Theoretical predictions of global monthly and seasonally averaged semidiurnal temperatures and winds between 80 and 150 km are presented and compared with seasonal means from measurements at the Arecibo, Chatanika, Millstone Hill, and Sondrestrom incoherent scatter radars. The model values above 100 km are obtained by utilizing a set of tabulated ''Hough mode extension'' functions to extrapolate the (2,2), (2,3), (2,4), and (2,5) propagating tidal modes at 100 km from a spectral model (not valid in the E region due to the way molecular dissipation is parameterized) to higher altitudes and by superimposing a component driven by in situ excitation (important mainly above 140 km).

The monthly simulations demonstrate the close coupling that exists between seasonal and latitudinal structures and argue for an observational program emphasizing adequate seasonal and latitudinal coverage. Despite constraints imposed by the distributions of existing data with respect to month and year, comparisons with the model reveal several trends that appear to be real: (1) the model amplitudes are usually 20--50% smaller than those observed; (2) vertical wavelengths are well-predicted by the model, but model phases generally lead the observations at middle and low latitudes by approximately 2--3 hours at all heights for both winds and temperatures; (3) semidiurnal temperatures at middle and low latitudes, in both the observations and model, indicate significant seasonal changes in the phase of the temperature oscillation, but not in the winds; and (4) the seasonal trends in the model and observations are consistent, with ''summer leading winter'' at Millstone Hill, while the reverse is true at Arecibo.

The sources of the absolute amplitude and phase discrepancies in the model, which are probably connected with the adopted mean wind and thermal excitation specifications, need to be investigated. In addition, a concerted effort must be made within the Coupling Energetics and Dynamics of Atmospheric Regions (CEDAR) program to better declineate the seasonal-latitudinal thermal structure and dynamics of the E region, following the example of the extensive F region observational data base which has been accumulated over the last decade. ¿ American Geophysical Union 1991