Convection Permitting Climate Simulations of Mesoscale Convective Systems in a Global Variable-Resolution Model

Organized mesoscale convection systems (MCSs) are responsible for ~60% of summer rainfall in the U.S. Great Plains. Deficiency in representing MCSs contributes largely to climate model biases in simulating the diurnal variability of precipitation over the Central US, with important implications to model fidelity in simulating the regional water cycle in many regions influenced by MCSs worldwide. Simulating MCSs is a significant challenge because cumulus parameterizations commonly used in climate models do not represent the organization of convection. Convection permitting regional climate simulations have demonstrated more skill in capturing characteristics of MCSs. This motivates the use of global variable resolution models with regional refinement at convection permitting resolution to simulate MCSs and their interactions with the large-scale circulation. We explore the non-hydrostatic global variable-resolution modeling framework, Model for Prediction Across Scales (MPAS), for simulating MCSs over the Central US. MPAS experiments are conducted for the summer season of 2011 at multiple resolutions including global quasi-uniform resolutions of 120 km and 30 km and global variable resolution at 30 km with a regional mesh refinement at 4 km over the US Great Plains. An MCS detection algorithm is applied to the simulations and observations to evaluate the sensitivity of MCS characteristics such as frequency, size, and precipitation amount to model resolution and convective parameterizations in the MPAS global variable resolution modeling framework.