Sensitivities of storm divergence and stratiform rain production to microphysics and cumulus parameterizations

Mesoscale model simulations of precipitating upper-level disturbances and frontal systems in southeast Texas have shown that storms that form in less baroclinic environments can generate more elevated convergence despite producing lower associated stratiform area fractions. However, the model-derived levels of non-divergence (LNDs) and stratiform area fractions exhibit some bias depending on the microphysics and cumulus parameterization schemes employed. These sensitivities are less pronounced for storms with higher degrees of baroclinicity that typically contain larger stratiform area fractions, but are most pronounced for those barotropic and weakly baroclinic storms with elevated divergence profiles. Quantifying why these differences in divergence and stratiform rain production occur for these storms is critical because of the potential implications a more elevated upper-level heating profile (particularly within stratiform regions) may have on driving additional convection in the subtropics.

This study utilizes a suite of MM5 simulations with nested domains of 27, 9, and 3-km for several well-modeled storms to evaluate model-derived divergence profiles and stratiform rain production for varying sets of cumulus and microphysics parameterization schemes. Although cumulus schemes are used on the 27-km coarse domain for all cases, additional model runs present differences between incorporating cumulus parameterizations on the 9-km domain versus explicitly resolving convection. The Goddard and Reisner schemes with graupel are utilized to analyze microphysical sensitivities while the Grell and Kain-Fritsch with shallow convection schemes are used to evaluate differences in cumulus parameterizations. These results should be somewhat analogous for similar parameterization schemes in other mesoscale models like WRF. Some attempt is also made to evaluate how different schemes simulate the transition of deep convection to anvil cloud and stratiform rain regions, particularly in frontal leading line-trailing stratiform MCSs. Although this study does not attempt to determine which schemes are most valid, it does show how stratiform precipitation processes and mean divergence (and thus heating) profiles are affected by microphysics and cumulus parameterization schemes, a result that is likely applicable to other mesoscale models and potentially climate models.