4.1. The Stratosphere During the Northern Winter
Based on results published in 1982, Labitzke (1987) found that a signal of the 11-year SSC emerged when the arctic stratospheric temperatures and geopoten-tial heights were grouped into two categories determined by the direction of the equatorial wind in the stratosphere (QBO). The reality and significance of using this approach have been confirmed by Naito and Hirota (1997) and Salby and
Callaghan (2004). Meanwhile, 19 more years of data have become available and the correlations remained stable. This phenomenon exists during the whole winter but maximizes during late winter (February) when the planetary wave activity is largest (e.g., van Loon and Labitzke, 2000). An example is given in Figure 2 which shows on the left hand side for the Northern Hemisphere in February maps of the correlations between the solar 10.7 cm flux and the 30 hPa heights, with the winters (Februarys) in the east phase of the QBO in the upper part of the Figure, and the winters in the west phase of the QBO in the lower part. The pattern of correlations is clearly very different in the two groups, with negative correlations over the Arctic in the east phase and large positive correlations there in the west phase. Outside of the Arctic the correlations are positive and strong in the east phase, but very weak in the west phase.
The respective height differences between solar maxima and minima are given on the right hand side of Figure 2. In the east phase of the QBO the stratosphere is colder in solar maxima than in solar minima, and the heights tend to be below normal over the Arctic (about one standard deviation, c.f. Figure 1); they are well above normal towards the equator (up to three standard deviations over Canada). In the west phase, the arctic heights tend to be well above normal (about two standard deviations, c.f. Figure 1) in solar maxima; there are only small anomalies outside the Arctic.
Figure 3 shows for February vertical meridional sections of correlations between the solar 10.7 cm flux and zonally averaged temperatures, as well as the corresponding temperature differences between solar maxima and minima. When all years are used in February, the correlations (top left) and the corresponding temperature differences (top right) are small. But, in the east phase of the QBO, the correlations of the zonally averaged temperatures in the lower stratosphere with the solar data are strongly positive from 60°N to the South Pole in the summer hemisphere, and negative north of 60°N, in the winter hemisphere. These negative correlations are connected with seven Major Midwinter Warmings (MMWs) during East/Min (solar flux below 110, Labitzke et al., 2006); but because also three such MMWs took place in solar maxima, the correlations are weaker than in the west phase of the QBO, see below.
On the right hand side in the middle panel are the zonally averaged temperature differences between solar maxima and minima in the east phase of the QBO which correspond to the correlations on the left side; the shading is the same as that in the correlations where it denotes correlations above 0.4. As the standard deviations in the tropics and subtropics are very small (Figure 1), the temperature differences of 1 to 2 K reflect a solar signal of the order of 1 to 2 standard deviations.
In the west phase of the QBO (Figure 3, bottom), the pattern of the correlations with the 10.7 cm solar flux is completely different from the east phase winters: highly positive correlations are found over the Arctic while they are near zero or weakly negative elsewhere. The large positive correlations over the Arctic are associated with the frequent MMWs which occur at solar maxima when the QBO
Was this article helpful?