Supersaturation and droplet growth statistics in turbulent moist convection: Applications for LES subgrid modeling

Turbulent fluctuations of scalar fields are critical for cloud microphysical processes. Variability of supersaturation (a joint scalar of water vapor and temperature) can affect cloud droplet formation and droplet size distribution in turbulent clouds. This study investigates water vapor, temperature, supersaturation, and droplet statistics in DNS of moist Rayleigh-Bénard convection to provide a foundation for modeling subgrid-scale interactions in atmospheric models.



Supersaturation behaves differently than independent scalars due to nonlinearity, contrasting with commonly assumed independent scalar-like behavior in models. Although supersaturation has autocorrelation and structure functions close to independent scalars, the autocorrelation timescale of supersaturation differs. Relative scalar fluxes from the sidewalls (like during the entrainment-mixing process in atmospheric clouds) in DNS without cloud droplets make supersaturation PDFs less skewed than the adiabatic sidewalls, where they are highly negatively skewed. However, droplet condensation changes the PDF shape response: it becomes positively skewed for the adiabatic case and negatively skewed when the relative sidewall fluxes are large. Condensation also affects correlations between water vapor and temperature, suppressing supersaturation variability in non-adiabatic cases and increasing it in adiabatic cases. A subgrid model for supersaturation variance in LES is proposed and evaluated against DNS. Further evaluation of the coupled LES subgrid model with droplet growth variabilities will also be discussed.