Neelin and Held assumed that the moist static energy is approximately conserved in tropical circulations in the absence of diabatic effects. They then asserted that air entering, ascending through, and exiting a moist convective region gains a certain amount of moist static energy as a result of the net effect of surface energy fluxes and radiative flux divergences in the troposphere. Turning this around, one sees that the strength of the ascending mass current in a convective region is equal to the integrated moist static energy creation rate in the region divided by the specific gain in moist static energy in parcels flowing through the system. Neelin and Held call this specific gain the ``gross moist stability'' (GMS). The smaller the GMS, the greater the ascending mass current and associated heating and precipitation rates for a given moist static energy creation rate. Recent observations of deep convective systems in TOGA COARE confirm that the GMS is indeed positive for these systems (Lopez and Raymond, presented at this conference).
In a time-dependent situation, creation of moist static energy either further moistens the environment or it results in upward mass currents. If cloud physics forces the precipitation rate to be an increasing function of humidity, then the system tends toward the time-independent balance of Neelin and Held, with the relaxation time being a decreasing function of the strength of the existing convection (Raymond, in press, Quart. J. Roy. Met. Soc.). The Neelin-Held relationship is promoted from a steady-state diagnostic relationship into a prognostic relationship with this closure.
In radiative-convective equilibrium the creation of moist static energy by surface fluxes is balanced by radiative losses. A positive creation rate therefore can be caused either by an increase in surface heat flux or a decrease in radiative cooling from radiative-convective equilibrium values. Since persistent deep convection, such as that which occurs in ITCZs, often has weaker surface fluxes than surrounding regions, it follows that the radiative losses must be sufficiently suppressed there to overcome the reduced surface fluxes and produce net moist static energy creation, as is needed to sustain the convection. This can only result from cloud-radiation interactions. Thus, strong deep convection in regions of average or weaker surface fluxes is a telltale sign of significant suppression of radiative losses by clouds.