The National Center for Atmospheric Research (NCAR) thermospheric general circulation model (TGCM) is used to calculate the time-dependent thermospheric response to the May 30, 1984, annular solar eclipse. The path of maximum obscurity begins at sunrise in the Pacific Ocean near 2¿N and 135¿W. It moves northeastward, passing across central Mexico, the eastern United States, and then the Atlantic before ending near 28¿N and 4¿E in Algeria. The area of the partial shadow is relatively large, and the total solar flux incident on the dayside of the earth is decreased by about 6% during the eclipse. The TGCM calculates the time-dependent response of the winds, temperature, and the mass mixing ratios of the major constituents throughout the thermosphere. Perturbations follow the path of the annular eclipse, with maximum deviations occurring near 1700 UT at about 300 km for the temperature and at higher altitudes for the winds and composition.

The perturbation winds converge from all directions toward the shadow at speeds reaching 75 m s-1 in the upper thermosphere. The maximum temperature anomaly (-55 K) and vertical wind anomaly (-7 m s-1) occur near the center of the shadow. At a constant altitude of 300 km, both the N2 density and the O density decrease by about 10% and 6%, respectively. The path of maximum obscuration passes within 3¿ of latitude of the incoherent scatter radar at Millstone Hill, Massachusetts (42.6¿N, 71.5¿W). The station experiences a maximum solar obscuration of 86% at 1700 UT (1200 LT). A time-dependent one-dimensional numerical model of the ionosphere that use the TGCM-calculated winds, temperature, and composition at Millstone Hill is used to calculate the electron and ion densities and temperatures and the densities of the odd-nitrogen species NO, N(4S), and N(2D) during the eclipse. The calculated electron density decreases by about factors of 2 in the F region, 4 in the F1 region, and 3 in the E region compared to a similar control run without an eclipse. The F1 region emerges during the eclipse with an increase in the NO+/O+ ratio.

The calculated electron temperature decreases by 460 K during the eclipse but then increases 200 K following the eclipse because of the intense solar heating in a region of reduced electron densities. The calculated ion temperature generally follows the changes in neutral temperature. The calculated thermospheric and ionospheric responses agree well with measurements made by the Millstone Hill incoherent scatter radar.