Nightglow within the mesopause region of the atmosphere is dominated by the infrared emissions from vibrationally excited hydroxyl (OH*) radicals at ~87km altitude. These emissions provide a signal from which the behaviour of the middle atmosphere can be studied and in recent years a wealth of observational data on the global distribution of OH* has been collected by instruments such as SABER on board NASA’s TIMED satellite and SCIAMACHY on board the ESA’s ENVISAT. General Circulation Models (GCM’s) represent the best tools available to test our understanding of the physical, chemical and radiative processes of the middle atmosphere. In this thesis, a one-dimensional (1-D) model of the first nine vibrational states of OH* is developed and incorporated into the University College London CMAT2 GCM. Unlike in other GCMs these vibrational levels are treated as distinct chemical species from in the ground state as suggested in Pickett et al. (2006). The results are compared to both satellite observations and predictions from other models. It is found that the 1-D model successfully predicts many of the observed features including the magnitude of OH* concentrations and a vibrational level dependence on the emission peak altitudes which is consistent with previously published results. The influence of atomic oxygen on the peak altitude of individual emission layers is found to be a more complex function than suggested in previous studies. Increased atomic oxygen increases the rate of OH* production but this is modulated by increased quenching rates particularly at the upper edge of the layers. On incorporation of the 1-D model into CMAT2, it is found that the model can generally make accurate predictions of both the temporal and spatial variability in the emissions, although the magnitude of the emissions are typically a factor of two or more higher than expected. This is attributed to increased O concentrations predicted by the model. Diurnal and seasonal variability is well represented, as are changes with respect to altitude and latitude. At equinox, a single peak in the emission is observed over the equator while during solstice, additional peaks occur in the mid- to high-latitude regions resulting from transport of oxygen by the mean circulation. Longitudinal variations are not well represented and suggest an underestimation of the effects of non-migrating tides. Migrating tides are shown to modulate the emission, in particular at the low latitudes although reduced effects are also observed in the mid- to high-latitude regions. Migrating tides have proven difficult in previous modelling studies prompting investigators to double tidal amplitudes in an effort to obtain better agreement with observations (Marsh et al., 2006). Similar tuning of the model’s dynamical routines may be required in the case of CMAT2.