Climatology and variability of historical Soviet snow depth data: some newperspectives in snow - Indian monsoon teleconnections

This study presents the monthly climatology and variability of the historic al soviet snow depth data. This data set was developed under the bilateral data exchange agreement between United States of America and the former Uni on of Soviet Socialist Republics. The original data is for 284 stations for periods varying from 1881 upto 1985. The seasonal cycle of the mean snow d epth has been presented both as spatial maps and as averages over key locat ions. The deepest snow (similar to 80 cms/day) areas are found over Siberia (in Particular over 80 degrees-100 degrees E, 55 degrees-70 degrees N) dur ing March. Over the course of the annual cycle average snow depth over this region changes dramatically from about 10 cms in October to about 80 ems i n March. The variability is presented in the form of spatial maps of standa rd deviation. To investigate the interaction of snow depth with Indian mons oon rainfall (IMR), lag and lead correlation coefficients are computed. Res ults reveal that the winter-time snow depth over western Eurasia surroundin g Moscow (eastern Eurasia in central Siberia) shows significant negative (p ositive) relationship with subsequent IMR. Following the monsoon the signs of relationship reverse over both the regions. This correlation structure i s indicative of a midlatitude longwave pattern with an anomalous ridge (tro ugh) over Asia during the winter prior to a strong (weak) monsoon. As the t ime progresses from winter to spring, the coherent areas of significant rel ationship show southeastward propagation. Empirical orthogonal function ana lysis of the snow depth reveal that the first mode describes a dipole-type structure with one centre around Moscow and the other over central Siberia, depicting similar pattern as the spatial correlation structure. The decada l-scale IMR variations seem to be more associated with the Northern Hemisph ere midlatitude snow depth variations rather than with the tropical ENSO (E l Nino Southern Oscillation) variability.