Quasi-Stationary Waves in the Southern Hemisphere. Part II: Generation Mechanisms

Abstract

In this Part II the authors investigate the role that Antarctic elevations, the rest of the world orography, thermal forcing from lower latitudes, and the transient eddy component of the flow play on the generation of the quasi-stationary wave field in the Southern Hemisphere. An approach based on the UCLA GCM is followed. Results from a control simulation with full orography and from experiments without the Antarctic elevations and without the rest of the world orography, suggest that the quasi-stationary wave with zonal wavenumber 1 (QS-wave 1) around Antarctica is primarily generated by mechanisms other than the Antarctic elevations. Comparison of a three-dimensional Eliassen-Palm flux vector in the control simulation, and those where the Antarctic elevation and the rest of the world orography are removed, suggests that wave activity propagates both from the subtropics and from polar latitudes. Although in qualitative agreement with results of Part I, the horizontal and vertical structure of these remote forcings is different in the simulations where a more barotropic wave train is generated from lower and polar latitudes. Antarctica is indeed a source of wave activity but unlike observations it is confined to polar regions at tropospheric levels. Additional evidence of thermal forcing was found in an experiment without orographic elevations and zonal asymmetries south of 45°S. It is found that QS-wave 2 is most affected by the zonal asymmetries in sea ice and SST. The effects of the transient component of the flow were also analyzed. The heat transport by the transient eddies in the absence of Antarctic elevations is greater than in the control simulation consistent with a warming of the polar region. Analysis of the contribution by the low-pass and high-pass transients to QS-wave 1 in the control simulation reveals a very different behavior than in Part I. In the control simulation, the low-pass transients and QS-wave 1 are mostly in opposition of phase. High-frequency transients are uncorrelated with QS-wave 1 in all cases. In the experiments without Antarctic elevations or the rest of the world orography, low-pass transients are in phase with QS-wave 1 over high and polar latitudes. In summary, the results of this study suggest that the generation of QS-wave 1 at high latitudes is predominantly from lower latitudes.