Three-dimensional structure of quasi-stationary circulation anomalies observed in the course of a typical life cycle of an interannual seesaw between the surface Aleutian and Icelandic lows (AL and IL, respectively) is examined by using a reanalysis data set for the last three decades. A diagnosis is applied in a particular framework where the 31-day mean anomalies are regarded as stationary Rossby waves embedded in the zonally varying climatological-mean flow. It reveals that the upward propagation of wave activity into the stratosphere occurs in late winter primarily from the tropospheric anomalies corresponding to the anomalous IL, which develops below the entrance region of the lower-stratospheric polar night jet as a remote influence from the North Pacific. Accordingly, the North Atlantic anomalies exhibit apparent amplification and westward phase tilt with height, whereas the North Pacific anomalies are much more like the external mode. It appears that, in the presence of the zonal wavenumber 1 (k = 1) component of the climatological-mean planetary waves, the polar-night jet over the Pacific is shifted too far north to allow the stationary anomalies associated with the anomalous AL to propagate upward as stationary Rossby waves.
Unlike the predominant signal of the Arctic Oscillation (or annular mode) that alters the intensity of the polar vortex, the seesaw modifies the stratospheric planetary-wave patterns in a modest but signifi-cant manner. In late winter when the AL is weaker than normal associated with a particular phase of the seesaw, the k = 1 component is masked by the enhanced zonal wavenumber 2 (k = 2) component embedded in the intensified polar-night jet. In contrast, the predominant k = 1 component embedded in the relatively weak polar-night jet dominates over the diminished k = 2 component in late winter, when the AL is stronger associated with the other phase of the seesaw. Our examination of seven late-winter major events of stratospheric sudden warming over the three recent decades suggests that the polarity of the AL-IL seesaw might set up a condition of which planetary wave component (k = 1 or k = 2) is more strongly involved in a late-winter warming event.