关于青藏高原2485年温度的季节和空间代表性问题

SEASONAL AND SPATIAL REPRESENTATIVENESS OF THE TREE-RING BASED 2485-YEAR ANNUAL MEAN TEMPERATURE RECONSTRUCTION IN THE NORTHEASTERN TIBETAN PLATEAU

利用树轮宽度重建的青藏高原东北部2485年年均温度曲线一经发表就引起了学界很大关注。本文就该温度曲线的空间代表性问题进行了再次分析,该研究的样本来自青海乌兰和都兰地区,有一定的空间覆盖度。尽管宽度年表用于年均温度重建,但是我们也指出,它与冬半年(上年9月～当年3月)平均温度也显著高相关(r=0.686,p<0.0001)。重建温度曲线不仅与青藏高原东北部地区,而且与我国中北部地区的气象台站同期平均温度记录显著相关。本文进一步通过重建曲线与不同时空尺度CRU TS3.1网格点资料的相关分析,证明2485年年均温度曲线具有一定的全球空间代表性。

Tibetan Plateau（TP） is widely acknowledged as one of the most sensitive regions in the world to global climate change,and it has attracted lots of foreign and domestic dendroclimatologists to carry out work in it. Using a combined tree-ring chronology from living trees and archaeological wood, collected from Wnlan and Dulan, about 100km apart from each other in the Qinghai Province, northeastern TP, we have reconstructed the annual mean temperatures（ last January to December）variability for the past 2485 years in 2009. This reconstruction could alsobe considered as a winter-time（last September to current March） temperature series. As we elaborated in the 2009 paper,this 2485-year temperature reconstruction could not only disclose the temperature variation in the mid-eastern Tibetan Plateau,but also for large parts of north-Central China, on the basis of statistical analysis. Furthermore, it also closely corresponded to those temperature series of the entire northern Hemisphere over long time scales. It has drawn great attention since the publication of this tree-ring based 2485-year annual mean temperature reconstruction in 2009. However, it also leads to some controversy about the spatial representativeness of this temperature series. In view of this reason, the present paper further analyzed the spatial and seasonal representativeness of this 2485-year temperature series, to elucidate the following points. First,the studied tree-ring samples were collected from Wulan and Dulan of Qinghai Province, suggesting certain space coverage. Second,the tree-ring width chronology could be used to reconstruct both the annual mean temperature and the winter half year mean temperature. Third, except for the statistical analysis in the published paper in 2009, in this paper, we analyzed the spatial correlations between the reconstructed temperature series and the annual mean temperature grid dataset,as well as the winter half year mean temperature grid dataset from CRU TS3. 1 in different temporal and spatial scale. The reconstructed temperature series showed high correlations with the concurrent mean temperatures not only in the north eastern Tibetan Plateau,but also in the large parts of north-Central China. In the global scale, it also revealed a certain global representativeness of this temperature series.