Winter snow depth variability over northern Eurasia in relation to recent atmospheric circulation changes

Abstract

AbstractMean snow depth time‐series for February (1936–2001) over northern Eurasia (incl. Norway, Finland and the former USSR), interpolated into 5 × 5° grid points, are studied using empirical orthogonal function (EOF) analysis. First, five statistically significant rotated PCs are correlated to Northern Hemisphere (NH) teleconnection patterns at the 700 hPa height: North Atlantic Oscillation (NAO), Polar‐Eurasia (Pol), Pacific‐North American (PNA), West Pacific (WP), and Scandinavia (Scand). The impact of the NH circulation modes on snow depth variations is evaluated using the multiple stepwise backward regression (MSBR).Analyses of the snow depth PCs indicate that within the northern Eurasia territory, there are several regions with snow accumulation, respondent to certain circulation modes. PC1 describes low‐frequency snow depth variation to the north from 55 to 60°N between the White Sea and the Lena river basin, and is positively correlated with NAO and negatively—with Scand. MSBR shows that in 1951–1974 the leading role in snow depth variability belongs to Scand. After 1975, Scand has passed over the leading role to NAO. Scand and NAO are also responsible for the surface air temperature changes over the northern Eurasia. Snow depth PC1 and wintertime temperature are closely related to each other. PC2 describes quasi‐decadal snow depth variability over eastern Europe and is negatively correlated with NAO. For the Baltic and White Sea coasts, Fennoscandia, and the center of the East European plain, decrease of snow accumulation, related to a positive NAO phase, seems to be caused by mild winters. For the southwestern and central regions of eastern Europe, negative snow depth anomalies could also be caused by decrease of precipitation associated with the eastward shift of cyclone tracks related to the positive NAO phase. Two regions, where snow depth variations are described by PC1 and PC2, respectively, reveal the border between the opposite recent tendencies of snow accumulation. Copyright © 2007 Royal Meteorological Society