Abstract. Volcanic aerosols in the stratosphere produce significant transitory solar and infrared radiative perturbations, which warm the stratosphere, cool the surface and affect stratospheric circulation. In this study, using the Geophysical Fluid Dynamics Laboratory SKYHI general circulation model (GCM) with a high vertical resolution and a recently improved radiative transfer code, we investigate the aerosol radiative forcing and the stratospheric temperature response for the June 15, 1991, Mount Pinatubo eruption, the most well observed and largest volcanic eruption of the 20th Century. The investigation is carried out using an updated, comprehensive monthly and zonal-mean Pinatubo aerosol spectral optical properties data set. While the near-infrared solar spectral effects contribute substantially to the total stratospheric heating due to aerosols, over the entire global domain the longwave component exceeds the solar in causing a warming of the lower stratosphere (30-100 hPa). In contrast, the magnitude of the solar perturbation (increased reflection) in the overall surface-atmosphere radiative heat balance exceeds that due to the longwave (infrared trapping effect). The troposphere affects the stratospheric radiative forcing, mainly because of the dependence of the reflected solar and upward longwave radiation on cloudiness, and this adds to the uncertainty in the calculation of the stratospheric temperature response. A four-member ensemble of 2-year GCM integrations (June 1991 to May 1993) were performed using fixed sea surface temperatures and a cloud prediction scheme, one set with and another without the volcanic aerosols. The temperature of the tropical lower stratosphere increases by a statistically significant 3 K, which is almost I K less than in previous investigations that employed coarser vertical resolution in the stratosphere, but is still larger than observed. In the low latitudes the evolution of the simulated temperature response •nimics that observed only through about the first year. Thereafter, despite a significant aerosol optical depth perturbation in the tropical atmosphere, there is a lack of a signature in the temperature response that can be unambiguously attributed to the Pinatubo aerosols, suggesting other forced or unforced variations (e.g., ozone changes, quasi-biennial oscillation) occurring in the actual atmosphere which are unaccounted for in the model. In the high latitudes the large interannual variability prohibits a clear quantitative comparison between simulated and observed temperature changes and renders the aerosol-induced thermal signals statistically insignificant. In the global mean the evolution of the simulated lower stratospheric temperature response is in excellent agremnent with the observation for the entire 2-year period, in contrast to the model-observation comparison at the low latitudes. This arises because in the global mean the stratospheric response is not sensitive to dynamical adjustments within the atmosphere caused by internal variations, and dep...