Sunspots, the QBO, and the Stratosphere in the North
Polar Region – 20 Years later
KARIN LABITZKE
1 ∗
, MARKUS KUNZE 1 ;
STEFAN BRÖNNIMANN
2
1
Institute for Meteorology, Free University of Berlin, Germany
2
Institute for Atmospheric and Climate Science, ETH Zürich, Switzerland
Corresponding author: Karin Labitzke, Institute for Meteorology, Free University of Berlin, Carl-Heinrich-Becker-Weg 6-10, 12165 Berlin,
e-mail: [email protected]
∗
submitted to Meteor. Z. on 27 Oct. 2005
Abstract
We have shown in earlier studies the size of the changes in the lower
stratosphere which can be attributed to the 11-year sunspot cycle (SSC).
We showed further that in order to detect the solar signal it is necessary
to group the data according to the phase of the Quasi-Biennial Oscillation
(QBO). Although this is valid throughout the year it was always obvious
that the effect of the SSC and the QBO on the stratosphere was largest
during the northern winters (January/February).
Here we extend our first study (Labitzke 1987) by using additional data.
Instead of 30 years of data, we now have 65 years. Results for the entire
data set fully confirm the early findings and suggest a significant effect of
the SSC on the strenght of the stratospheric polar vortex and the mean
meridional circulation.
Zusammenfassung
In früheren Arbeiten haben wir gezeigt, wie groß der Einfluss des 11jährigen Sonnenfleckenzyklus auf die untere Stratosphäre ist. Um diesen
Einfluss zu isolieren, müssen die Daten nach der Phase der Quasi-Biennial
Oscillation (QBO) sortiert werden. Dies ist während des ganzen Jahres notwendig, aber der Einfluss von QBO und 11-jährigem Sonnenfleckenzyklus
ist am stärksten während des Nordwinters (Januar-Februar).
Für unsere erste Veröffentlichung (Labitzke 1987) standen nur Daten
von 30 Jahren zur Verfügung. Aber inzwischen können wir Ergebnisse zeigen, die auf 65 Jahren basieren und die unsere frühen Arbeiten bestätigen:
1
Der 11-jährige Sonnenfleckenzyklus hat einen signifikanten Einfluss auf die
Stärke des stratosphärischen Polarwirbels und auf die mittlere Meridionalzirkulation.
1
Introduction
Even after two hundred years of research, the relation between solar variability
and Earth’s climate remains a matter of debate in the scientific literature and a
topic of foremost interest to the Earth science community. Effects of solar variability related to the 11-year sunspot cycle are most obvious in the stratosphere,
though still not fully understood (Crooks and Gray, 2005; Matthes et al., 2006).
Labitzke suggested in 1982 that the Sun influences the intensity of the north
polar vortex (i.e., the Arctic Oscillation (AO)) in the stratosphere in winter, and
that the Quasi-Biennial Oscillation (QBO) is needed to identify the solar signal.
Based on these results, Labitzke found in 1987 that a signal of the 11-year
Sunspot Cycle (SSC) emerged when the arctic stratospheric temperatures and
geopotential heights were grouped into two categories determined by the direction
of the equatorial wind in the stratosphere (QBO). This first study was based on
30 years of data (1957-1986), that is barely three solar cycles.
Several publications criticized the short data record and suggested that the
correlations are due to aliasing caused by dividing the data according to the phase
of the QBO (e.g., Teitelbaum and Bauer, 1990; Salby and Shea, 1991).
But even when 20 more years of data became available, the correlations remained stable, see Table 1 (Labitzke, 2006).
For our earlier investigations (e.g., Labitzke and van Loon, 1988; van Loon
and Labitzke, 2000) we used either the FU-Berlin data set (Labitzke and collaborators, 2002) or the NCEP/NCAR re-analyses (Kalnay et al., 1996) starting
with the year 1958, because the available data were considered less reliable before
the IGY (International Geophysical Year 1957/58), especially over the Southern
Hemisphere due to a lack of radiosonde stations.
