The climatology of the Brewer-Dobson circulation and the contribution of gravity waves

The climatology of residual mean circulation – a main component of the Brewer–Dobson circulation – and the potential contribution of gravity waves (GWs) are examined for the annual mean state and each season in the whole stratosphere based on the transformed-Eulerian mean zonal momentum equation using four modern reanalysis datasets. Resolved and unresolved waves in the datasets are respectively designated as Rossby waves and GWs, although resolved waves may contain some GWs. First, the potential contribution of Rossby waves (RWs) to residual mean circulation is estimated from Eliassen–Palm flux divergence. The rest of residual mean circulation, from which the potential RW contribution and zonal mean zonal wind tendency are subtracted, is examined as the potential GW contribution, assuming that the assimilation process assures sufficient accuracy of the three components used for this estimation. The GWs contribute to drive not only the summer hemispheric part of the winter deep branch and low-latitude part of shallow branches, as indicated by previous studies, but they also cause a higher-latitude extension of the deep circulation in all seasons except for summer. This GW contribution is essential to determine the location of the turn-around latitude. The autumn circulation is stronger and wider than that of spring in the equinoctial seasons, regardless of almost symmetric RW and GW contributions around the Equator. This asymmetry is attributable to the existence of the spring-to-autumn pole circulation, corresponding to the angular momentum transport associated with seasonal variation due to the radiative process. The potential GW contribution is larger in September to November than in March-to-May in both hemispheres. The upward mass flux is maximized in the boreal winter in the lower stratosphere, while it exhibits semi-annual variation in the upper stratosphere. The boreal winter maximum in the lower stratosphere is attributable to stronger RW activity in both hemispheres than in the austral winter. Plausible deficiencies of current GW parameterizations are discussed by comparing the potential GW contribution and the parameterized GW forcing.