Synoptic Scale Weather Patterns Associated with Annual Snow Accumulation Variability in North-Central Greenland

Abstract

Abstract Previous studies on the synoptic forcing of high elevation areas of central Greenland have mostly relied on ice cores, snow pits, mesoscale models, and climate models. In this study, a radar-measured 118-year annual snow accumulation record (1889-2006) along a 375 km traverse between NGRIP and NEEM ice camps in Greenland is used. A Self-Organizing Maps (SOM) algorithm is applied to a reanalysis forecast model (20th Century Reanalysis Data Version 2 (1870 to 2008)) to identify recurring patterns in the sea level pressure (SLP) field that impact meteorological processes and explain annual variations of accumulation over North-Central Greenland for the 118 year period. The SOM algorithm identified 36 representative daily SLP patterns over the North Atlantic region. Synoptic weather patterns shown in these SLP patterns include cyclone splitting, cyclone-blocking, and cyclone tracks indicating changes in cyclone position and cyclone intensity. Based on radar-measured annual snow accumulation, common SLP patterns for wet years (more accumulation) over North-Central Greenland are characterized by low pressure systems surrounding Greenland or cyclones approaching the west coast of Greenland, conveying moisture through a topographically lifted onshore flow; these patterns are mostly associated with negative/neutral NAO index. In dry years (less accumulation), prevailing patterns are characterized by cyclones positioned a long distance away in the Atlantic Ocean south of Greenland, which are mainly associated with a positive NAO index. Extreme wet year prevailing patterns of NGRIP-NEEM traverse southern portion show a distinct departure from the above described North-Central Greenland general patterns by having more frequent positive-NAO days, which is similar to dry and extreme dry years. Model precipitation amounts over North-Central Greenland from the 20th Century Reanalysis Data are found to be overestimated by up to 10 cm/year.