Here, we extend our work on the Sun-QBO-relationship backwards in time,
incorporating the full data set of the NCEP/NCAR re-analyses which starts in
1948, i.e. adding 10 more years to the data on the arctic winters (1948 till 1957).
Further, we can add 6 more years of data giving the intensity of the stratospheric polar vortex by using statistical reconstructions based on historical
upper-air data (see Section 2). Altogether, 65 years of data are now available,
i.e., 6.5 solar cycles, (see Fig. 4).
2
Data and Methods
The data (30-hPa heights) available for the investigation of the Sun-QBO relationship are listed in Table 1, together with the correlations between the SSC
2
Table 1: Available 30-hPa Height Analyses. For details see text.
1)
2)
3)
4)
Periods of Data
FU-Berlin: 1958 - 1986
NCEP/NCAR 1958 - 2006
early NCEP/NCAR: 1948 - 1957
REC-Index: 1942 - 1947;
Total
Number of years corr.west
29 years
0.78
49 (+20)
0.68
59 (+10)
0.67
65 (+6)
0.69
65 years
0.69 (99%)
and the 30-hPa heights over the Arctic in the west phase of the QBO, see Section 4. In the beginning of our investigations (Labitzke 1987, Labitzke and van
Loon, 1988) only FU-Berlin analyses were available, but limited to the Northern
Hemisphere (Labitzke and Collaborators, 2002).
Later, after very promising comparisons between the Berlin data and the
NCEP/NCAR re-analysis data over the arctic stratosphere (e.g., van Loon and
Labitzke, 1998; Labitzke and Kunze, 2005) we used extensively the global NCEP/NCAR
re-analyses, and with 2006 we have 20 more years since the first publication in
1987, where all data available until 1986 had been incorporated. In this paper
we extend parts of our studies back to the year 1948.
No earlier gridded data are available for 30-hPa height. However, Brönnimann et al. (2005) have compiled large amounts of historical upper air data from
radiosondes, aircraft, and pilot balloons (for a documentation of the data see
Brönnimann (2003), references in Brönnimann et al. (2005, 2006) as well as documents RJ0167 and RJ0168 at http://dss.ucar.edu/docs/papers-scanned/papers/html).
Even though the coverage in the Arctic is not good and most profiles do not reach
the stratosphere, these data can be used to statistically reconstruct monthly 30hPa heights over the North Pole from 1942 until 1947 (REC-Index in Fig. 3).
The method used for reconstructing 30-hPa heights (REC-Index) was a principal component regression, performed in exactly the same way as described in
Brönnimann et al. (2006). The historical upper-air data were continued into the
present, using corresponding data from NCAR/NCEP re-analyses (Kalnay et al.,
1996), and random noise was added to mimick the uncertainty of the historical
data. In addition, surface temperature and SLP data were used (see Fig. 1, top
and Brönnimann et al., 2006). Principal component models were then calibrated
in the 1958-2001 period using FU Berlin 30-hPa heights over the Pole for calibration. Two split-sample validation experiments were performed in which only
1958-1987 or 1972-2001 were used for calibration and the remaining (independent) data for validation (see Brönnimann et al., 2006). This allows calculating
one correlation coefficient (or other skill measure) between the reconstructions
and (independent) FU-B data for each occurring combination of predictor vari3
ables (i.e., for each month and year in the reconstruction period) and for each
validation experiment. For the month of February, which is the focus for further analysis, the twelve correlation coefficients obtained in this way are between
0.41 and 0.81 (median 0.7). We expect this to be a conservative estimation of
the quality, as the calibration period for the final reconstruction was considerably
longer (yielding probably better reconstructions) than the split-sample validation
experiments.
The significance of our results depends on the number of solar cycles available
(Salby and Callaghan, 2004). With these new data we have reached 6.5 solar
cycles and we can now safely say that the results for the northern winters, especially in the west phase of the QBO, with r reaching 0.7, are highly significant;
(r=0.5 ∼ 95%; r=0.66 ∼ 99%).
The QBO is an oscillation in the atmosphere which is best observed in the
stratospheric winds above the equator, where the zonal winds change between
east and west with time. The period of the QBO varies in space and time, with
an average value near 28 months at all levels, see reviews by Naujokat (1986) and
Baldwin et al. (2001).
