Tree-ring studies

We analyzed tree ring width and density of  Picea Shrenkiana samples collected by our team in 2001-2003 (Photo 3) (Yurina et al., in press) and by F.Schweingruber in 1995 (International Tree-Ring Data Bank) at the upper tree limit (2800-3100 m) in Terskey Alatoo, Nuura-too, and Moldo-too. All samples inter-correlate well and are used to construct regional ring width (YUSCH) and density (D max) chronologies. Both ring width and maximum density correlate with summer temperature, though in case of D max the correlation is higher (R=0.64 for Chon-Kizil-su met.station) (Fig. 1).

Maximum density is also positively correlated with the ablation of glaciers (e.g. Karabatkak), which support the dependence of D max from summer temperature (Fig. 2). The correlation of D max with mass balance is negative. The correlations with precipitation are insignificant both for ring width and density.

Basing of D max regional chronology we reconstructed May-August Temperature of Chon-Kizil-su met.station for the period 1626-1995. According to this reconstruction (Fig. 1) the years of extremely low May-August temperature occurred in 1989, 1981, 1974, 1972, 1964, 1957, 1940, 1934, 1920, 1869, 1841, 1816, 1813, 1803, 1793, 1783, 1761, 1755, 1751, 1731, 1722, 1698, 1696, 1694, 1676, 1674, 1664, 1650, 1632. All deeps occurred in 20th century coincide with the instrumentally recorded low summer temperatures. 1816 minimum coincides with the “year without summer” after Tambora eruption. The period of low ablation (low summer temperature) in Central Tien Shan occurred in 1641-1655, 1662-1702, 1778-1793, 1810-1818, 1832-1852, 1881-1909, 1951-1977. According to the Wavelet analysis the major cycles in the tree-ring series are close to 200 and 40 years (Fig. 3).

Photo 3. Upper tree limit, Engilchek Valley, 2005. Photo by V. Mikhalenko

Fig.1.  Reconstruction of May-August temperature by maximum late wood density (mean of three chronologies SJ, SK, KAR). A – linear regression and coefficient of correlation of May-August temperature and indices of maximum density chronology. B– May-August temperature 1 – measured at the Chon-Kizil-su met.station and adjusted to the elevation of 3600 m (for the comparison with Tien Shan met.station), 2 – measured at the Tien Shan met. station, 3 – reconstructed by the maximum density on the base of linear regression coefficient. C – May-August temperature, measured at the Chon-Kizil-su met. station and adjusted to the elevation of 3600 m and reconstructed by maximum density for 1626-1995. D – number of samples in the chronology used for the reconstruction.

Fig. 3. Spectral density of the maximum density mean chronology (1), ring width (YUSCH chronology) (2), summer temperature (3), annual sum of precipitation (4) and mass balance of Gregoriev glacier (5),  (wavelet analysis).

← back