新疆西北部萨吾尔山西伯利亚落叶松树轮宽度的气候响应特征

树木生长对气候变化的响应研究不仅能为全球变化背景下森林生态系统的响应提供关键参数，也是树轮气候学研究的基础。本文利用位于新疆西北部萨吾尔山南坡的西伯利亚落叶松树木年轮宽度资料，建立了不同海拔高度的4个树轮宽度年表，采用相关分析和冗余分析法分析了树轮年表特征及其对气候的响应特征。结果表明：（1）不同海拔西伯利亚落叶松树轮宽度年表均具有较高的敏感度和一致性，可能包含较为丰富的气候信息。树轮年表参数有较一致的规律性：平均敏感度、标准差和信噪比随海拔高度降低逐渐增大；反映树轮对环境要素响应滞后效应的一阶自相关系数随海拔高度降低而减小。（2）随着海拔高度的降低，树轮宽度年表与生长季温度（5—9月）的正相关逐渐减弱，直至变为负相关；与反映水分状况的气象要素（降水量、相对湿度和水汽压）由不显著相关变为显著正相关（P<0.01）。萨吾尔山南坡海拔2000~2100m是西伯利亚落叶松树轮宽度对气候响应关系发生转折的关键高度。（3）位于下树线区域西伯利亚落叶松树轮宽度受年降水量（上年7月至当年6月）和当年生长季（5—7月）最高气温共同影响，具备开展区域干旱指数重建的潜力。

Climate response of tree-ring width of Larix sibirica Ledeb. from Sawuer Mountains, northern Xinjiang

Tree-ring growth responses to climate change provide not only crucial information to evaluate the response of forest ecosystem under global change, but the basis for dendrochronological research. Along an elevational gradient, tree ring samples from 4 sites at the southern slope of Sawuer Mountains were collected and the chronologies were developed. The characteristics of chronologies and their responses to climate were also analyzed. The results showed that, the high mean sensitivity and consistency suggested their potential for climate reconstruction. Generally, mean sensitivity, standard deviation and signal-noise-ratio increase with the elevation decreasing. In the four chronologies, the mean sensitivity and percentage of missing ring of HGD is the highest, illustrating the high potential for climate information. At the same time, the first autocorrelation coefficient which reflected lag effect decreased with the decreasing elevation. With the decreasing altitude, the positive correlation coefficient between tree ring width index and growing season temperature decrease, even to negative. The correlation coefficients between tree ring width index and moisture conditions (precipitation, relative humidity and water vapor pressure) change from relation to significantly positive correlation. The elevation at about 2000~2100m is the key transition zone for tree ring width response to climate. Tree-ring width was dominated by both annual precipitation (from July in the previous year to June in the current year) and the maximum temperature in the growing season of the current year, showing its potential for regional drought index reconstruction.