| description abstract | A parameterization of cumulus ensemble effects on the large-scale vorticity is tested to interpret the vorticity budget residual, Z, observed during Phase III of GARP Atlantic Tropical Experiment (GATE). The parameterization is derived consistently from the parameterization of cumulus ensemble effects on the momentum equation. Values of the parameterized Z are computed using the cumulus properties (mass flux, mass detrainment and cloud momentum) diagnosed by a spectral cumulus ensemble model. The results confirm the inferences made in Part I of this paper that organized cumulus convection produces significant residuals in the large-scale vorticity budget mainly through 1) the detrainment of excess momentum from clouds, and 2) the vertical advection of the large-scale vorticity due to the subsidence of environmental air compensating the convective mass flux. In addition, the twisting of the horizontal component of the large-scale vorticity into the vertical component due to nonuniform spatial distributions of the convective mass flux plays a significant role in producing Z at the levels where the vertical wind shear is large. Deep cumulus convection decelerates the mean flow over the area of convection by the detrainment of smaller cloud momentum transported from below. This deceleration produces a positive vorticity tendency to the right of the convective area facing downstream and a negative tendency to the left. In addition, the curl of the excess momentum tends to produce a positive vorticity tendency over the area of convection. These effects explain the observed features of Z in the upper troposphere, i.e., the horizontal dipole pattern and the positive mean values. The vertical advection of the large-scale vorticity by the cumulus-induced subsidence is the dominant mechanism producing negative Z in the middle troposphere where the gradient of vorticity, ???/?p, is positive. In the lower troposphere where ???/?p is negative, the vertical advection effect produces positive Z. In addition, detrainment of momentum from shallow clouds is found to be significant near 650 mb and responsible for generating localized patterns in the horizontal distribution of Z. Results of additional experiments show improvements of the parameterized Z in the lower troposphere by including downdrafts in the diagnosis of mass flux and the potential importance of pressure interactions between the clouds and the environment in the cumulus momentum budget. | |
