Predicting East Coast Winter Storm Frequencies from Midtropospheric Geopotential Height Patterns

Abstract

Relationships between a previously developed U.S. East Coast winter storm (ECWS) climatology and atmospheric circulation patterns derived from the NCEP?NCAR reanalysis data during 1958?97 have been investigated. Composites of normalized height difference, defined as the difference in monthly mean standardized 500-mb height anomalies between active and inactive ECWS seasons, reveal two dipoles, one over the eastern North Pacific Ocean and the other over the central North Atlantic Ocean, which may have seasonal forecasting utility. An October-mean dipolar pattern yielding stronger (weaker) than normal 500-mb geostrophic westerlies over the eastern Pacific (North Atlantic) skillfully predicts, in the data, greater than normal ECWS activity during the following November?April period. Active ECWS seasons follow other combinations of October-mean dipoles in the data, depending upon the phase of the El Niño?Southern Oscillation. On shorter time scales, 7-day-mean ECWS activity was positively (negatively) correlated with antecedent 7-day-mean 500-mb geostrophic westerly wind anomalies over the eastern Pacific (North Atlantic) under El Niño conditions. Otherwise, the lag correlations indicated that 500-mb geostrophic westerly winds decreased (increased) in magnitude with time over the eastern Pacific (North Atlantic) prior to the start of an active 7-day ECWS period. The results suggest that downstream energy propogation from the eastern Pacific to the western Atlantic may influence ECWS activity, and that this in turn may be modulated by the El Niño?Southern Oscillation.