Two sounding rockets launched from Peru as part of Project Condor confirm and extend a number of previous rocket measurements of the wave number spectrum of equatorial spsread F irregularities. Other papers in this series present investigations of the intermediate- and long-wavelength regimes; here, we concentrate on wavelengths less than 100 m. The Condor density fluctuation spectra display a break at a wavelength near 100 m, identical to that found in the PLUMEX experiment (Kelley et al., 1982b). The Condor data also confirm a subrange in which the density and the wave potential obey the Boltzmann relation--a strong indication of the presenc of low-frequency electrostatic waves with finite wavelength parallel to the magnetic field, perhaps low-frequency drift waves as proposed by Kelley et al. (1982b). The Condor data are also consistent with the previous conjecture that drift waves only exist above 300 km altitude. To investigate the difference in spectra observed over two altitude ranges, we have fit to the data a form for the power spectrum taken from Keskinen and Ossakow (1981). The fitted spectrum, along with empirically determined growth and dissipation rates, is next used to calculate the energy pumped into the spectrum at long wavelengths as well as the energy dissipated at shorter wavelengths. It is found that the energy is balanced by classical collisional effects in the low-altitude case, but energy balance in the high-altitude case requires an enhanced dissipation of about 500 times that due to classical diffusion. This result again implies the existence of electrostatic waves, which cause anomalous diffusion due to two mechanisms: nonambipolar diffusion and wave-particle interactions. We propose that the enhanced diffusion coefficient should also be consistent with a convective transport model for anomalous diffusion.

In this model, the long-wavelength portion of the spectrum consists of coherently steepened (k-2) Rayleigh-Taylor generated structures, while the short wavelengths are a turbulent cascade of low-frequency drift waves which cause convective transport. We have used the measured electric field spectrum in the drift wave regime to calculate the effective diffusion due to stochastic E¿B scattering. The resulting diffusion coefficient is comparable to that indicated by the observations, although its value depends on the outer scale selected for the turbulent spectrum. Furthermore, the experimental data show that the wavelength at which the drift wave cascade originates varies inversely with the power spectral density. If stochasic E¿ scattering is the dominant enhanced diffusion mechanism, then in order to duplicate the observed behavior, the turbulent portion of the spectrum must extend to longer scales than the wavelength at which the drift waves are excited. Thus the model is consistent with but does not uniquely imply an inverse cascade of drift wave turbulence in equatorial spread F.