The electron density fluctuation spectra measured during the first and third flights of the STATE experiments (STATE 1 and STATE 3) are compared to present theories for the structuring of mesopheric electrons by neutral atmospheric turbulence. The STATE 1 electron turbulence occurs in a region of vertical electron density gradient just above the mesopause. The spectra show a nearly classic spectral form with evidence for buoyancy, inertial, and viscous subranges. A close inspection, however, reveals that the electron gas is characterized by a much small microscale than can reasonably be expected for neutral turbulence. Although at first very surprising, it is important to note that the very existence of the observed intense summer polar mesospheric 50--MHz radar backscatter virtually requires such a result. We explain the extension of electron structure to smaller scales than the neutral gas by invoking a high effective Schmidt number for the passive scalar electrons.

The dimensionless number is the ratio of the kinematic viscosity of the neutrals to the diffusion coefficient of the electrons and can only be significantly larger than unity if D is anomalously small. We suggest that these conditions may occur and may be due to heavily hydrated positive ions near the region of extremely low temperature at the polar mesopause. The maximum in the STATE 1 electron density fluctuation strength was found just above the mesopause and seemed to be organized with a vertical scale of several hundred meters. This variation in fluctuation spectral density correlates with the outer scale of the electron density profile itself. This suggests that the turbulence in this case was dependent on the phase of a wavelike neutral gas disturbance. The STATE 3 turbulence, on the other hand, was centered on a deep electron density ''bite-out'' and was correlated in strength with the steep gradients in that region. The STATE 3 spectrum does not display a classic inertial subrange, but it does have a viscous range which is nearly identical to the STATE 3 data. Both data sets imply neutral density fluctuations of the order of 1%. ¿ American Geophysical Union 1988