Past and Future.Vol.2. London: Methuen & Co Ltd, 1977. 835

122 Parrish J T. Climate of the supercontinent Pangea. J Geol, 1993, 101: 215―233

123 Fluteau F, Besse J, Broutin J, et al. The late Permian climate. What can be inferred from climate medelling concerning Pangea scenarios and Hercynian range altitude? Palaeogeogr Palaeoclimatol Palaeoecol, 2001, 167: 39―71[DOI]

124 Gordon W A. Distribution by latitude of Phanerozoic evaporate de-posit. J Geol, 1975, 83(6): 671―684

125 Loope D B, Rowe C M, Joeckel R M. Annual monsoon rains recorded by Jurassic dunes. Nature, 2001, 412: 64―66[DOI]

1136 WANG PinXian Chinese Science Bulletin | April 2009 | vol. 54 | no. 7 | 1113-1136

126 Soreghan G S, Soreghan M J, Hamilton M A. Origin and significance of loess in late Paleozoic western Pangaea: A record of tropical cold? Palaeogeogr Palaeoclimatol Palaeoecol, 2008, 268: 234―259[DOI]

127 Dubial R F, Parrish J T, Parrish J M, et al. The Pangaean megamon-soon―― Evidence from the Upper Triassic Chinle Formation, Colo-rado Plateau. Palaios, 1991, 6: 347―370[DOI]

128 Olsen P E. A 40-million-year lake record of Early Mesozoic orbital climatic forcing. Science, 1986, 234: 842―848[DOI]

129 Olsen P E, Kent D V. Milankovitch climate forcing in the tropics of Pangaea during the Late Triassic. Palaeogeogr Palaeoclimatol Pa-laeoecol, 1996, 122: 1―26[DOI]

130 Weissert H, Mohr H. Late Jurassic climate and its impact on carbon cycling. Palaeogeogr Palaeoclimatol Palaeoecol, 1996, 122: 27―43[DOI]

131 Bush A B G. Numerical simulation of the Cretaceous Tethys circum-global current. Science, 1997, 275: 807―810[DOI]

132 Johnson K R, Ellis B. A Tropical rainforest in Colorado 1.4 million years after the Cretaceous-Tertiary boundary. Science, 2002, 296: 2379―2383[DOI]

133 Sewall J O, Sloan L C. Come a little bit closer: A high-resolution climate study of the early Paleogene Laramide foreland. Geology, 2006, 34: 81―84[DOI]

134 Greenwood D R. Eocene monsoon forests in central Australia?: Aust Syst Biol, 1996, 9 : 95―112[DOI]

135 Parrish J T, Ziegler A M, Scotese C R. Rainfall patterns and the dis-tribution of coals and evaporites in the Mesozoic and Cenozoic. Pa-laeogeogr Palaeoclimatol Palaeoecol, 1982, 40: 67―101[DOI]

136 Zhou T. Chinese Natural Geography (in Chinese). Paleogeography, Vol. 1. Beijing: Science Press, 1984. 262

137 Wang P X. Neogene stratigraphy and paleoenvironments of China. Palaeogeogr Palaeoclimatol Palaeoecol, 1990, 77: 315―334[DOI]

138 Liu T S, Zheng M P, Guo Z T. Initiation and evolution of the Asian monsoon system timely coupled with the ice-sheet growth and the tectonic movements in Asia (in Chinese). Quat Sci, 1998, 3: 193―204

139 Guo Z T, Ruddiman W F, Hao Q Z, et al. Onset of Asian deserti-fication by 22 Ma ago inferred from loess deposits in China. Na-ture, 2002, 416: 159―163[DOI]

140 Jia G D, Peng P A, Zhao Q H, et al. Changes in terrestrial ecosystem since 30 Ma in East Asia: Stable isotope evidence from black carbon in the South China Sea. Geology, 2003, 31: 1093―1096[DOI]

141 Chen L X, Liu J, Zhou X J, et al. Impact of uplift of Tibetan Plateau and change of land-ocean distribution on climate over Asia. Acta Meteorol Sin, 2000, 14: 459―474

142 Sun X J, Wang P X. How old is the Asian monsoon sys-tem?―Palaeobotanical records from China. Palaeogeogr Palaeocli-matol Palaeoecol, 2005, 222: 181―222[DOI]

