Implementation of Comprehensive Cloud Microphysics in Global Climate Models

Abstract

Implementation of comprehensive cloud microphysics in 50km resolution global climate models with appropriate vertical mixing is the main objective of this study. The comprehensive cloud microphysics was obtained from the Goddard Cumulus Ensemble (GCE) model which is a kind of Cloud Resolving Models (CRM). To implement the cloud microphysics to the global climate models, the resolution dependency of cloud microphysics was examined using CRM simulations with different resolutions, and it is improved by modifications of fractional saturation and terminal velocity. Appropriate vertical mixing for 50km resolution global climate models was examined using General Circulation Model (GCM) simulations with different horizontal resolutions and a 3-dimensional CRM simulation. As the horizontal resolution of the GCM is increased, the resolved grid-scale vertical transport is increased. Accordingly, the sub-grid vertical transport parameterized by a convection scheme has to be decreased appropriately. Using the 3-dimensional CRM simulation with 1km horizontal resolution, the appropriate ratio of parameterized sub-grid scale vertical transport to total vertical transport is investigated as the horizontal resolution changes. The ratio obtained from the 3-dimensional CRM simulation is used as a guideline to decrease the parameterized sub-grid scale vertical transport in GCMs. The cumulus base mass flux is used as a control factor to decrease the parameterized sub-grid scale vertical transport. The 50km global climate model with comprehensive cloud microphysics and appropriate vertical mixing was improved to simulate the vertical profiles of temperature and moisture, precipitation frequency as a function of precipitation strength, and Madden-Julian Oscillation (MJO). The MJO simulation diagnostics developed by MJO Working Group and the process-oriented MJO simulation diagnostics developed by MJO Task Force were applied to developed model in this study and the model intercomparison with CMIP5 models was conducted. The developed model relatively well simulates the MJO propagation, the convective-moisture coupling and the longwave radiation feedback processes.