Eight years of NMC (National Meteorological Center) temperature and SBUV (solar backscattered ultraviolet) ozone data were used to calculate the monthly mean heating rates and residual circulation for use in two-dimensional photochemical model in order to examine the interannual variability of modeled ozone. Fairly good correlations were found in the interannual behavior of modeled and measured SBUV ozone in the upper stratosphere at middle to low latitudes, where temperature dependent photochemistry is thought to dominate ozone behavior. The calculated total ozone is found to be more sensitive to the interannual residual circulation changes than to the interannual temperature changes. The magnitude of the modeled ozone variability is similar to the observed variability, but the observed and modeled year to year deviations are mostly uncorrelated. The large component of the observed total ozone variability at low latitudes due to the quasi-biennial oscillation (QBO) is not seen in the modeled total ozone, as only a small QBO signal is present in the heating rates, temperatures, and monthly mean residual circulation. Large interannual changes in tropospheric dynamics are believed to influence the interannual variability in the total ozone, especially at middle and high latitudes. Since these tropospheric changes and most of the QBO forcing are not included in our model formulation, it is not surprising that the interannual variability in total ozone is not well represented in our model computations. ¿ American Geophysical Union 1991