Low-frequency Variations of the Zonal Mean State of the Southern Hemisphere Troposphere

Abstract

This paper presents an observational study of the low-frequency variation in the Southern Hemisphere troposphere, using the global analyses for 1980-85 provided by the European Center for Medium Range Weather Forecasts. An empirical orthogonal function (EOF) analysis is made for the zonal mean geopotential height at 1000 mb to capture the variation. Based on time series of the second EOF coefficients, which represent the dominant non-seasonal low-frequency variation in the Southern Hemisphere, we define four typical events: negative extreme (D-), positive extreme (D+), negative to positive transition (T+), and positive to negative transition (T-) events.
The variations on a hemispheric scale show a barotropic seesaw pattern with an almost axisymmetric node around 60°S and wavenumber 3 anomalies superimposed on it. Maximum westerlies at 500 mb are located at higher latitudes in D- event (50-60°S) than in D+ event (30-40°S). In association with the location of maximum westerlies, storm activity (defined as temporal variances of the high-pass (≤6 day) height field data) shows that large variation around 50°S occupies the entire latitude circle for a D- event but it is weaker in the western hemisphere for a D+ event.
Composite analyses on a daily basis are made of several physical quantities at 500 mb for the four events. It is found that the latitudinal movement of maximum westerlies is quicker in T+ events than in T- events, in addition to the higher-latitude maximum westerlies in D- events than in D+ events. The zonal mean temperature is colder at high latitudes and warmer at middle latitudes in D- events than in D+ events, corresponding to the stronger polar vortex resulting from the steeper temperature gradient and more vigorous storm activity in D- events than in D+ events.
The eddy momentum flux plays an important role especially during transition events; there is large equatorward transport of momentum at high latitudes around the key day of a T+ event and larger poleward transport at middle latitudes around the key day of a T- event. The heat flux seems to play a less important role. The acceleration of mean zonal winds is mainly determined as a residual between the momentum flux convergence and the Coriolis force.