During February 2001, a major stratwarm was registered. The wind field of the mesopause region over Collm strongly reacted on this event. The data shown below are partly available from the monthly reports on http://www.uni-leipzig.de/~jacobi/collm/reports/2001. A description of the measurements is found on http://www.uni-leipzig.de/~jacobi/collm/win_e.htm or in Jacobi et al. (1997).

Profiles during February 2001

The major period of the stratwarm occurred during mid-February. To show the reaction of the mesopause region on the warming, profiles of the prevailing winds (zonal v_oz, meridional v_om) and the semidiurnal tidal amplitudes (v_2z) and phases (T_2z, defined here as time of maximum eastward winds) are shown for the time interval of the stratwarm and the rest of the month. It will be shown below that this period was also disturbed by the stratosphere.

The profiles are calculated by fitting mean winds and tidal harmonics to the half-hourly mean winds from the time interval indicated in the legend.  The regression coefficients are taken as height-dependent. This procedure (instead of calculating mean winds and tidal parameters for each height level separately) is necessary because of the quasi-regular change of the reflection height during the course of each day.

The stratwarm leads to a reversal of the zonal prevailing wind, which has frequently been described in literature. The meridional prevailing wind was influenced also and turned to stronger southward  directed winds. This can be explained by a possible residual circulation effect. As can be seen also, the tidal amplitudes strongly decrease, while, however, its vertical structure remains the same. There is also a stratwarm effect on the tidal phases. Sometimes in literature the effect of stratospheric warmings has been described as a sort of  temporarily returning to summer conditions in the upper mesosphere, which describes the easterly winds in the stratosphere and the reversed vertical gradients. This is also seen in the tides: the amplitudes are smaller, and the phase gradient very small, but in the lower part of the height range seen here showing the effect of a different mode.

The wind profiles in Figure 1 show the overall circulation during stratwarm. However, the circulation patterns are highly variable and therefore in the mean profiles in Figure 1 some features are possibly missing. Therefore in Figures 2 - 4 profiles, each based on 4 days of data and calculated assuming quadratically height-dependent coefficients. When interpreting the figures it has to ba taken into account that a database of 4 days only for a regression analysis is a rather small one, and that therefore particularly the uppermost and lowermost wind values may be affected by uncertainties due to a possible insufficient decomposition of the wind field into mean winds and the semidiurnal tide.

Nevertheless, some interesting features can be seen particularly from the zonal prevailing wind field in Figure 2. It is shown there that the wind field is already disturbed in mid January, and that wind reversals, especially at lower altitudes, are already seen before the major stratwarm was really established. As can be seen from the blue lines, however, a wave with a period of about 10 days was present, which leads to easterly (negative) winds also before the strong stratwarm event. After the end of the major warming the stratospheric circulation remains disturbed, but the mesopause region zonal westerlies appear to recover.

The meridional winds in Figure 3 show the strong southerlies during the major stratwarm. The pattern looks somewhat similar to the one of the zonal prevailing wind, but there are no waves that can easily be indentified from the contour plots. The semidiurnal amplitudes (Figure 4) during stratwarm are very weak, but show a modulation, which is connected with the zonal wind variability.

Time series

Some key features of the circulation can be seen easier from time series than from contour plots. Therefore in Figure 4 time series of the wind parameters are shown for two height (98 and 93 km). The data are taken from the contour plots in Figures 2 - 4.

It can be seen from Figure 5 that

• the zonal prevailing wind, starting late January, is very variable. The "stratwarm period" with strong easterlies was preceded by a period of extraordinarily strong westerlies. From the vertical gradients it can be seen that this is part of an oscillation with a period of about 10 days, indicating the presence of a planetary wave,
• the meridional wind, considering absolute values, shows a similar run of its curve. However, the strong positive values and the following turn towards negative (northerly) winds around day number 40 are not accompanied by a change in the vertical gradient. The vertical meridional wind gradients changes not before the decay of the major warming (after day number 50),
• the tidal amplitude gradient is positive during the entire time interval. During stratwarm the amplitudes are smaller than during the rest of the winter. It can be seen, however, that the amplitudes are affected from late January on, which indicates that the stratwarm influences the upper mesopshere during a longer time interval than the one marked by the stratospheric wind reversal,.
• the tidal phase changes during stratwarm to somewhat earlier values. This is typical for summer conditions, and shows the obvious dependence of the semidiurnal tide propagation on the mean stratospheric and mesospheric circulation. Around day number 45, however, the  "winter" conditions with later phases re-establish. This is in close connection with the strong zonal mean winds during that period. Probably during this time interval a different tidal mode is dominant. During late February and early March the phase changes again, already indicating the change to spring conditions.

Waves

In Figure 6 the vertical wind gradients are shown, calculated from the time series in Figure 5. To both the zonal and meridional wind gradients a 7-day averaging was applied and added as solid lines. It can be seen that during the warming period the meridional wind gradient was reversed; this, however, appears not before the major stratwarm was nearly finishing. After the mean zonal wind field has been recovered in late February, the meridional wind gradient turns back to small values again.

On the contrary, the zonal prevailing wind gradient exhibits a distinct 10-day oscillation. This oscillation is not visible in the wind values themselves. This can be seen from the spectra in Figure 7. The zonal prevailing wind shows a 20-day oscillation, while the gradient exhibits an addition 10-day peak. The missing 10-day oscillation in the 93 km can also be inferred from the contour plot (Figure 2), where the strongest variability is found only at greater or lower altitudes, respectively.
 

Inspecting the trend of the zonal prevailing wind gradient in Figure 6, it can be seen that the major stratwarm influence on the mesopause region consists of  two effects: a planetary wave, which is overlaid by a continuous decrease of the wind gradient. The superposition of these two effects leads to the strong turn in zonal prevailing winds visible after day number 40 in Figures 2 and 5, but in fact the stratwarm mean effect is a more smooth and continuous zonal wind change that started in late January, long before the stratwarm definitely turns to a major stratwarm and the stratospheric wind reversal was found.

Conclusions

During February 2001 the mesopause region winds showed a very strong reaction - also in comparison with the effect during previous stratwarm winters - on the major stratospheric warming. During the main phase of the warming  in mid February the zonal as well as the meridional prevailing winds showed strong negative values. This was probably of different origin: the zonal wind reversal was owing to a combination of strong planetary wave activity and a gradual decrease of the prevailing wind, while the meridional wind change probably was due to the residual circulation forced by planetary wave-mean flow interaction at lower altitudes.

The semidiurnal tidal amplitudes were strongly reduced during the warming. However, particularly the phases show a distinct tendency of interaction with the mean zonal winds, indicating a dependence either of the tidal forcing in the stratosphere or of tidal propagation in the mesosphere on the mean circulation.

References

Jacobi, Ch., R. Schminder, and D. Kürschner, 1997: Measurements of mesopause region winds over Central Europe from 1983 through 1995 at Collm, Germany.
Contrib. Atmos. Phys. 70, 189 - 190.
 

Back to Collm results page

Last modification: 28.3.2001