13:45 〜 15:15
[MIS01-P16] APPLYING TREE-RING ANALYSIS FOR CLIMATE RECONSTRUCTION IN THE ALTAI MOUNTAINS.APPLYING TREE-RING ANALYSIS FOR CLIMATE RECONSTRUCTION IN THE ALTAI MOUNTAINS
キーワード：Blue intensity, Delta Blue intensity, tree-ring width, Larix sibirca, Altai
Recent decades have shown unprecedented climate change in Siberia, with rising temperatures over the global average. These changes are affecting important forest processes such as phenology, productivity and carbon uptake capacity, altogether risking the critical role of Siberian forests in global dynamics. Dendrochronology is a powerful tool to comprehend historical climate fluctuations and decrease the ambiguity in forecasting global climate change. This study aims to evaluate the temperature response of Larix sibirca Ledeb trees in the cold continental upper treeline in the Altai Mountains (Russia), near the Aktru Glacier (2109 m asl) and Seminsky Pass (1666 m asl), by analyzing tree-ring width (TRW), Blue intensity (BI), and Delta Blue intensity (DBI).
Sixty uneven-aged trees (30 per site) were sampled between 2021 and 2022. Wood cores were air-dried and resin was extracted with 96% ethanol in a Soxhlet extractor. Then, cores were polished with a sanding machine and digitalized using an Epson Perfection V800 professional flatbed scanner and Silverfast SE software. The scanner was calibrated with an IT8 calibration target color card. The scanning was carried out in a dark environment to prevent external light interference. RW and BI were measured in CooRecoder version 9.3 software, cross-dated using CDendro, and standardized using a "Detrend" smoothing spline function in R.
The resulting individual series showed high consistency within the forest stands and the study area. The mean correlation coefficients of individual series in Aktru were 0.7 (TRW) and 0.61 (BI), and 0.61 (DBI), while for Semensky, they were 0.75 (TRW) and 0.66 (BI), and 0.65 (DBI). This enabled the creation of a 338 year long tree-ring chronologies for the upper treeline site Aktru, and 213-year-long in Seminsky.
In Aktru, the three proxies were positively correlated with summer temperatures (June-August; Fig. 1). RW showed the strongest relationship with June temperature (r = 0.4), while DBI had a longer temporal window from May to July (0.34-0.71). On the other hand, BI had a lower relationship with temperature compared to DBI. Nevertheless, BI and DBI also showed a significant correlation with August temperature in recent times. For Seminsky, correlations between RW and temperature showed lower strength and were found in July and most pronounced in recent times, whereas BI and DBI showed stronger signals than RW and over a longer period, with maximal signals occurring in May (BI r = 0,48, p< 0.05; DBI r = 0,43, p< 0.05). No significant correlations with precipitation were found at any of the sites.
For the Aktru region, the optimal window for the RW climate response was found between May 25 and July 20, with a temporally stable correlation of 0.54 for the 1940–1969 period and 0.45 for the 1970–2020 period. The optimal window for the DBI daily correlations occurred between May 23 and August 20, with a correlation of 0.75 for the 1971–2020 period. For the Seminsky region, the optimal window was found in the period May 10 – June 25 with a correlation of 0.6 for the 1973-2020 period. The results of the study highlighted the potential of BI- and DBI-based chronologies as proxies for climate in the region, showing stronger climate response than in the commonly used RW-based chronologies. Thus, both parameters are relevant for growing season temperature reconstructions.
Acknowledgments
This work was supported by the Ministry of Science and Education of the Russian Federation [FSRZ-2020-0014]. Climate data was provided by the Tuvan State University “System of experimental bases located along the latitudinal gradient” TSU.
Sixty uneven-aged trees (30 per site) were sampled between 2021 and 2022. Wood cores were air-dried and resin was extracted with 96% ethanol in a Soxhlet extractor. Then, cores were polished with a sanding machine and digitalized using an Epson Perfection V800 professional flatbed scanner and Silverfast SE software. The scanner was calibrated with an IT8 calibration target color card. The scanning was carried out in a dark environment to prevent external light interference. RW and BI were measured in CooRecoder version 9.3 software, cross-dated using CDendro, and standardized using a "Detrend" smoothing spline function in R.
The resulting individual series showed high consistency within the forest stands and the study area. The mean correlation coefficients of individual series in Aktru were 0.7 (TRW) and 0.61 (BI), and 0.61 (DBI), while for Semensky, they were 0.75 (TRW) and 0.66 (BI), and 0.65 (DBI). This enabled the creation of a 338 year long tree-ring chronologies for the upper treeline site Aktru, and 213-year-long in Seminsky.
In Aktru, the three proxies were positively correlated with summer temperatures (June-August; Fig. 1). RW showed the strongest relationship with June temperature (r = 0.4), while DBI had a longer temporal window from May to July (0.34-0.71). On the other hand, BI had a lower relationship with temperature compared to DBI. Nevertheless, BI and DBI also showed a significant correlation with August temperature in recent times. For Seminsky, correlations between RW and temperature showed lower strength and were found in July and most pronounced in recent times, whereas BI and DBI showed stronger signals than RW and over a longer period, with maximal signals occurring in May (BI r = 0,48, p< 0.05; DBI r = 0,43, p< 0.05). No significant correlations with precipitation were found at any of the sites.
For the Aktru region, the optimal window for the RW climate response was found between May 25 and July 20, with a temporally stable correlation of 0.54 for the 1940–1969 period and 0.45 for the 1970–2020 period. The optimal window for the DBI daily correlations occurred between May 23 and August 20, with a correlation of 0.75 for the 1971–2020 period. For the Seminsky region, the optimal window was found in the period May 10 – June 25 with a correlation of 0.6 for the 1973-2020 period. The results of the study highlighted the potential of BI- and DBI-based chronologies as proxies for climate in the region, showing stronger climate response than in the commonly used RW-based chronologies. Thus, both parameters are relevant for growing season temperature reconstructions.
Acknowledgments
This work was supported by the Ministry of Science and Education of the Russian Federation [FSRZ-2020-0014]. Climate data was provided by the Tuvan State University “System of experimental bases located along the latitudinal gradient” TSU.