What the dynamic configuration is?

We propose a dynamic configuration that may reveal circulation anomalies from the perspective of PWs, that is, the occurrence of eastward-moving wave dissipation in the middle stratosphere and the strong equatorward-propagating wave packet moving westward in the lower stratosphere, accompanied by an enhanced negative forcing to the mean flow.

Why the dynamic configuration is important?

The zonal wind in the tropical stratosphere shows a quasi-biennial oscillation (QBO), which is an important reference for climate prediction and stratosphere-troposphere interaction. However, the periodicity of the QBO was disrupted during the 2015/16 and 2019/20 Northern Hemisphere winters, raising big challenges to its predictability and attracting widespread attention. Previous studies have indicated that planetary wave (PW) activity from mid-latitude in the lower stratosphere is one of the main sources for the reversal of the zonal wind field, our results show that in addition to that, the presence of dissipated eastward-moving waves in the middle stratosphere can be important for the formation of QBO disruptions. The above two wave signs appear successively before the disruption, exert strong negative forcing on the background flow, and together form a dynamic configuration that occurs before the tropical stratospheric circulation anomalies. It may be necessary to take this PW dynamic configuration into account when predicting future QBO disruptions.

How do we find the wave dissipation in the middle stratosphere during QBO disruptions？

Although previous studies showed that PWs accompanied by strong momentum forcing from the mid-latitudes contribute significantly to the two QBO disruptions, the propagation and dissipation process of the PW and specific wave-mean flow interaction mechanism throughout the whole stratosphere are still open questions. Besides, the 2010/11 winter also provides comparable wave momentum flux divergence in the equatorial region with strong meridional EP flux, but lacks the zonal wind reversals, indicating that strong equatorial PW transport in the lower stratosphere alone cannot fully explain this circulation anomaly.

By drawing the result of PW characteristics during the two events compared to climatology,  we find that at 10 hPa, the wave amplitudes stronger than climatology at high latitudes during 2015/16D and 2019/20D all become smaller than climatology at low latitudes. Notably, this change is not observed for the 2010/11 winter. Thus, we hypothesize that in addition to the normal dissipation process during the equatorward propagation of the PW, there may be other mechanisms that accelerate the decline of the wave amplitude and produce additional forcing to the mean flow, which may also be connected to the formation of QBO disruptions. The results show that in the two QBO disruptions, there are obvious eastward-propagating waves in the middle stratosphere, accompanied by significantly dissipation during the equatorward-propagation.

How do we verify this phenomenon？

Using MLS satellite observation data and MERRA2 reanalysis data, the PW  dissipation phenomenon during its equatorial propagation in the middle stratospherie is reproduced simultaneously. It is also proved that the momentum is gradually deposited downward during the process of wave propagation to the equatorial regionand, by including momentum budget analysis in different latitudes.

With the decrease of latitude, it can be seen that the height of EP flux divergence that significantly exceeds the climatology gradually decreases from 6 hPa to 40 hPa, showing that the gradual downward effect of wave forcing is continuous as the latitude decreases. During the period of dissipation of the eastward-moving wave (January in 2015/16D, June and early September in 2019/20D), strong negative forcing in the middle stratosphere over extra-equatorial region (17.5° N/S) can be extended to the equatorial region (7.5° N/S) in the lower stratosphere in the immediate subsequent period. The results wellreflect that the eastward-moving wave dissipation in the middle stratosphere over the extra-equatorial region can affect the lower stratosphere in the equatorial region, and the time period exactly corresponds to the wave packet dissipation in the middle stratosphere and the strong wave transmission in the lower stratosphere.

Conclusion

In historical observations, there has never been such stratospheric atmospheric circulation anomalies in tropics showing as the two QBO disruptions, and the time interval between them is so close. Considering that most of the current circulation models that can generate a reasonable QBO use adjustable parameterizations with large uncertainties to resolve small-scale waves, the failure of this circulation anomaly to be predicted by models is likely due to the fact that abnormal wave activity during the corresponding time period is not adequately represented. Therefore, it is necessary to fully understand the wave activity underlying the two disruption events.

The results show that the PW activity behind the two disruption events is significantly stronger than climatology, and that the zonal wind at low latitudes is favorable for the equatorward propagation of PWs in the lower stratosphere. There are several single strong westward-moving wave packets in the lower stratosphere, accompanied by strong meridional momentum transport. The enhanced easterly winds in the middle stratosphere promote the eastward-moving waves (7–10 hPa) to penetrate to lower latitudes. The strong wind shear may promote the dissipation of eastward-moving waves, producing additional negative forcing on the mean flow. The influence of wave dissipation from the extra-equatorial middle stratosphere can extend to the equatorial lower stratosphere, indicating that in addition to strong westward-moving PWs from mid-latitude in the lower stratosphere, the presence of dissipated eastward-moving waves in the middle stratosphere is also indispensable for the formation of QBO disruptions.