Time and Space Spectral Analyses of Southern Hemisphere Sea Level Pressure Variability

Abstract

This study examines the time-space structure of the standard deviation of daily summer and winter mean sea level pressure over the Southern Hemisphere, as identified in 20 years of analyses generated by the Australian Bureau of Meteorology and two long simulations with a GCM. The unfiltered variability derived from the operational analyses generally display a deal of zonal symmetry, particularly during the summer period, with maxima in the midlatitudes. The percentage of the temporal variance which is explained by the bandpass and low-pass components is calculated; in January and July the percentage of the variance explained by the bandpass data is maximized between 30° and 60°S and assumes values of typically 25%. In general, the low-pass data account for more of the variance and tends to achieve its maxima in low and high latitudes. The greatest contribution to the low-frequency field comes from the planetary-scale waves, particularly at higher latitudes. The synoptic and small-scale waves are generally found to be the dominant contributors to the variance in the higher-frequency bandpass fields, particularly in the region of the hemispheric storm track. Similar analyses applied to the output of the GCM suggest that, overall, the model performs reasonably well, although the quality of the simulation of the low-frequency variability is inferior to that of the synoptic time scales. The tendency of the model to overpredict the winter daily mean sea level pressure variability in the South Pacific appears to be mostly due to this error in the low-frequency part of the field. These results reveal a considerable difference between the location of cyclone centers and bandpassed mean sea level pressure variability in the high southern latitudes. The model data imply that the maxima of the bandpassed variability tend to be some 30°?40° of longitude to the west and 5°?7° latitude to the north of those of cyclone centers. This serves to underline the dangers and ambiguity of referring to regions of high variability as ?storm tracks.?