天山巩乃斯河源区树轮年表特征

利用树轮资料建立了天山巩乃斯河源区5个树轮采点的宽度年表,并对各采点的标准化年表进行特征分析。结果表明：巩乃斯林场5个采点年表的平均敏感度都较低,这与该区的降水丰富、土壤层较厚、坡度较小有关。阿尔先沟、东站、南站A、南站显著水平达到0.01的自相关分别达到9阶、6阶、5阶和6阶,反映该区气候对树木生长影响的＂滞后效应＂非常明显。巩乃斯林缘采点的标准化年表在5个采点中平均敏感度最高、一阶自相关最小,样本对总体的代表性、信噪比、第一主成分方差都为最高,表明该年表的质量最好。标准化年表与新源站月气象资料的相关分析发现,年表与生长季及其前期的降水呈正相关,而与夏季平均最高气温呈显著负相关关系,其相关时段也具有明显的生理学意义。巩乃斯林缘采点的标准化年表所含的气候信息最为丰富,与新源气象站上年7月至当年4月的降水相关系数达0.62,而其余4个点记录更多的记录了低频气候信息,年表经5点平滑处理后与月气象资料相关明显增大。

Tree-ring Chronology Features of Gongnaisi Riverhead Area in Western Tianshan Mountain

Five tree-ring chronologies in Gongnaisi River source area were developed, and the features of five standardization chronologies were analyzed in this paper. It shows that all the mean sensitivity of five chronologies was relatively low, it may relate to the rich precipitation, thick soil layer and flat slope in that area. The autocorrelation significance level or=0.01 for Aerxiangou, Dongzhan, Nanzhan-A and Nanzhan attained 9 order, 6 order, 5 order and 6 order respectively, which reflected the ＂lag effect＂ of climate to tree-ring growth in the above area to be very obviously. Among all the five sampling sites, the standardization chronology of Gongnaisi edge was borne out the highest mean sensitivity, the least 1 order autocorrelation, and its sample representative, signal-noise ratio, and the variance of first principal component were all in tiptop,these indicated the standardization chronology of Gongnaisi edge was best and contained more climate information. By means of correlation analysis on the meteorology data of Xinyuan weather station, the standardization chronology was positive correlate with precipitation and negative with mean maximum temperature during growing season, which had obvious physiological significance. The coefficient between the chronology of Gongnaisi edge and precipitation from last July to April in Xinyuan was 0.62： the climate information of this chronology was most. The chronologies in the rest four sites recorded more low frequency climate information, after smoothing treatment with 5 points, the correlation of these chronologies and monthly meteorology data was increasing obviously.