The Impact of Mesoscale Convective Systems on the Multi-scale Modeling Framework (MMF): Rainfall and its Characteristics

The importance of the precipitating mesoscale convective cloud systems (MCS) has been quantified from the TRMM precipitation radar and microwave imager retrievals.  The MCSs generate more than 50% of rainfall in most regions.  The typical MCSs have horizontal scale of the a few hundred kilometers.  In cloud resolving model (CRM) simulations, a large domain and high resolution are required to realistic representation MCSs.  Almost all global and climate models have problem to represent the MCSs and their propagation.  Although multi-scale modeling frameworks (MMFs) are capable of capture some organized MCS-like storm signals and propagation, their embedded CRMs typically have small domain (128 km) and coarse resolution (4 km) that cannot realistic simulate MCSs.

The impact of number of grid points and resolution of CRM (Goddard Cumulus Ensemble or GCE) embedded in the Goddard MMF (GMMF) is examined by conducting both non-coupling (GCE) and coupling (GMMF) simulations.  The results indicate that the more grid points (i.e., 128 and 256) in the both GCE and GMMF produce better simulations compared to those simulations with less grid points (i.e., 32 and 64).  In GMMF simulations, both precipitation bias and root mean square error (RMSR) are reduced, and correlation is increased in embedded GCE with 256 grid points compared to cases with less grid points (especially compared to those with 32 grid points).