Exploring a Multi-resolution Approach Using AMIP Simulations

Journal Article · Fri Jul 31 00:00:00 EDT 2015 · Journal of Climate, 28(14):5549–5574

DOI:https://doi.org/10.1175/JCLI-D-14-00729.1· OSTI ID:1222892

Sakaguchi, Koichi; Leung, Lai-Yung R.; Zhao, Chun; Yang, Qing; Lu, Jian; Hagos, Samson M.; Rauscher, Sara; Dong, Li; Ringler, Todd; Lauritzen, P. H.

This study presents a diagnosis of a multi-resolution approach using the Model for Prediction Across Scales - Atmosphere (MPAS-A) for simulating regional climate. Four AMIP experiments are conducted for 1999-2009. In the first two experiments, MPAS-A is configured using global quasi-uniform grids at 120 km and 30 km grid spacing. In the other two experiments, MPAS-A is configured using variable-resolution (VR) mesh with local refinement at 30 km over North America and South America embedded inside a quasi-uniform domain at 120 km elsewhere. Precipitation and related fields in the four simulations are examined to determine how well the VR simulations reproduce the features simulated by the globally high-resolution model in the refined domain. In previous analyses of idealized aqua-planet simulations, the characteristics of the global high-resolution simulation in moist processes only developed near the boundary of the refined region. In contrast, the AMIP simulations with VR grids are able to reproduce the high-resolution characteristics across the refined domain, particularly in South America. This indicates the importance of finely resolved lower-boundary forcing such as topography and surface heterogeneity for the regional climate, and demonstrates the ability of the MPAS-A VR to replicate the large-scale moisture transport as simulated in the quasi-uniform high-resolution model. Outside of the refined domain, some upscale effects are detected through large-scale circulation but the overall climatic signals are not significant at regional scales. Our results provide support for the multi-resolution approach as a computationally efficient and physically consistent method for modeling regional climate.

• Research Organization: Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
• Sponsoring Organization: USDOE
• DOE Contract Number: AC05-76RL01830
• OSTI ID: 1222892
• Report Number(s): PNNL-SA-106250; KP1703010
• Journal Information: Journal of Climate, 28(14):5549–5574, Journal Name: Journal of Climate, 28(14):5549–5574
• Country of Publication: United States
• Language: English