We present the E region neutral winds and temperature observed by the Arecibo incoherent scatter radar (ISR) and a collocated spectrometer during the January 20--30, 1993, period and compare them with the thermosphere/ionosphere/mesosphere electrodynamics global circulation model (TIME-GCM) predictions. The model is most successful in predicting the meridional wind phase and magnitude throughout the altitude range of 95--145 km. Although the model can also predict the overall pattern of zonal wind and temperature, its daytime eastward wind exceeds the observations by about 35 m/s at 105 km and its temperature deviation from mean is typically 3 times smaller than the observations. The overestimation of the eastward meanflow in the model is likely associated with an overestimation in gravity wave forcing which prevents the propagation of the diurnal tide to 100 km. The diurnal tide, which is quite substantial at 100 km from the radar measurement, deposits its momentum to enhance the westward meanflow [Miyahara and Wu, 1989>. One reason for the smaller simulated temperature fluctuation is that the current model does not include any semidiurnal forcing at the lower boundary (30 km). According to a recent study by Hagan [1996>, water vapor absorption of solar near-infrared radiation in the troposphere can have a very large effect on the tides in the lower thermosphere. We further note that the observed semidiurnal tide experiences a fast growth rate at 105 km and saturates at slightly above 110 km, while the model prediction shows a much smaller growth rate in this region despite a much longer vertical wavelength. This suggests that the model may overestimate the molecular/eddy viscosity in the turbopause region. During the January 1993 period, the observed mean mesopause temperature, 160 K, is about 30 K colder than the model prediction and current prevailing beliefs. Although such a cold mesopause at a tropical latitude needs to be verified by more experiments, it cannot be precluded theoretically due to large uncertainties in radiative and dynamical heating/cooling rates, most particularly the radiative cooling rate from CO2.¿ 1997 American Geophysical Union