Toward Earth system modeling with resolved clouds and ocean submesoscales on heterogeneous many-core HPCs

Area of Science:

• Earth System Science
• Climate Modeling
• High-Performance Computing

Background:

• Advancements in high-resolution Earth system models (HR-ESMs) are crucial for understanding complex Earth processes.
• The development of powerful supercomputing infrastructure enables unprecedented model resolution and complexity.

Purpose of the Study:

• To introduce the newly developed Sunway heterogeneous-architecture supercomputer and its high-resolution coupled Earth system models (SW-HRESMs).
• To showcase the progress in SW-HRESMs development and their capability in simulating weather-climate extremes.
• To outline future directions for developing non-hydrostatic, cloud and ocean submesoscale-resolving Earth system models.

Main Methods:

• Development of SW-HRESMs with atmospheric resolution up to 5 km and oceanic resolution up to 3 km.
• Utilizing the Sunway supercomputer for high-performance computations.
• Analysis of preliminary model results focusing on weather-climate extremes and multiscale interactions.

Main Results:

• SW-HRESMs demonstrate capability in capturing major atmospheric and oceanic weather-climate extremes.
• The models highlight the importance of resolved clouds and ocean submesoscale eddies in simulating tropical cyclones and eddy-mean flow interactions.
• Preliminary results show the potential for multiscale interaction studies at various computational costs.

Conclusions:

• The developed SW-HRESMs represent a significant advancement in Earth system modeling, offering higher resolution and improved physics.
• Further development towards non-hydrostatic and submesoscale-resolving models is essential for more realistic simulations.
• These advancements pave the way for future research into finer-scale Earth system processes and climate prediction.