Because the QBO modulates the solar signal in the stratosphere, and in turn
is modulated by the sun (Soukharev and Hood, 2001; Labitzke 2005), it is necessary to stratify the data into years for which the equatorial QBO in the lower
stratosphere (at about 45 hPa, e.g., Holton and Tan, 1980) was in its westerly or
easterly phase, (QBO data set (starting in 1953) in: Labitzke and Collaborators,
2002).
Prior to 1953, we used zonal wind data from pilot balloon ascents made at
tropical stations (see Fig. 1, top). Ascents that reached the stratosphere were
sparse and irregularly spaced. Nevertheless, they show a clear QBO signal back
to 1950 and earlier during some periods (e.g., a change from a westerly to an
easterly phase is clearly visible in 1945). We averaged the data in bins of 2 km
(16-30 km) in three latitude bands (20o S-5o S, 5o S-5o N, 5o N-20o N). Using corresponding NCEP/NCAR data (1957-2004) we determined the mean annual cycle
as well as the lead or lag with respect to the 45-hPa QBO and corrected the
pilot balloon data accordingly (Fig. 1, bottom). Because the signal-to-noise ratio
rapidly decreases back in time, statistical methods are not suitable and we used
a visual fitting technique that is based on those periods where the eye is able to
identify a QBO signal. Assuming that the QBO has always been there, but with
an uncertain phase due to changes in the period between 24 and 32 months, we
plotted a sine curve with a period of 2.2 years. This curve was piece by piece
shifted and stretched in time (the amplitude was not changed) to visually match
the data from 1942 to 1958, progressing from clear features (1945, 1950-1958) to
more diffuse features. Thereby we attributed more weight to the equatorial latitude band and to the data at 20-24 km than to other data. Only slight stretching
and shifting was necessary. The final curve fits the data reasonably well except
in 1947 and 1949. The curve was then used to define the QBO phase for the
winters 1942 through 1952. Even though the uncertainty of the curve itself must
4
Figure 1: Top: Map showing the historical upper-air and surface series used in this
studies. The sites north of 20o N (solid line) were used for the reconstruction of 30-hPa
heights, the sites in the tropics were used to derive the QBO. Bottom: Zonal wind in
the tropical stratosphere from pilot balloons, visual fit (solid line), and attributed QBO
phase (see text). The dashed line shows the FUB QBO at 50 hPa.
be considered large, most of the early winters are close to the extreme phases of
the QBO and in this sense should be fairly robust with respect to errors in this
curve.
The monthly mean values of the 10.7 cm solar flux are used as a proxy for
variations through the SSC. The flux values are expressed in solar flux units:
1 s.f.u. = 10−22 W m−2 Hz−1 . This is an objectively measured radio wave intensity,
highly and positively correlated with the 11-year SSC and particularly with the
UV part of the solar spectrum (Hood, 2003).
For the earliest years, before the regular measurements of the 10.7 cm solar
flux became available in 1947, we derived the solar flux from a regression between
the sunspot numbers and the flux (1947 through 2005). The correlation between
these two indices is 0.98.
For the range of the SSC, the mean difference of the 10.7cm solar flux between
solar minima (about 70 units) and solar maxima (about 200 units) is used, i.e.,
130 units. Any linear correlation can be represented also by a regression line with
y=a+bx, where x in this case is the 10.7cm solar flux and b is the slope. This slope
5
Figure 2: Vertical meridional sections of the standard deviations for February (left)
and July (right) of the zonal mean monthly mean temperatures (K), upper panels, and
of the zonal mean monthly mean geopotential heights (geopot. m), lower panels, for
the period 1968-2002 (NCEP/NCAR re-analyses).
is used here, multiplied by 130, in order to get the differences of stratospheric
parameters between solar minima and maxima, as presented in Section 4.1 and
4.2 (Labitzke, 2003).
