We respond to several criticisms of the view that there is a physical linkage between solar activity and the dynamics of the troposphere and lower stratosphere, and we provide further evidence in support of a mechanism for such a linkage involving atmospheric electricity and cloud microphysics. The main criticisms are (1) that the decadal time scale variations in stratified data result from aliasing introduced by the sampling process and are not responses to a decadal time scale physical input; (2) that the observed correlations are due to chance coincidence or an atmospheric periodicity that is not uniquely related to solar variability; and (3) that there are no plausible mechanisms that can amplify one of the weak solar-varying inputs in the region where the correlations are found. We show that the aliasing criticism is inadequate because the real quasi-biennial oscillation departs from an ideal sine wave in a way that reduces aliasing effects to insignificant levels. The nonuniqueness of identification of the 11-year solar cycle as the period of the arctic forcing for the Arctic winter stratospheric temperatures is a problem only for the short 33-year record of polar temperatures; in much longer time series of unstratified climate data the periods of 11 and 22 years are prominent. Highly unique signatures of solar wind forcing of tropospheric dynamics exist on the day-to-day time scale via two independent inputs to atmospheric electricity. These are (1) through changes in tropospheric ion production as a result of solar wind modulation of galactic cosmic rays and (2) through changes in the potential difference between the polar ionospheres and the surface, forced by the solar wind By component.

The product of the cosmic ray flux and the ionospheric potential determines the vertical air-earth electrical current. In the presence of clouds of large horizontal extent, this current determines the rate of polarization charging of the clouds via the accumulation of positive electrostatic charges on droplets near cloud tops. The observed correlations, and theoretical and laboratory results for the effects of electrostatic charges on droplets and aerosols on the rates of ice nucleation, are consistent with the postulate that for certain regions and seasons and atmospheric levels the large-scale atmospheric electrical parameters have significant effects on the rates of initial ice nucleation. In such cases the chain of consequences includes changes in the rates of precipitation, net latent heat release, vertical motions, atmospheric vorticity, and ultimately in the general circulation. Much more work is required before the mechanism can be considered to have a secure basis in laboratory experiment and quantitative atmospheric modeling. ¿ American Geophysical Union 1993