Skip to main content
U.S. flag

An official website of the United States government

The CAM-MPAS model simulations for the HighResMIP experiments

Presentation Date
Thursday, December 10, 2020 at 4:00am



The High Resolution Model Intercomparison Project (HighResMIP) provides a new and unique opportunity to assess the impact of horizontal resolution on climate simulations at the global and century-long time scales through a large collection of multi-model simulations produced under a common protocol. This presentation introduces the contribution from the DOE-funded Water Cycle and Climate Extremes Modeling project to the HighResMIP, using a modified version of the Community Atmosphere Model version 5.4. Aiming for global simulations at meso-scale resolutions, the following changes have been made to the base model. (1) The default finite-volume dynamical core is replaced with the Model for Prediction Across Scales (MPAS), a global non-hydrostatic dynamical core on unstructured meshes configured using quasi-uniform meshes at 120km (LR) and 30km (HR) grid spacings. (2) The Zhang and McFarlane deep convection scheme is replaced by the Grell-Freitas scheme, known for its scale-awareness. The prognostic aerosol treatment is also replaced with the MACv2-SP scheme following the HighResMIP protocol. The LR and HR present-day simulations with prescribed ocean forcings (“HighResSST”) are currently running and will be followed by the future climate experiment. Preliminary results show that the global-mean climate of the LR model is close to that of the CAM5.4 base model, but with some significant regional differences. Compared to the base model, the zonal-mean tropical precipitation and Amazon precipitation in LR CAM-MPAS agree better with observations, while the Asian summer monsoon is weaker and the positive bias of the mid-latitude jet speed is worsened, especially in the Southern Hemisphere. These two degraded aspects, however, are improved in the HR CAM-MPAS simulation. As in previous studies, HR CAM-MPAS produces more intense daily precipitation than LR, but CAM-MPAS exhibits smaller resolution sensitivity compared to the base model, particularly for precipitation produced by parameterized convection. Overall, the preliminary results in the historical simulations give confidence to the credibility of the CAM-MPAS model and show some advantages of higher resolution. Other notable characteristics and resolution impacts will be presented with a focus on the dynamics and hydrological cycle.

Funding Program Area(s)