3
Variability in the Stratosphere
The arctic stratosphere reaches its highest variability in winter. In order to judge
the size of the solar signal, Fig. 2 gives an example of the variability of the
stratosphere during the northern winters (February) and summers (July). It is
remarkable that in the lower and middle stratosphere the standard deviations
in the arctic winters are three to four times larger than those in the antarctic
winters; this is due to the fact that the Major Mid-Winter Warmings (MMWs),
i.e. the breakdowns of the arctic polar vortex which create the large variability of
6
the Arctic, do usually not penetrate to the lower stratosphere over the Antarctic.
But the variability is large in the upper stratosphere over the Antarctic where
so-called Minor Mid-Winter Warmings occur frequently, (Labitzke and van Loon,
1972).
In July the variability is low in both hemispheres, because the southern winters
are still relatively undisturbed. A relative maximum of variability is observed on
the equator due to the QBO.
4
4.1
Influence of the 11-Year Sunspot Cycle on the
Stratosphere in Late Winter
The Arctic Stratosphere since 1942
Figure 3 shows in two scatter diagrams the 30-hPa heights over the North Pole
in February when the modulation of the solar signal by the QBO is at its
maximum. The correlations between the 30-hPa heights and the solar cycle are
shown, with the winters in the east phase of the QBO in the left part of the figure,
and the winters in the west phase of the QBO in the right part. The abscissas
indicate the SSC. The correlations are clearly very different in the two groups,
with negative correlations over the Arctic in the east phase of the QBO and large
positive correlations there in the west phase. (The correlation for all years is 0.1,
not shown.) The numbers in the scatter diagrams are the years of the individual
Februaries.
Including the February 2006, the total number of Februaries available is now
65, more than twice the number available in the beginning (Labitzke, 1987; Labitzke and van Loon, 1988) and comprises 7 minima and 6 maxima, see Fig. 4.
As mentioned above, the 20 years after the first publication in 1987 (filled
squares) fit very well into the scatter diagrams (Fig. 3) and confirm the earlier
results. But also the NCEP/NCAR re-analyses from 1948 until 1957 (10 years,
filled circles) as well as the six Februaries from 1942 until 1947 (REC-Index: open
diamonds) fit very well and the size of the correlations (particularly in the west
phase of the QBO) did not change much (see Table 1 and van Loon and Labitzke,
1994 and 2000).
The average height difference (∆ H in Fig.3) between solar maxima and minima is very large in the west phase winters, reaching 704 meters which is almost
2 standard deviations of the interannual variability, cf. Fig. 2.
Figure 4 presents the SSC based on the 10.7 cm solar flux for the period
1942-2006. It is indicated whether a winter (January/February) belonged to the
west or east phase of the QBO. Further, large filled symbols indicate if a MMW
was observed. This can be well decided since 1950. The definition of a MMW is
based on a reversal of the zonal wind over the Arctic at the 10-hPa level. As the
very early data do not reach that high we must rely on comparisons with more
7
Figure 3: Scatter diagrams of the monthly mean 30-hPa geopotential
heights (geopot.km) in February at the North Pole (1942 till 2006), plotted against
the 10.7 cm solar flux. Left: years in the east phase of the QBO (n=29); right: years
in the west phase (n=36). The numbers indicate the respective years; r=correlation
coefficient; ∆ H gives the mean difference of the heights (geopot. m) between solar
maxima and minima. Data: + = 1st period: 1958-1986; filled squares = 2nd period:
1987-2006; filled circles = 3rd period: 1948-1957; diamonds = 4th period: 1942-1947.
(REC-Index: 1942 until 1947; NCEP/NCAR re-analyses: 1948 until 2006)(van Loon
and Labitzke (1994), updated).
recent events. The 30-hPa height values derived for 1947 and 1949 (west phase of
the QBO) and 1942 (east phase) must be compared in Fig.3 with neighbouring
values, e.g., 1958, 1991,1970, 1960 in the west phase group, or 1985, 1963 in
the east phase group, which all represent well known MMWs. Therefore, we
speculate that also during the winters 1942 (Brönnimann et al., 2004), 1947 and
1949 MMWs took place.
There is a very clear tendency for the MMWs in the west phase of the QBO
to occur during solar maxima (solar flux above 150 units), Fig.3: out of 11 cases
10 took place in solar max and none in solar min and this leads to the large
positive correlations with the SSC over the Arctic, as discussed above.
