The Rice convection model (RCM) is utilized to investigate the electrodynamic coupling between the inner magnetosphere and the thermosphere including the effects of EUV- and convection-driven neutral winds under quasi-equilibrium conditions. It is shown that the parameters determining the coupling are the Pedersen and Hall ''effective winds'', which are the height integrals of the respective conductivity-weighted wind profiles divided by the respective layer conductivities. Their appearance in the RCM is equivalent to a two-slab formulation whereby the integrated Hall conductivity originates in the lower slab, the integrated Pedersen conductivity originates in the upper slab, and the height dependence of the neutral wind is accounted for by assuming different wind vectors for the lower and upper slab. A unique aspect of the study is that the convection-driven winds are included self-consistently and interactively; that is, a steady state wind parameterization is written analytically in terms of the electrostatic potential, which is in turn included in a closed-loop calculation for the electric potential itself.

Simulations are performed from 1400 UT to 1600 UT during the CDAW-6 interval on March 22, 1979, when the cross-cap electric potential attains values of order 140--180 kV. During the early phases of the disturbance when the normal shielding from high latitudes breaks down, the neutral winds do not modify appreciably the disturbance electric fields at middle and low latitudes. As the system approaches a quasi-equilibrium state, the neutral winds play a much more significant role. By comparison with the ''no-wind'' simulation, the fields driven by EUV winds counteract the fields of magnetospheric origin and give the appearance of a shielding effect. The convection driven component of the neutral wind similarly acts to reduce the southward field in the noon sector, but gives rise to an enhancement in the dusk sector field extending to middle latitudes. The parameterized Pedersen effective winds are of order 300 ms-1 and reflect the familiar two-cell pattern with antisunward flow over the polar cap and return flows in the dawn and dusk sectors. These amplitudes and similarity with the ion drift motions reflect the relatively large contribution to the Pedersen effective winds originating in the upper E region and lower F region of the ionosphere. Possibilities for introducing further sophistication into the wind parameterization are discussed, as well as ramifications of the present study on the possible merging of the RCM with the NCAR TGCM to attain a higher degree of self-consistency and reality in modelling efforts. ¿ The American Geophysical Union 1989