The sensible heat budget of a large-scale area containing an idealized tropical cloud cluster is analyzed. The cluster is assumed to have spatial dimensions and precipitation rates typical of observed cloud clusters.
In its early stages of development the idealized cluster consists of isolated deep precipitating convective cells, or "hot towers." A simple model using the assumed precipitation rates as input is employed to compute the condensation and evaporation rates and sensible heat fluxes associated with the precipitating hot towers. The condensation dominates the contribution of the hot towers to the large-scale heat budget, and the net effect of the towers is warming distributed through the full depth of the troposphere.
In its mature stage of development, the idealized cluster contains not only convective towers but a widespread cloud shield interconnecting the towers. The cloud shield is dynamically and thermodynamically active, and processes associated with it also contribute significantly to the large-scale sensible budget. Stratiform precipitation falls from the cloud shield, and in the stratiform precipitatioi region, condensation occurs in mesoscale updraft aloft, evaporation occurs in a mesoscale downdraft at low levels and melting occurs in a middle-level layer. The condensation, evaporation and sensible heat transports associated with the mesoscale updraft and downdraft are determined from simple models using the cluster's assumed stratiform precipitation rate as input. The evaporation and melting in the stratiform precipitation region are also estimated from vertical profiles of radar reflectivity in real cloud clusters. The total effects of the stratiform precipitation processes on the largescale heat budget are warming of the middle to upper troposphere, where condensation in the mesoscale updraft is the dominant effect, and cooling in the lower troposphere, where melting and mesoscale downdraft evaporation dominate.
The widespread cloud shield present in the mature and later stages of a cloud cluster's life cycle is also an important absorber and emitter of radiation. Radiative transfer models applied to tropical cloud shields show substantial heating effects in the middle to upper troposphere. These effects are nearly as important as the heating by convective towers and the heating and cooling associated with the stratiform precipitation processes.
As the idealized cloud cluster progresses from early to mature stages of development, its net effect on the large-scale heat budget changes. As the cloud shield develops, the mesoscale updraft condensation and radiation reinforce the heating by convective towers aloft, while the mesoscale downdraft evaporation and melting counteract the convective-tower heating at low levels. Thus, the net heating by the cluster increases in the upper troposphere and decreases in the lower troposphere as the system develops. Large-scale upward motion, which is required to balance the large-scale heat budget against the effects of the cluster, is thus expected to increase aloft and decrease at low levels. Vertical motions deduced from large-scale wind observations in the tropics confirm this expectation. Thus, it is concluded that the mesoscale stratiform and radiative processes associated with the cloud shields of developing cloud clusters are sufficiently strong to alter the large-scale vertical motion field in the tropics.