High Performance, High Fidelity: A GPU‐Accelerated Doubly‐Periodic Configuration of the Simple Cloud‐Resolving E3SM Atmosphere Model Version 1 (DP‐SCREAMv1).

The development of the Simplified Cloud Resolving Energy Exascale Earth System Atmosphere Model (SCREAMv1) enables global storm‐resolving simulations on modern GPU‐based supercomputers. However, the high computational cost of SCREAMv1 limits its routine use for process‐level studies, creating a need for efficient proxy configurations. This study addresses this gap by introducing DP‐SCREAMv1, a doubly periodic cloud‐resolving model designed to be fully consistent with SCREAMv1 while enabling high‐resolution, long‐duration simulations at significantly reduced computational expense by simulating a limited doubly periodic domain rather than the entire globe. Built on a C++/Kokkos architecture, DP‐SCREAMv1 achieves exceptional performance scalability on GPU systems and includes a rich library of cases for validation and scientific exploration. In this work, we demonstrate short wall‐clock times at SCREAMv1's default resolution and show that DP‐SCREAMv1 supports routine execution of large‐domain, high‐resolution experiments that were previously challenging in practice. Furthermore, we show that DP‐SCREAMv1 enables routine execution of "Giga‐LES" style simulations and facilitates large‐domain, high‐resolution simulations that were recently considered burdensome to perform. These results document an efficient, fully consistent process‐level configuration for SCREAMv1 (DP‐SCREAMv1) and illustrate its use for long‐duration and large‐domain experiments at cloud‐resolving to eddy‐permitting resolution. Plain Language Summary: Understanding and predicting weather and climate relies on advanced computer models that simulate atmospheric processes like storms and clouds. However, the most detailed and accurate models, which resolve these processes at very fine scales, are extremely computationally expensive and difficult to run for long periods or over large areas. To address this, we developed DP‐SCREAMv1, a streamlined version of a state‐of‐the‐art global atmospheric model. DP‐SCREAMv1 is designed to focus on smaller, more controlled areas of the atmosphere while maintaining the same high level of detail and accuracy as the global model. By harnessing modern supercomputers and their powerful graphics processing units (GPUs), DP‐SCREAMv1 can perform simulations that were previously impossible, such as modeling a large area of the tropics at resolutions as fine as 200 m. These simulations provide new insights into how precipitation forms and organizes, improve understanding of key atmospheric processes, and help refine the global model. DP‐SCREAMv1 also allows scientists to test new ideas more efficiently, paving the way for better climate and weather predictions in the future. This work demonstrates how cutting‐edge computing can push the boundaries of what we can learn about our atmosphere and associated models. Key Points: DP‐SCREAMv1 has been developed, leveraging the high‐performance benefits and achieving full implementation consistency with the global modelSimulations with default resolution and standard domain size are now trivial with DP‐SCREAMv1High‐resolution, long‐duration, and relatively large limited‐area simulations are now possible with DP‐SCREAMv1 on GPUs [ABSTRACT FROM AUTHOR]

Copyright of Journal of Advances in Modeling Earth Systems is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)