| description abstract | We use a global, primitive equation model to study the evolution of waves growing in a zonal mean state that is initially baroclinically unstable. The waves produce changes in the zonal mean state that we compare with changes predicted by baroclinic adjustment theories We examine mean state adjustment by representative zonal wavenumbers 3, 7 or 12. In the absence of surface processes, as the wave grows to its maximum amplitude, it reduces the zonal mean state's potential vorticity gradient through the lower troposphere, in accord with adjustment theories. Over the latitudes with largest wave amplitude, changes in the static stability and the zonal wind's vertical shear contribute about equally to the potential vorticity gradient adjustment. However, during the last day of a wave's growth, momentum fluxes strengthen the barotropic component of the zonal wind and the potential vorticity gradient in the middle troposphere, changes that are not anticipated by adjustment theory. The static stability adjustment occurs across the latitudinal band occupied by the growing wave. Further experiments show that the static stability adjustment alone is very effective in reducing the instability of the flow and restricting the maximum amplitude attained by growing waves. Adjustment of the zonal wind's vertical shear is confined to a narrower range of latitude and is partially reversed as the wave decays. Additional experiments indicate that the barotropic governor mechanism of James does not contribute strongly to the mean flow's stabilization in the cases we examine, though it way inhibit secondary growth at latitudes adjacent to the initial disturbance. When the model includes surface friction and heat flux, the waves adjust the zonal mean state less effectively, especially near the surface. Surface heat flux inhibits static stability adjustment, and surface friction inhibits adjustment of the zonal wind's vertical shear. In the absence of surface processes, the adjusted state produced by the wave is quite different from observed mean structures. However, with both surface processes included, the vertical profiles of the adjusted static stability, wind shear and potential vorticity gradient are similar to observed profiles. The model' interaction between the waves and the mean flow corroborates results from previous studies of baroclinic adjustment that used simpler representations of atmospheric dynamics. | |
