Cloud-Resolving Modeling Intercomparison Study of a Squall Line Case from MC3E – Convective Cores and Microphysics-Dynamics Feedback

Description

The large spread in CRM model simulations of deep convection and aerosol effects on deep convective clouds (DCCs) makes it difficult to (1) further our understanding of deep convection and (2) define “benchmarks” and then limits their use in parameterization developments. Past model intercomparison studies used different models with different complexities of dynamic-microphysics interactions, making it hard to isolate the causes of differences between simulations. In this intercomparison study, we employed a much more constrained approach – with the same model and same experiment setups as well as a piggybacking approach to explore the major microphysical processes controlling the model differences in the regimes of warm rain, mixed-phase, and ice phase, and the relative importance between latent heat and hydrometeor loading in terms of the feedback to dynamics. Real-case simulations are conducted for the squall line case May 20, 2011 from the MC3E field campaign. The intensity of strong updrafts differs by a factor of 2 among eight microphysics schemes, which is mainly contributed by ice microphysics. All schemes underestimate convection below 4 km (likely due to a stronger low-level jet). For the three schemes that dramatically overestimate convective strength at the upper levels, the middle-level pressure gradient, and upper-level latent heating and hydrometer mass are also among the largest as well, and they well correlate with strong updraft speeds. Piggybacking simulation results suggest (1) latent heating varies strongly with microphysical representation, (2) microphysics-dynamics feedback significantly buffers the model differences, and (3) the coupling intensity between microphysics and dynamics varies a lot with different microphysics schemes due to large differences of the latent heating and hydrometeor loading calculated by different microphysics schemes.