INTERNATIONAL JOURNAL OF CLIMATOLOGY Int. J. Climatol. 22: 1313–1325 (2002) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/joc.809 RELATIONSHIP BETWEEN SOVIET SNOW AND KOREAN RAINFALL R. H. KRIPALANI, a, * BAEK-JO KIM, b JAI-HO OH c and SUNG-EUII MOON b a Indian Institute of Tropical Meteorology, Pune 411008, India b Meteorological Research Institute, Korea Meteorological Administration, Seoul 156-720, Korea c Pukyoung National University, Busan, Korea Received 24 October 2001 Revised 25 April 2002 Accepted 25 April 2002 ABSTRACT In this study the statistical relationship between winter–spring Soviet snow depth and summer monsoon rainfall over South Korea has been investigated by lag correlation coefficients and composite analysis. Data for the 1966–95 period are used. Results reveal that the winter/springtime snow depth over western Eurasia (over Kazakhstan) is negatively related, whereas the snow depth over eastern Eurasia (over Manchuria–eastern Siberia) is positively related with Korean monsoon rainfall. The dipole correlation pattern suggests that heavy snow over eastern Eurasia and light snow over western Eurasia is favourable, whereas the reverse situation is unfavourable for monsoon activity over South Korea. The NCEP–NCAR re-analyses data reveal that the dipole correlation configuration is indicative of a mid-latitude long-wave pattern with an anomalous ridge over north Asia during the winter prior to a weak Korean monsoon and an anomalous trough prior to a strong monsoon. The re-analyses data further suggest that the position, shape, and strength of the summertime North Pacific subtropical high and the low-level jet over the East Asian sector could be considerably influenced by the snow distribution over Eurasia. Copyright  2002 Royal Meteorological Society. KEY WORDS: Soviet snow depth; Korean monsoon rainfall; mid-latitude circulation; East Asian monsoon circulation 1. INTRODUCTION The prolonged influence of Eurasian–Himalayan snow on the climate conditions over India is well docu- mented. Blanford (1884) was apparently the first to correlate some measure of Himalayan snow with Indian rainfall. With the advent of satellite technology, several studies have examined the relationship between satellite-derived snow cover estimates and monsoon variability over India (e.g. Hahn and Shukla, 1976; Rao et al., 1996; Bamzai and Shukla, 1999). All these studies have shown a negative relationship between the Eurasian snow cover and the Indian monsoon rainfall (IMR), implying that extensive (little) Eurasian snow cover in winter–spring was followed by deficient (excess) IMR. Several groups have reproduced the observed relationships by means of numerical experiments based on general circulation models (e.g. Barnett et al., 1989; Vernekar et al., 1995). In an empirical study, Kripalani and Kulkarni (1999) related the observed Soviet snow depth with IMR. They found that the snow depth variations over western Eurasia surrounding Moscow are negatively related with IMR, whereas the snow variations over eastern Eurasia in central Siberia are positively related, depicting a dipole-type correlation configuration. The primary result of that study was a strong positive relationship of snow depth over central Siberia with IMR. Another important result of that study and earlier studies (Kripalani et al., 1996, 1997) was the identification of localized regions where snow variations–mid-latitude circulation anomalies have an impact on monsoon variability. * Correspondence to: R. H. Kripalani, Indian Institute of Tropical Meteorology, NCL PO, Pashan, Pune 411008, India; e-mail: krip@tropmet.res.in Copyright  2002 Royal Meteorological Society