143 Kroon D, Steens T N F, Troelstra S R. Onset of monsoonal related upwelling in the western Arabian Sea. In: Prell W L, Niitsuma N, et al., eds. Proceedings of the ODP. Sci Results, 1991, 117: 257―264

144 Burbank D W, Derry L A, France-Lanord C. Reduced Himalayan sediment production 8 Ma ago despite an intensified monsoon. Nature, 1993, 364: 48―54[DOI]

145 Sepulchre P, Ramstein G, Fluteau F, et al. Tectonic Uplift and Eastern Africa Aridification. Science, 2006, 313: 1419―1423[DOI]

146 Quade J, Cerling T E, Bowman J E. Development of Asian monsoon revealed by marked ecological shift during the latest Miocene in

northern Pakistan. Nature, 1989, 342: 163―166[DOI] 147 Cerling T E, Harris J M, MacFadden B J, et al. Global vegetation change

through the Miocene/Pliocene boundary. Nature, 1997, 389: 153―158[DOI] 148 Ruddiman W F, Raymo M E, Prell W L, et al. The uplift-climate

connection: A synthesis. In: Ruddiman W F, ed., Tectonic Uplift and Climate Change. Plenum, 1997. 471―515

149 Prell W L, Kutzbach J E. Sensitivity of the Indian monsoon to forcing parameters and implications for its evolution. Nature, 1992, 360: 647―653[DOI]

150 An Z S, Kutzbach J E, Prell W L, et al. Evolution of Asian monsoons and phased uplift of the Himalaya-Tibetan plateau since Late Miocene times. Nature, 2001, 411: 62―66[DOI]

151 Wei G. J, Li X H, Liu Y, et al. Geochemical record of chemical weathering and monsoon climate change since the early Miocene in the South China Sea. Paleoceanography, 2006, 21: PA4214, doi: 10.1029/2006PA001300[DOI]

152 Kump L R, Arthur M A. Global chemical erosion during the Cenozoic: Weatherability balanced the budget. In: Ruddiman W F, ed., Tectonic Uplift and Climate Change. New York: Plenum Press, 1997. 399―426

153 Wang P X. Cenozoic deformation and the history of sea-land interac-tions in Asia. In: Clift P, et al., eds. Continent-Ocean interactions in the East Asian Marginal Seas. Geophysical Monograph 149, AGU, 2004, 1―22

154 Ramstein G, Fluteau F, Besse J, et al. Effect of orogeny, plate motion and land–sea distribution on Eurasian climate change over the past 30 million years. Nature, 1997, 386: 788―795[DOI]

155 Cane M A, Molnar P. Closing of the Indonesian seaway as a precursor to east African aridification around 304 million years ago. Nature, 2001, 411: 157―162[DOI]

156 Lamb H H. Climate: Present, Past and Future. Vol.1. London: Methuen & Co Ltd, 1972. 613

157 CLIMAP Project Members. The surface of the ice-age earth. Science, 1976, 191: 1131―1144[DOI]

158 Berger A, Loutre M F, Melice J L. Equatorial insolation: from pre-cession harmonics to eccentricity frequencies. Climate of the Past, 2006, 2: 131―136

159 Park J, D'Hondt S L, King J W, et al. Late Cretaceous precessional cycles in double time: A warm-earth Milankovitch response. Science, 1993, 261: 1431―1434[DOI]

160 Westerhold T, Röhl U, Raffi I, et al. Astronomical calibration of the Paleocene time. Palaeogeogr Palaeoclimatol Palaeoecol, 2008, 257: 377―403[DOI]

161 Pierrehumbert R T. The hydrologic cycle in deep-time climate prob-lems. Nature, 2002, 419: 191―198[DOI]

162 Wang B, Clemens S C, Liu P. Contrasting the Indian and East Asian monsoons: Implications on geological timescale. Mar Geol, 2003, 201: 5―21[DOI]

163 Fedorov A V, Dekens P S, McCarthy M, et al. The Pliocene Paradox (mechanisms for a permanent El Niňo). Science, 2006, 312: 1485―1489[DOI]

164 Soreghan G S, Montanez I P. Special issue on the late Paleozoic Earth system. Palaeogeogr Palaeoclimatol Palaeoecol, 2008, 268: 123―125[DOI]

165 Sewall J O, Sloan L C, Huber M, et al. Climate sensitivity to changes in land surface characteristics. Glob Planet Change, 2000, 26: 445―465[DOI]