树轮最大密度记录的吉尔吉斯斯坦天山山区公元1650年以来的7—8月温度变化

July-August Temperature Variability Since A.D.1650 Inferred from Tree-Ring Maximum Density of Picea Schrenkiana in the Tianshan Mountains,Kyrgyzstan

利用吉尔吉斯斯坦天山山区3个采样点的雪岭云杉树轮最大密度数据建立区域树轮最大密度年表(RC)。气候响应分析发现,区域树轮最大密度年表与7—8月平均温度显著相关。在此基础上,利用区域树轮最大密度年表重建了吉尔吉斯斯坦天山山区自1650年以来的7—8月温度变化,重建方程的方差解释量高达45.3%(R2adj=44.7%,N=95,F=77.029)。重建结果显示,在过去的346 a中吉尔吉斯斯坦天山山区7—8月平均温度的异常高温年份为40 a,异常低温年份为46 a,且存在11个低温时段(1650—1654,1662—1678,1693—1703,1779—1794,1801—1805,1811—1819,1834—1854,1882—1910,1917—1923,1952—1975和1986—1992年)和11个高温时段(1655—1661,1679—1692,1704-1778,1795—1800,1806—1810,1820—1833,1855—1881,1911—1916,1924—1951,1976—1985和1994—至今)。突变和周期分析揭示该地区的温度变化受到了火山、太阳活动和海陆气交互作用的共同影响。此外,在过去的346 a来气候以暖干/冷湿为主,近20a来出现了明显暖湿化趋势。

Due to their accurate dating, high resolution,wide distribution and good replication,tree-ring data are a major proxy for paleoclimate research. Tree-ring data from moisture-sensitive trees provide one of the best sources of proxy information that can extend our knowledge of the hydroclimatic variability on annual time-scales beyond the relatively short instrumental records.Although the 20 th century warming has proven true, studies have also indicated that there are large differences in temperature change in different regions. As a result, the efforts to understand climate change over central Asia are in need. In this study,we chose the Tianshan Mountains, Kyrgyzstan as the research region, as the mountain areas is of critical importance to ecology and the livelihood of the people in the arid regions of Central Asia. Previous tree-ring research showed that there is complex relationship between tree growth and climate factors. Therefore, we collected the tree-ring maximum density(MXD) data of Picea Schrenkiana at the upper treeline(2800 m). The most significant relationship is found between the tree-ring maximum density(MXD) of Picea Schrenkiana and July-August temperature.Due to the common responses to climatic influences, we combined the three MXD chronologies into a synthesis chronology(RC). The reliable period is from 1650 to 1995(expressed population signal >0.8). The chronology is correlated positively with temperature, with statistically significant correlations(p<0.01) found in July-August temperature(r=0.673). Based on the relationship, the mean July-August temperature in the research region was reconstructed using the function T =1.033 +11.847RC(N =95,R2 =0.453,F =77.029,P <0.001),where T is the mean July-August temperature and RC is the synthesis MXD chronology.The reconstructed and observational temperature series agree well at both high and low frequency bands. Both sign test and product mean test show that the temperature reconstruction is significant at 0.01 leve1. Relatively cold periods are identified for AD 1650-1654,1662-1678,1693-1703,1779-1794,1801-1805,1811-1819,1834-1854,1882-1910,1917-1923,1952-1975 and 1986-1992. Warm conditions prevailed during AD 1655-1661,1679-1692,1704-1778,1795-1800,1806-1810,1820-1833,1855-1881,1911-1916,1924-1951,1976-1985 and1994-now. Multi-taper method(MTM) spectral analysis revealed some significant low and high frequency cycles(68.3 years, 26.9 years, 17.0 years, 11.3 years, 2.5 years and 2.1 years). In addition, temperature mutations occurred in 1707,1777,1818,1835,1856,1882,1912,1923 and1951,respectively. Some of the temperature variations corresponded to variations in solar activity and major volcanic eruptions. The local climate is mainly characterized by cold/wet and warm/dry over the past 346 years. There is a great potential in reconstructing summer temperature using the MXD data of Picea Schrenkiana trees growing at high elevations in the Tianshan Mountains.