The National Center for Atmospheric Research thermospheric general circulation model (TGCM) for the Earth's thermosphere has been modified to examine the three-dimensional structure and circulation of the upper mesosphere and thermosphere of Mars (MTGCM). The computational framework and major processes unique to a CO2 thermosphere are similar to those utilized in a recent Venus TGCM. Solar EUV, UV, and IR heating along combine to drive the Martian winds above ~100 km. An equinox version of the code is used to examine the Mars global dynamics and structure for two specific observational periods spanning a range of solar activity: Viking 1 (July 1976) and Mariner 6--7 (August--September 1969). The MTGCM is then modified to predict the state of the Mars thermosphere for various combinations of solar and orbital conditions. Calculations show that no nightside cryosphere of the type observed on Venus is obtained on the Mars nightside.

Instead, planetary rotation significantly modifies the winds and the day-to-night contrast in densities and temperatures, giving a diurnal behavior similar to the Earth under quiet solar conditions. Maximum exospheric temperatures are calculated near 1500 LT (≤305 K), with minimum values at 0500 LT (≤175 K). The global temperature distribution is strongly modified by nightside adiabatic heating (subsidence) and dayside cooling (upwelling). The global winds also affect vertical density distribution; vertical eddy diffusion much weaker than used in previous one-dimensional models is required to maintain observed Viking profiles. A solar cycle variation in dayside exospheric temperatures of ~195--305 K is simulated by our Viking and Mariner runs. The dayside heat budget responsible consists of a balance of EUV heating, adiabatic cooling, and molecular thermal conduction, with CO2 cooling having little influence. Solar heating is probably not the only driving mechanism of the Mars upper atmosphere; gravity wave and tidal influence from below will also likely be significant, especially during dust storm periods. ¿American Geophysical Union 1990