For the MMWs in the east phase of the QBO the situation is less clear, but
more MMWs took place during solar minima (solar flux below 110 units): 10
8
Figure 4: Time series of the 10.7 cm solar flux, 1942 until 2006,(January+February)/2.
Squares denote winters in the west phase of the QBO, circles winters in the east
phase. Large filled symbols characterize the occurrence of Major Midwinter Warmings (MMWs). (Labitzke and van Loon (1990), updated).
out of 15 MMW cases, against 4 in solar maxima. This leads to the negative
correlations in the east phase, see Figs. 3, 5 and 6.
4.2
The Northern Hemisphere Stratosphere since 1948
Figure 5 (left) shows for the Northern Hemisphere the correlations between the
10.7 cm solar flux and the detrended 30-hPa heights in February for the period 1948 - 2006, i.e. 59 years and 6 solar cycles. On the right hand side the
height differences (geopotential meters) between solar maxima and minima are
given. Upper panels: east phase of the QBO, lower panels: west phase of the
QBO. The patterns of the correlations and the respective height differences are
very different in the two phases of the QBO: the correlations are strongly positive (up to 0.68) over the Arctic in the west phase winters, reflecting an intensification of the Brewer-Dobson Circulation (BDC) connected with MMWs and
downwelling/warming over the Arctic in solar maxima. The correlations are negligible outside the Arctic because of dynamically forced upwelling/cooling, which
reaches as far as 30o S,(e.g., Kodera and Kuroda, 2002; Salby and Callaghan,
2004; van Loon et al., 2004; Labitzke, 2005; Matthes et al., 2006).
9
Figure 5: Left: Correlations between the 10.7 cm solar flux (the 11-year solar cycle) and
30-hPa heights in February, shaded for emphasis where the correlations are above 0.4;
upper panel: years in the east phase of the QBO; lower panel: years in the west phase
of the QBO. Right: Respectively, height differences (geopot. m) between solar maxima
and minima, shaded where the height differences are larger than 80 m. (NCEP/NCAR
re-analyses, period: 1948 through 2006); (Labitzke (2002), updated).
In the east phase of the QBO the correlations are weakly negative over the
Arctic but positive over the tropics and subtropics (r = 0.61), connected here
with an enhanced downwelling/warming, i.e. a weakening of the BDC in solar
maxima. This is consistent with our earlier results. It is of interest to compare
these results based on 59 years of data (6 solar maxima and 6 solar minima) with
the results published earlier for the period 1968 until 2003 (Labitzke, 2005), i.e.
36 years (4 solar maxima and 3 solar minima): the patterns are very similar and
the size of the height differences is practically the same. So, one can safely say
that the earlier years since 1948 fit very well into the results obtained before with
10
Figure 6: Vertical meridional sections between 200 and 10 hPa (11 and 32 km) of (left):
the correlations between the detrended zonally averaged, monthly mean temperatures
for February and the 10.7 cm solar flux (shaded for emphasis where the correlations
are larger than 0.4). Right: The respective temperature differences (K) between solar
maxima and minima, shaded where the corresponding correlation on the left hand side
are above 0.4. Upper panels: all years; middle panels: only years in the east phase
of the QBO; lower panels: only years in the west phase of the QBO. (NCEP/NCAR
re-analyses, 1948 through 2006). (Labitzke (2002), updated).
fewer data.
Figure 6 shows for February and for the period from 1948 till 2006 correlations and temperature differences based on zonal mean data, in vertical meridional sections ranging from 200 to 10 hPa and from 90o N to 90o S. The correlations
for all Februaries and for the years in the respective phases of the QBO are given
on the left hand side, and the resulting temperature
differences on the right hand side. There is practically no signal of the SSC
using all years. As discussed above the patterns of the correlations and of the
differences are very different between the two phases of the QBO and the structure
described above for the 30-hPa level is found to be consistent throughout the
11
Figure 7: Same as Fig. 6, but for the geopotential heights.
