Midwinter Suppression of Northern Hemisphere Storm Track Activity in the Real Atmosphere and in GCM Experiments

Abstract

The seasonal cycle of the North Pacific and the North Atlantic storm track activity is investigated on the basis of daily National Meteorological Center (now known as NCEP) upper-air analyses (1946?89) and of data from the ECHAM3 T42 atmospheric general circulation model. Emphasis is put on the midwinter suppression of the Pacific storm track. This feature of seasonal variability is not sensitive to a particular definition of midlatitude synoptic wave activity, as is shown by applying a common definition of area mean storm track intensity. The suppression is reproduced by the atmospheric model with very similar characteristics. It is attributed to a negative correlation between the storm track intensity and the speed of the subtropical jet at 250 hPa for average zonal winds exceeding the threshold of approximately 45 m s?1, contrasting with a positive correlation below this value. The lack of an analogous behavior over the Atlantic may be assigned to the lower wind speeds there. In a 3·CO2 time-slice experiment with the ECHAM3 model, very intense jet streams occur more often in winter and the suppression becomes more pronounced. At the same time, the level of climatological storm track activity over the Pacific during winter is higher than in the control run. This is explained by the fact that the time-slice experiment produces statistically higher levels of activity for every given jet intensity. The suppression is dominated by a decrease in synoptic-scale wave activity. Two possible reasons for this decrease were investigated but had to be rejected: there is neither a seasonal shift in the energy spectrum to frequencies that are outside the range sampled by the typical bandpass filter, nor evidence that the suppression is attributed to Pacific blocking activity occurring preferably during midwinter.