CLIMATE HISTORY OF THE TIBETAN PLATEAU DURING THE LAST TWO MILLENNIA

The reconstruction of long climatic series spanning the past few centuries and millennia is important both for understanding natural climate variations at those timescales and for estimating anthropogenic influence on the climate system. Our view of past climatic conditions, however, is heavily biased by the data from highlatitude northern and southern hemispheres. To test and validate climatic models on a global scale, it is essential to expand this database to all regions of the Earth's surface. The Tibetan plateau affects the largescale atmospheric circulation over Asia including the monsoon over South Asia and the south China sea. Therefore, the climate history of the Tibetan plateau is an important component in reconstructing the Earth's climate history. In the last decades, many paleoclimatic records, such as ice cores, tree rings, lake sediments and glacier fluctuations covering the last two millennia, have been recovered on the Tibetan plateau. In this paper, the above various proxy data are used to reconstruct the regionalscale temperatures of the Tibetan plateau during the last two millennia. The reconstructions display temperature fluctuations in different parts of the Tibetan plateau. Generally, the northeastern Tibetan plateau experienced distinct warming in 8001100 AD, slight warming in 11501350 AD, and three cold periods during the Little Ice Age between 14001900 AD. In contrast, the southern Tibetan plateau witnessed an early warming in the 68th centuries, an obvious cooling in 8001150 AD, a dramatic warming in 11501400 AD, and the coldest conditions in the 17th century. In addition, the warmer and cooler periods correspond well to glacier fluctuations in the northeastern and southern Tibetan plateau. For the western Tibetan plateau, the δ18O records of the Guliya ice core indicate that the temperatures in the Little Ice Age were higher than that in the middle ages, which would contrast with all other regions of the plateau. On the other hand, a 1400yearlong treering chronology established in the northwest Karakorum, Pakistan, shows that the warmest period was from 800 AD and 1139, whereas an obvious cool period was recorded in 11401874 AD with the coldest period in the first half of the 17th century. More proxy records from the western Tibetan plateau are needed to arrive at firm conclusions. A number of lines of evidence suggested that the Medieval Warm Period was not global in extent, although relatively warm conditions might have prevailed during some parts of the middle ages for some areas of the globe. As described above, in China, the warm climate during this period varied in time and space. Evidence from Dunde, Dulan, Tianjun, Qinghai lake, and other independent proxy records indicates that a warm period between 800 AD and 1100 AD occurred in the northeastern Tibetan Plateau and eastern China, which was also indicated by treering chronology in the Karakorum. Five lines of evidence (Qamdo, Ximen lake, Hidden lake, Chencou lake, and the southern Tibet records)indicate that the southern Tibetan plateau experienced a warm period in 11501400 AD, which was also recorded in the temperature reconstructions for the whole plateau. Therefore, the two warm periods of 8001100 AD and 11501400 AD can be referred to as correlative of the Medieval Warm Period, regardless of their regional differences. Several circulation patternswinter monsoon, west wind drift, and summer monsoon (southwest and southeast monsoon), might have been responsible for these regional temperature differences. 