(geopot. m). (Labitzke (2002), updated).
Height differences
height range investigated here. The whole lower stratosphere is influenced by
the SSC, with the different positive or negative correlations reaching down to the
200-hPa level over the Arctic, and with the opposite correlations reaching as far
as 30o S. The size of the correlations and of the differences are almost the same
as published, e.g., by Labitzke (2005) for the period 1958-2003, with n=46 years.
Figure 7 shows the vertical sections of correlations and differences for the
geopotential heights in February. Again, the pattern of the correlations and of
the height differences are very similar to earlier publications using data starting
in 1958, e.g., Labitzke 2005.
5
Summary
65 years of data are now available for the study of the influence of the SSC on
the arctic stratosphere in winter – this is more than twice the amount of data
12
with which we began our investigations in 1987. Since then we have regularly
updated our results and reached gradually 49 years of data, Table 1. Now we
can add 16 more years going back to 1942. The new data fit very well and it is
rewarding to see that the early results are corroborated by the additional data.
The results for the entire data set fully confirm the early findings and suggest a
significant effect of the SSC on the strength of the stratospheric polar vortex and
on the mean meridional circulation.
Acknowledgements
We thank the members of the Stratospheric Research Group, FUB for professional support, Harry van Loon for many years of close cooperation, and Roy
Jenne (NCAR) for providing the historical pilot balloon data. Tracy Ewen, Andrea Grant, and Thomas Griesser (ETH Zürich) helped in the digitising and
re-evaluation work of the historical upper-air data as well as in the reconstructions. The 10.7cm solar flux data are from the World Data Center A, Boulder,
Colorado. SB was funded by the Swiss National Science Foundation.
References
Brönnimann, S., 2003: A historical upper-air data set for the 1939-1944 period.
Int. J. Climatol., 23, 769-791.
Brönnimann, S., J. Luterbacher, J. Staehelin, T. Svendby, 2004: An extreme
anomaly in stratospheric ozone over Europe in 1940-1942. Geophys. Res. Lett.,
31, L08101, doi: 10.1029/2004GL19611.
Brönnimann, S., G. P. Compo, P. D. Sardeshmukh, R. Jenne, A. Sterin, 2005:
New approaches for extending the 20th century climate record. Eos, Trans AGU
86, 2-7.
Brönnimann, S., T. Ewen, T. Griesser, R. Jenne, 2006: Multidecadal signal of
solar variability in the upper troposphere during the 20th century, accepted for
publication in: Solar Variability and Planetary Climates, SSSI 24, L. Gray, J.
Langen, M. Lockwood, R. M. Bonnet, and Y. Calisesi (eds), Springer; to appear
also as a special issue of Space Science Reviews.
13
Crooks, S. A., L. J. Gray, 2005: Characterization of the 11-year solar signal using a multiple regression analysis of the ERA-40 dataset. J. o. Climate, 18 (7),
996-1015.
Holton, J., Tan, H., 1980. The influence of the equatorial Quasi-Biennial Oscillation on the global circulation at 50 mb. J. Atmos. Sci., 37, 2200-2208.
Hood, L. L., 2003: Thermal response of the tropical tropopause region to solar
ultraviolet variations. Geophys. Res. Lett., 30, No. 23, 2215, doi: 10.1029/2003
GL018364.
Kalnay, E., R. Kanamitsu, R. Kistler, W. Collins, D. Deaven, L. Gandin, M.
Iredell, S. Saha, G. White, Y. Zhu, M. Chelliah, W. Ebisuzaki, W. Higgins, J.
Janowiak, K. C. Mo, C. Ropelewski, J. Wang, R. Reynolds, R. Jenne, J. Joseph,
1996: The NCEP/NCAR 40-year re-analysis project. Bull. Am. Meteor. Soc.,
77, 437-471.
Kodera, K., Y. Kuroda, 2002: Dynamical response to the solar cycle. J. Geophys.
Res., 107, (D24), 4749 doi: 10.1029/2002JD002224.
