The climatology of the stratospheric ‘THIN AIR’ model

An improved two‐and‐a‐half dimensional chemical–dynamical–radiative model shows remarkably realistic behaviour in a number of respects. The stratospheric/mesospheric westerly jet slopes equatorwards with height and is split at mid‐latitudes in the southern hemisphere. The parametrized orographic gravity waves exert a mid‐latitude drag and are important in splitting the jet, and are possibly important in pre‐conditioning the stratosphere for a sudden warming. A large easterly drag is required in the winter high‐latitude mesosphere in order to obtain a tropical mesospheric westerly jet. The summer easterly jet maximum lies above the tropospheric jet, due to gravity‐wave absorption in the lower stratosphere. Very realistic annual and semi‐annual cycles are produced in the model, with a mesospheric semi‐annual oscillation forced by tidal drag and gravity waves. Planetary‐wave temperatures are also modelled well, and wave amplitudes are correlated with the westerly jets. In the lower stratosphere, zonalmean Ertel's potential vorticity gradients strongly influence the wave activity, horizontal convergence and heating rates, with descent being strongest equatorwards of the vortex edge. This forms the basis for an explanation of the inter‐hemispheric asymmetry in the ozone column. Nitrous oxide distributions indicate too much equatorial ascent but show a ‘double peak’ connected with the tropical circulation, and steep horizontal gradients near the winter vortex edge.