and hence the location of wave-induced zonal mean momentum forcing and
the resulting Brewer-Dobson circulation, is dynamically determined by
the distribution of zonal mean potential vorticity; steep potential
vorticity gradients at the vortex edge, for example, act as a
wave-guide, allowing wave propagation to higher altitudes before wave
breaking occurs. We consider a series of forced-dissipative
experiments to test the effect of the potential vorticity distribution
on the Brewer-Dobson circulation in models of varying horizontal
resolution, the latter acting to limit the extent to which wave
breaking may steepen potential vorticity gradients at the vortex edge.
Under perpetual January radiative conditions and with steady wave
forcing at the tropopause the model system exhibits strong internal
variability, with the polar vortex undergoing a series of
quasi-periodic sudden warmings separated by a period of gradual vortex
recovery under radiative forcing. Although under certain forcing
parameters the horizontal resolution is able to influence
qualitatively the nature of the internal variability, in general, the
time-average vertical upwelling at low latitudes in the lower
stratosphere is relatively insensitive to changes in the horizontal
resolution; changes in the wave breaking are balanced by changes in
diffusion, resulting in similar zonal mean momentum forcing. The
distribution of wave flux convergence, details of the potential
vorticity distribution, and structure of the Brewer-Dobson circulation
are investigated across a range of horizontal resolutions spanning
those currently used in climate modeling.