Labitzke, K., 1982: On the interannual variability of the middle stratosphere
during the northern winters.- J. Met. Soc. Japan, 60, 124-139.
Labitzke, K., 1987: Sunspots, the QBO, and the stratospheric temperature in
the north polar region.- Geophys. Res.Lett.14, 535-537.
Labitzke, K., 2002: The solar signal of the 11-year sunspot cycle in the stratosphere: Differences between the northern and southern summers. J. Met. Soc.
Japan, 80, 963-971.
Labitzke, K., 2003: The global signal of the 11-year sunspot cycle in the atmosphere: When do we need the QBO? Meteorolog. Z. 12, 209-216.
Labitzke, K., 2005: On the solar cycle - QBO relationship: a summary. J. Atm.
Sol.-Terr. Phys., 67, 45-54.
Labitzke, K., 2006: Solar variation and stratospheric response, accepted for publication in: Solar Variability and Planetary Climates, SSSI 24, L. Gray, J. Langen,
M. Lockwood, R. M. Bonnet, and Y. Calisesi (eds), Springer; to appear also as
a special issue of Space Science Reviews.
Labitzke, K. and Collaborators, 2002: The Berlin Stratospheric Data Series; CD
from Meteorological Institute, Free University Berlin.
14
Labitzke, K., M. Kunze, 2005: Stratospheric temperatures over the Arctic: Comparison of three data sets. Meteorolog. Z. 14, 65-74.
Labitzke, K., H. van Loon, 1972: The stratosphere in the Southern Hemisphere,
Chapter 7, 113-138, in: Meteorology of the Southern Hemisphere, Met. Monogr.,
13, No. 35 (C. W. Newton, Ed.).
Labitzke, K., H. van Loon, 1988: Associations between the 11-year solar cycle,
the QBO and the atmosphere. Part I: The troposphere and stratosphere in the
northern hemisphere winter. J.Atmos.Terr.Phys., 50, 197-206.
Labitzke, K., H. van Loon, 1990: The state of the atmosphere on the Northern
Hemisphere at solar maximum, July 1989 - February 1990. Beilage zur Berliner
Wetterkarte, SO 6/90. In: Meteorolog. Abh., FU-Berlin, Band 64, No.4.
Matthes, K., Y. Kuroda, K. Kodera, U. Langematz, 2006: The transfer of the
solar signal from the stratosphere to the troposphere: Northern winter. J. Geophys. Res., 111, doi:10.1029/2005JD006283.
Naujokat, B., 1986. An update of the observed Quasi-Biennial Oscillation of the
stratospheric winds over the tropics. J. Atmos. Sci., 43, 1873-1877.
Salby, M., D. Shea, 1991: Correlations between solar activity and the atmosphere:
An unphysical explanation. J. Geophys. Res., 96, 22.579-22.595.
Salby, M., P. Callaghan, 2004: Evidence of the solar cycle in the general circulation of the stratosphere.- J. Clim. 17, 34-46.
Soukharev, B., L.L. Hood, 2001: Possible solar modulation of the equatorial
quasi-biennial oscillation: additional statistical evidence. J. Geophys. Res., 106,
14.855-14.868.
Teitelbaum, H., P. Bauer, 1990: Stratospheric temperature eleven-year variations: Solar cycle influence or stroboscopic effect? Ann. Geophysicae, 8, 239-242.
van Loon, H., K. Labitzke, 1994: The 10 - 12 year atmospheric oscillation. Review article in: Meteorolog. Z., 3, 259-266, 1994.
van Loon, H., K. Labitzke, 1998: The global range of the stratospheric decadal
wave. Part I: Its association with the sunspot cycle in summer and in the annual
mean, and with the troposphere. J. o. Climate, 11,1529-1537.
15
van Loon, H., K. Labitzke, 2000: The influence of the 11-year solar cycle on the
stratosphere below 30km: A review. Space Sci. Rev., 94, 259-278.
van Loon, H., G. A. Meehl, J. M. Arblaster, 2004: A decadal solar effet in the
tropics in July-August. J. Atm. Sol.-Terr. Phys., 66, 1767-1778.
16