A two-dimensional model with input parameters from a general circulation model: Ozone sensitivity to different formulations for the longitudinal temperature variation

Net heating and temperature derived from the middle atmospheric version of the National Center for Atmospheric Research's Community Climate Model (MA CCM2) history tapes are used to evaluate three different approaches to account for zonal temperature asymmetries in the calculation of gas phase and heterogeneous chemical reaction rate constants and polar stratospheric cloud (PSC) surface area in a two-dimensional chemistry transport model (2-D CTM). The first method uses the daily (and monthly) averaged three-dimensional (3-D) temperature distribution derived from the MA CCM2 to calculate chemical and heterogeneous reaction rates at each 3-D grid point, followed by zonal averaging (pseudo-3-D method). The second method uses 3-D daily temperature statistics from the MA CCM2 to calculate the monthly averaged probability function (stochastic approach). The third method is based on a planetary wave superposition on the zonally averaged temperature (wave approach). The sensitivity of the gas phase reactions to the longitudinal temperature asymmetry is small, while the sensitivity of the heterogeneous reaction rates is comparable to the ozone response to aircraft emissions. All three methods of accounting for longitude temperature asymmetry give similar PSC morphologies in the southern hemisphere, in good agreement with climatological data and independent model calculations. In the northern hemisphere, where the CCM2 winter temperatures at high latitudes are known to be warmer than those observed, the PSCs predicted by the pseudo-3-D and wave methods are much scarcer than those observed or calculated by other authors using climatological temperatures. For the same reason, all other methods employed in the present study failed to predict any PSCs in the northern hemisphere.