Relationships between circulation strength and the variability of growing-season and cold-season climate in northern and central Europe

Abstract

This paper examines variability in surface climate and the relationship with two circulation indices using recently produced 200–250 year long European daily and monthly climatic series. Assessments are made of the influence of circulation indices (zonal over western Europe and average pressure over central Europe) on regional temperature, precipitation and drought series for northern and central Europe. Changes in the circulation are shown to be very important during the winter in northern Europe, but only over western and central Europe in summer. Circulation indices in winter, however, only explain up to 25% of surface-climate variability and, except for the British Isles, do not explain any decadal and longer timescale variability in summer. Extreme temperatures and growing-season and frost-season durations and degree-day series are calcu lated for four sites (Central England, Stockholm, Uppsala and St. Petersburg). Degree-day counts are shown to be very strongly correlated (r > 0.9) to average extended summer (‘growing’ season, May to September) and extended winter (‘frost’ season, October to April) temperatures. Durations of the growing and frost seasons are only weakly correlated (r ~ 0.2–0.4) with the appropriate extended seasonal temperatures or degree-day counts. Warmer growing/frost seasons, therefore, need not necessarily be longer/shorter. Counts of extremely warm days in each year are more highly correlated with the growing-season mean temperature than with coldseason temperature. Similarly, counts of extremely cold days in each year are more highly correlated with cold-season temperature than with growing-season temperature. Both the growing-season temperature and the corresponding degree-day counts clearly show warmer temperatures over all series in Fennoscandia before 1860. Over central Europe, and for the Central England temperature series, growing seasons now are of similar warmth to those before 1860. Annually averaged temperatures across the three regions indicate long-term warming which is, however, almost entirely the result of warming during the October to April period.