The sensitivity of simulated shallow cumulus convection and cold pools to microphysics

The simulations differ only in their use of two widely used double-moment rain microphysical schemes: the Thompson and Morrison schemes. Both simulations produce similar mesoscale variability, with the Thompson scheme producing more weak cold pools and the Morrison scheme producing more strong cold pools, which are associated with more intense shallow convection.

The most robust difference is that the cloud cover and LWP are significantly larger in the Morrison simulation than in the Thompson simulation. One-dimensional kinematic simulations confirm that dynamical feedbacks do not mask the impact of microphysics. These also help elucidate that a slower autoconversion process along with a stronger accretion process explains the Morrison scheme's higher cloud fraction for a similar rain mixing ratio.