Ozone and temperature responses to solar variability, based on satellite data, have been reported in a companion paper (Keating et. al., this issue). The present is intended to present a theoretical interpretation of this analysis with the purpose of better understanding the chemical behavior of the stratosphere and the coupling between temperature and ozone concentration, when a periodic forcing is applied to the solar ultraviolet (UV) flux. The response of the temperature and of the trace species concentrations, including ozone, to short-term variations in the solar UV irradiance is calculated by a one-dimensional chemical-radiative time-dependent model. The applied solar variability is assumed to be sinusoidal with a period of 27 days (in accordance with the rotation period of the sun) of 13.5 days (when two active regions are on opposite sides of the sun). The amplitude varies with wavelength, which is consistent with observations made by the Nimbus 7 solar backscattered ultraviolet (SBUV) experiment. The maximum ozone sensitivity in the stratosphere appears to be located near 3 mbar. The calculated amplitude and phase of the ozone response are significantly modified when the feedback between ozone and temperature is taken into account. The ozone/temperature coupling tends to modify the ozone phase lag such that, in the upper stratosphere and in the mesosphere, the ozone peak occurs a few days before the UV peak.

Comparison of the model results with the observed ozone and temperature responses, based on satellite data, shows that the theory is consistent in many respects with observations. The calculated time lag of the temperature response, however, is approximately a factor of 2--4 smaller than the time lags derived from the measurements, suggesting evidence for some additionl process not included in the model calculation. Large negative ozone sensitivities and positive temperature responses are predicted in the mesosphere as a result of the absorption by O2 of the solar irradiance at the Lyman α wavelength. The model also shows that the expected variation in the stratosphere nitric acid mixing ratio is a factor of 2 larger than the corresponding opposite variation in the ozone concentration. ¿ American Geophysical Union 1987