Interpretation¶
In the time period from 4 until 3 weeks before the SSW onset the zonal wind is more westerly (eastward) than normal (in the mean for the certain calendar day) at CMA station only. This could be the upper edge of the intensified polar-night jet (PNJ) between stratosphere and mesosphere, which is typical in the time before SSWs. For all other northern hemispheric stations the zonal wind speed is decreasing from around two weeks before the SSW onset and onward. This decrease corresponds to an increased PW activity below and up to 80 km altitude around 12 days before the SSW onset.
At Collm Observatory this decrease is followed by a strong increase in zonal wind speed around one week before the onset of the SSWs. It is apparent in altitudes from 80 to 90 km and corresponds to a strong increase in PW activity in the same altitudes at the same time. With weaker intensity this enhanced PW activity reaches even up to 100 km. This corresponds to the zonal wind increase during this time mentioned above.
This signature is not found at Kiruna and the other stations at higher latitudes, where the decrease of zonal wind speed continues in this altitude range at this time. Possibly these two processes superpose but the later one is stronger in 80 to 90 km altitude and therefore overpowers at Collm so that this feature can be seen. Whether the zonal wind increase triggers the planetary wave activity or vice-versa is not yet clear and further investigations have to be made.
In the time period around the SSW onset latitudinal differences in the zonal wind component can be seen. The stations at lower latitudes, i.e. Collm and CMOR, show a maximum of the zonal wind reverse around 3-5 days after the SSW onset. At higher latitudes (Kiruna, and Sodankyla and their combined dataset) the decrease of zonal wind speed and reverse to easterly directions is already apparent a few days earlier, that is, around the SSW onset or up to 3 days afterwards for Sodankyla.
This observation corresponds to the planetary wave (PW) activity, which is enhanced earlier for Kiruna etc. and later for Collm etc. This leads to the assumption that latitudinal differences are apparent in the PW activity as well.
In the weeks after the SSW onset the zonal wind reverses to westerly directions again in altitudes up to 85 km. The strongest increase in the zonal wind component can be found around 2-3 weeks after the SSW onset. This is very well pronounced at Collm where the zonal wind increase corresponds to an strong increase in PW activity. This observed zonal wind increase is a phenomenon known as the re-establishment of the stratopause, which often takes place in altitudes between 75 and 85 km some weeks after an SSW event.
The meridional wind shows kind of a periodicity at the stations on the NH. It is fluctuating between positive and negative values i.e., northern and southern wind directions. Especially at Collm this is well pronounced, where the fluctuations seem to have a quite even periodicity of around 10 days. Apparent for all NH stations is, that the strongest meridional wind speeds can be found around 10 days before the SSW onset (northern direction) and around the SSW onset (southern direction). From about 2 weeks after the SSW onset, the wind becomes mainly northerly. Also remarkable is, that the anomalies in the meridional wind are very high for all NH stations compared to the climatology values.
Concerning gravity waves the stations on the northern hemisphere show in general an increased level of GW kinetic energy before and around - some stations even short time after - the onset of the SSWs. Especially during the time period of 3-6 weeks before the SSW onset the radar observations show as well increased values of total GW kinetic energy as strongly positive anomalies in an altitude range from 85 to 95 km. This is best pronounced at Kiruna. The positive anomalies after the SSW onset are weaker in magnitude and of less vertical extent.
The reason for the enhanced level of GW kinetic energy some weeks before the SSW onset could be the PNJ, which intensifies typically before the beginning of an SSW event. Because of the increased wind speed the GW filtering is changed in that way, that more waves with larger positive phase speeds are filtered and can’t propagate further upward to the mesosphere. As a result there are less GWs with positive phase speed in the mesosphere. Because of that, the GWs with negative phase speed predominate even more than normal. This leads to the observed enhanced GW activity in higher altitudes (here: 85 to 95 km).
The GWD composite for the combined data of Sodankyla and Kiruna (SES) showed strong differences in amplitudes depending on the time of the measurements. The amplitude may be underestimated for measurements before 2008. Therefore a separate analysis only for dates after 2008 was made for this station. This analysis showed minor differences in the regarded quantities compared to the analysis for all dates: The GWD shows lower values around 20 to 30 days before the SSW onset. The maximum of the PW activity around the SSW onset reaches higher altitudes and is more narrow in time. The zonal wind reverse shortly after the SSW onset is more pronounced. Concerning meridional winds the maximum of the northerly wind around lag -10 is less pronounced; the maximum of the southerly wind is time-shifted to some days before the onset instead of around the onset.
Another interesting feature can be seen in the analysis of the momentum fluxes. While the climatology of the meridional component remains negative, the anomalies are positive in the time before the SSW onset for all NH stations except Kiruna. At Collm Observatory this is broken by a negative anomaly around 10-12 days before the SSW onset, corresponding to a maximum in southward (i.e., “negative”) momentum fluxes. This feature may be attributed to the observed enhanced PW activity at this time.