Cloud Microphysical Control on the Dehumidifying Power of Deep Convection

Deep convection is the primary mechanism by which energy is vertically transported in the tropical atmosphere. This energy transfer, manifested through latent heating by clouds, counteracts top of the atmosphere radiative cooling. Radiative-convective equilibrium (RCE) is a state in which these two modes of energy transfer balance, and has been shown to be a good approximation of the tropical atmosphere. It has long been conjectured that this behavior deems moist convection a heat engine. However, Pauluis and Held’s (2002) seminal paper on the entropy budget of an atmosphere in RCE argues that the convectively active moist atmosphere behaves more like a dehumidifier rather than a heat engine. The basis of the argument being that the vertical heat transport is dominated by latent heat which is efficiently rained out.

For over two decades now, numerical simulations of radiative-convective equilibrium at the cloud resolving scale have unraveled a phenomenon termed convective self-aggregation, whereby deep convection clusters into a single region of the domain. The implication is a polarized atmosphere, in which one region experiences deep convection while the other experiences radiatively driven subsidence. As in the tropical circulation, the dry subsiding air penetrates the moist boundary layer and converges at low levels into the convective region. Convectively aggregated atmospheres are on average drier and rainier than their non-aggregated counterparts, which lends theoretical support to the dehumidifier hypothesis.

Much progress has recently been made on quantifying the feedbacks necessary for convective self-aggregation to occur and sustain itself. However, very little attention has been given to the cloud microphysical process which implicitly come into play in many of these feedbacks. Here, we explore the hypothesis that the precipitating efficiency determined by the cloud microphysics can have an impact on the RCE state. We conducted RCE simulations with the ICOsahedral Non-hydrostatic atmosphere large eddy model (ICON-LEM) on a long channel domain (2000 km x 120 km) with periodic boundary conditions, a two moment microphysics scheme, and an interactive radiation scheme. Changes made to each simulation were the number concentration of cloud condensation nuclei which was varied from 50 cm^-3 (extremely pristine) to 6000 cm^-3 (extremely polluted). We found that at concentrations lower than about 500 cm³ increases in CCN led modest reductions in the strength of the aggregation. Using the moist static energy budgeting framework developed by Wing and Emanuel (2014), we quantified the feedbacks on convective aggregation for our simulations and found that increasing CCN in this range led to a reduction in the surface flux feedback but an enhancement in the longwave feedback. This can be nicely explained by the increased amount of high level clouds which partially compensates for the reduction in precipitating efficiency. However, at concentrations above 500 cm^-3, the aggregated state is significantly disturbed with reductions in surface precipitation on the order of 20-30 %, detrainment of cirrus clouds into the dry regions which weakens the subsidence, the emergence of a mid-level cloud distribution, and a reduction in low level cloud amount. The system experiences substantial reduction in the strength of the aggregation which stems from changes in both the diabatic mechanisms and the circulations they drive.

We go one step further and attempt to interpret what is happening in terms of the dehumidifier hypothesis. To do so, we analyze the temperature, moisture, and condensate loading contributions to the total mechanical work done by the system. We find that the decreased precipitating efficiency (due to increased CCN concentrations) always leads to a reduction of the work done to vertically transport water vapor at the expense of frictional dissipation to the enhanced mass of condensate. Throughout the range of CCN concentrations simulated, the percentage of mechanical work lost to frictional dissipation increases from 77% to 96%. We discuss the theoretical and practical implications of these results on aerosol-cloud interactions, cloud microphysics, and deep convection.

References

Pauluis, O. and Held, I. M.: Entropy Budget of an Atmosphere in Radiative–Convective Equilibrium. Part I: Maximum Work and Frictional Dissipation, J. Atmos. Sci., 59(2), 125–139, doi:10.1175/1520-0469(2002)059<0125:EBOAAI>2.0.CO;2, 2002.

Wing, A. A. and Emanuel, K. A.: Physical mechanisms controlling self-aggregation of convection in idealized numerical modeling simulations, J. Adv. Model. Earth Syst., 6(1), 59–74, doi:10.1002/2013MS000269, 2014.