Arctic lake sediments as paleoclimatic archives

• Main Author: Andreev, Andrei
• Format: Conference Object
• Language: unknown
• Published: 2005
• Subjects: Arctic; Arctic Lake; Arga; Nikolay; lena river
• Online Access: https://epic.awi.de/id/eprint/12164/; https://hdl.handle.net/10013/epic.22597

The important data for the understanding of climate fluctuations and the related environment changes since the Late Weichselian were obtained by the study of lacustrine cores from the Russian Arctic. It has been proved that such continuous records provide the high-resolution environmental history up to the 30,000 years. Although, they are mostly not yearly, but decade or centennial resolution records, they are only available data from vast Eurasian regions, providing a base to reconstruct climate variability on time scales from decades to hundreds or thousands of years.Several lake cores from the Lake Lyadhej-To (68°15'N, 65°45'E) situated at the NW rim of the Polar Ural to the Lake Nikolay (73°20N, 124°12E) from Arga Island in the Lena River Delta. Three statistical approaches were used to reconstruct climate from the pollen records: the Best Modern Analogue method (Guiot, 1990), the Plant Functional Types method (Peyron et al., 1998) is based on the biomization concept of Prentice et al. (1996) and Information-Statistical method (Klimanov, 1976, 1984). The methods show comparable results in reconstruction of thermal and hydrological changes during the last 10,000 years. The time of maximum warmth took place during the Early-Middle Holocene. The differences are relatively small compared to the statistical errors of the reconstructions.Surface pollen spectra tests demonstrate that the Best Modern Analogue method reconstructs modern climate variables at Arctic with a reasonably high accuracy (Tarasov et al. 2002). Correlation coefficients between reconstructed and calculated climate variables have suggested that TVII and annual sum of the day temperatures above 5ºC (the so-called sum of growing-degree-days with temperatures above 5ºC) can be reconstructed with reasonably high confidence (R=0.80 and 0.82 respectively). Among the other tested variables P and runoff (difference between P and evaporation (E)) were reconstructed from the modern spectra with relatively high accuracy (R=0.68 and 0.61 respectively).