The effect of cloud size distribution on explicitly resolved entrainment and detrainment in LES of shallow convection

We analyze cloud mass entrainment for individual tracked clouds from large-eddy simulations (LES) for a set of quasi-steady state shallow cloud fields spanning a range of cloud size distributions. The large-eddy simulations are based on the CGILS (CFMIP-GASS Intercomparison of Large Eddy Models) trade cumulus control case (S6) of Blossey et al., (2013, J Adv. Model. Earth Syst.), typified by cloud tops below 3 km and cloud fractions of less than 10%. Shallow cumulus regimes with a range of cloud size distributions are generated using uniform sea-surface temperatures ranging between 295 K and 301 K, with large-scale subsidence and radiative cooling determined using a framework which minimizes the number of parameters used to describe the shallow cumulus regime, following Bellon et al. (2012, J. Atmos. Sci.) Runs are performed on a 9 x 9 km horizontal domain at a grid cell resolution of 25 meters.

Previous work by Dawe and Austin (2013, Atmos. Chem. Phys.) determined the joint probability density functions for cloud core fractional mass entrainment/detrainment rates and horizontal mean cloud core properties, as well as empirical best-fit relations between entrainment/detrainment and measures of cloud buoyancy, environmental moisture and stability in cloud populations of the simulated BOMEX (ocean equilibrium) and ARM (land diurnal cycle) LES datasets. We extend this work by determining the dependence of these relationships on the quasi-steady state cloud size statistics of the CGILS-based cloud fields. We further analyze the hypothesis proposed by Dawe and Austin (2014, J. Adv. Model. Earth Syst., submitted) that the bulk fractional entrainment rate of a cloud field is primarily determined by the size of the largest clouds in the field and not by the average cloud size.