Important progress has been made recently in developing an understanding of the effects of geomagnetic storms in the thermosphere and ionosphere. Numerical simulations of theoretical storms with the coupled thermosphere ionosphere model (CTIM) have provided a better understanding of the dynamics of the upper atmosphere and have also permitted the identification of the processes responsible for global storm effects at highlatitude and midlatitude. The theory developed based on the model simulations can explain most of the apparent coherence of local time and seasonal dependencies and the apparent randomness in the longitudinal response of the global ionosphere, uncovered through statistical analysis of storm observations. A true test of the model and the theory is their ability to predict the large-scale distribution of storm effects for specific storms. In this paper, CTIM simulation results for the December 7--9, 1982, period are presented. We compare model results with DE 2 temperature and plasma density data. We also compare modeled electron densities with ionosonde data from several sectors in both hemispheres. The global characteristics of the response are reproduced by the model, and we are able to explain the pronounced longitude differences in the summer hemisphere. The Australian sector passes through midnight during the main driven phase of the storm and experiences the largest energy input and the largest neutral composition changes. The deepest ionospheric negative phase seen in ionosonde data is over Australia and is consistent with this interpretation. Given the large uncertainties in our knowledge of the magnitude and spatial distribution of energy input during a particular storm, predicting local changes is still a challenge.¿ 1997 American Geophysical Union