• Introduction
• Model
• Chemical species in the AGCM
• Results
• Conclusions
• References

FIGURES

• Figure 1
• Figure 2a, 2b
• Figure 3a, 3b
• Figure 4
• Figure 5
• Figure 6
• Figure 7

Introduction

The sun has a 11-year sunspot cycle. Its related solar constant shows a small variation (0.1%) which is more significant in the ultraviolet spectrum than in the visible. Recently, the relationship between solar activity cycles observed summer changes of stratospheric temperature and ozone concentrations has been studied.
These observations reveal indeed that the difference between the ozone content at the solar minimum and the one at the solar maximum for an 11-year cycle differs by more or less 4%. Concerning the temperature in the stratosphere a difference of 2.0K is noticeable [Hirooka (1994)].
On the other hand, Kodera(1995) reported that in November, westerly anomalies due to the sunspot cycle appear in the midlatitudes of the upper stratosphere. These anomalies then shift poleward and downward in the following months until February.
During the same period, easterly anomalies develop first in the lower latitudes to finally intrude into the polar regions of the stratopause. As a result these easterly and westerly abnormal winds form a rotating dipole in the mid-high latitudes in winter.

The purpose of the present study is to investigate the causes of these phenomena by the use of a General Circulation Model with chemistry.

Model

Two experiments of 21 years' duration were completed to simulate the solar maxima and solar minima events respectively. It should be mentioned that in both experiments the simulated extremum was kept constant all along the simulation.
The Center for Climate System Research / National Institute for Environmental Studies chemical model [Takigawa et al. (1999)] with extended module to take account of heterogeneous process has been used for both simulations. The model resolution is T21 with 34 vertical layers from surface to about 80km altitude. Solar flux data were obtained from Lean et al. (1997). The first three years of integration are ignored in analysis of the results.

Chemical species in the AGCM

Ox?ANOx?AHOx?AClx?ANOy?ACly
N2O?AH2O?ACH4?AH2O2?AHNO3
HNO4?AN2O5?AClONO2?ACCl4
CFCl3?ACF2Cl2?ACH3Cl?AHCl?ACl2
HOCl?AClONO2?APSC scheme

The amplitude of the changes in the incident irradiance from quiet to active sun conditions varies with wavelength in the spectral range of the Schmann-Runge bands. Nicolet (1984) suggested a semi-empirical relation of this variation for the photolysis rate of molecular oxygen by:

J(O2;SRB active sun)=(1.11?}0.04)J(O2;SRB quiet sun)

The coefficient at zero optical depth J?‡(H2O;Lyƒ¿) varies from 4.0 to 6.5 x 0.000001/s according to the level of the solar activity by Nicolet (1981).

Results

The solar minimum and maximum discussed in the results represent each an
averaged value of the 18years integration.

Figure 1

Figure 2a, 2b

Figure 2a and Figure 2b represent the annual mean temperature and ozone content differences between solar maxium and solar minimum for a tropical latitude.

Figure 3a and Figure 3b are the annual mean temperature and ozone differences for the zonal mean.

Figure 4 is zonal wind difference between solar maximum and solar minimum. We can see that westerly and easterly anomalies appear in the mid-high latitude each month.

In Figure 5 and Figure 6, westerly and easterly anomalies appear at latitude 70N every month.

Figure 6 shows that the anomalies shift from south to north with time.


Figure 7 shows that EP-flux anomalies direction changing every month. The scaling factor of Fz is 125.

Conclusions

The two experiments allowed showing that the difference of the annual mean temperature and ozone amount between at the solar maximum and at the solar minimum can be noticeable in the upper stratosphere and mesosphere (Fig.2).

Over the higher latitudes however, such a difference is less clear because of the high standard deviation related to small insolation in this region in winter. Hirooka (1994) 's analysis of measurement made at 1 hPa showed a local temperature difference of 2K, while the model simulated a difference temperature of 1K. Moreover we have shown that zonal wind anomalies in the Northern Hemisphere during winter have about 2-month periodicity (Fig.5 and 6). Those anomalies appear in the mid-high latitudes and shift poleward with time from October to March.

The change in the EP-flux anomalies highlights the influence of the solar cycle on the magnitude of the planetary waves (Fig.7). However, the factors involving this magnitude variation are not yet known.
A model taking the parametrization of the photodissosiation rates of NOx, and HOx into account will be implemented in the future to improve the results. An insight of the factors influencing the variation of the amplitude of the planetary
waves will also be investigated.

References

Hirooka, T., Proceeding of conference in Meteorological Society of Japan,1994

Kodera, K., J.Geophys.Res.,100,14,077-14,087, 1995

Lean, J.L., Rottman, G.J., Kyle, H.L., Woods, T.N., Hickey, J.R. and Puga, L.C., J.Geophys.Res.,102,29,939-29,956, 1997

Nicolet, M., J.Geophys.Res.,86,5203, 1981

Nicolet, M., J.Geophys.Res., in press, 1984

Takigwa, M., Takahashi, M. and Akiyoshi, H., J.Geophys.Res.,104,14,003-14,018, 1999

We thank Dr. Lean for providing 10.7cm solar flux data.