From the Program Manager, May ’25: Welcoming New PI, Advancing The Vision
Xujing Davis, E3SM Program Manager, DOE BER
Welcoming New PI, Advancing The Vision
Welcome to the E3SM newsletter. I hope this message finds you well, and thank you very much for your continued interest in, support of, and contribution to the E3SM science.
Major Project Announcement: PI and Leadership Team Transition
First, I would like to announce that Dr. David Bader has passed his baton to Dr. Peter Caldwell as the E3SM Project’s Principal Investigator (PI). During his decades of science endeavors across multiple DOE national laboratories and the DOE headquarters, Dr. Bader has made transformative contributions to atmosphere and earth systems science. Thanks to his distinguished leadership, strategic vision, and steadfast dedication, E3SM now stands as an internationally recognized, top-tier, computationally advanced Earth system model with high resolution capabilities. This positions the DOE to make groundbreaking progress in overcoming critical obstacles in predicting Earth and energy systems, ultimately bolstering future energy security.
Dr. Peter Caldwell, the new E3SM PI, is a well-recognized leader in the atmosphere science and earth system modeling community nationally and internationally. During the past few years, he has established himself as a pioneer by successfully leading the development of the first exascale-ready model code, i.e., the Simple Cloud Resolving E3SM Atmosphere Model (SCREAM), in km scale Earth system modeling, and in exploration of AI technologies in advancing earth system modeling. In addition to his technical achievements, Dr. Caldwell has demonstrated outstanding leadership within the E3SM community, serving in key roles and advocating for collaboration across disciplines.
I would like to join the project team to express my deepest gratitude to Dr. Bader, for his exceptional leadership, enormous contribution and the legacy of excellence! Please also join me in congratulating and welcoming Dr. Caldwell, who will build on the strong foundation and lead the team toward future successes. With this, I believe, E3SM is well-positioned to continue its tradition of excellence and innovation in earth system science, serving the evolving critical DOE mission needs.
Due to the PI transition, personnel changes and evolving priorities, the E3SM project welcomes new Group Leads and Deputy Leads including Drs. Andrew Roberts, Darin Comeau, Chris Terai, Xue Zheng, Stephen Price, Hailong Wang and Aaron Donahue. Congratulations to all and thanks in advance for your contribution to a new chapter of the E3SM!
E3SM Future Vision
Looking into the future, the E3SM team is committed to continue pushing the high resolution frontier by fully embracing exascale and post-exascale architectures at DOE and to enthusiastically and strategically explore ML/AI technologies for improved model accuracy, accelerated performance and automated workflows toward a modeling system across a full spectrum of temporal and spatial scales.
ML/AI Status and Strategies
The E3SM project is on track to remain at the forefront of Earth system modeling research. The team’s plan for E3SMv4 (scheduled for release in 2028) would mark a significant milestone: the first coupled model release for exascale computers. This release will be accompanied by a set of emulators, enabling a wide range of scientific discoveries and applications aligned with DOE needs. This progress would build upon the strong momentum in ongoing ML/AI efforts, guided by the newly developed 5-pillar E3SM ML/AI strategy, along with fruitful collaborations with private industry and the leveraging of lab investments, exemplified by the recent E3SMv3 emulator release. These advancements would provide essential scientific insight into DOE’s energy infrastructure planning, resource management, and American energy innovation.
Exciting Achievement and Progress on Exascale Exploration
Atmosphere
During the past few years, the team has been steadfastly working on scaling up the simulation length of the internationally award winning SCREAMv1 (i.e., EAMxx 3km configuration), the first global model running on DOE’s exascale computer Frontier. After countless machine crashes and debugging, the team recently reached a major milestone of completing a 10 year simulation on Frontier! This simulation offers unprecedented data for improved scientific understanding of weather events statistics vital for energy infrastructure and national security.
Ocean
The Ocean Group has made exciting progress in the future E3SM Ocean Model for E3SM Global Applications (OMEGA), with the performance of OMEGA-0 promising that the final OMEGA model will provide E3SM with the ability to efficiently deliver ultra-high resolution simulations on DOE’s world-leading computing systems.
Land
The E3SM Land team is well on its way toward exascale readiness. The team has developed a portable Docker environment to support the ultra-high-resolution E3SM Land Model (uELM), targeting exascale computers with NVIDIA GPUs and OpenACC. Future plans include incorporating support for AMD GPUs and OpenMP, as well as establishing a multi-user uELM development environment on HPC clusters.
Cross Team Synergies
Given the synergies across different groups on exascale voyage, the EAMxx and OMEGA teams had a productive meeting at Sandia National Laboratories to collaborate on rewriting E3SM components in C++/Kokkos for DOE exascale computers, seeking to leverage each other’s work, avoid redundancy, and guide future component transitions. The teams mutually expressed their desire to break through component silos, maintain open communication, and continue in-person meetings to advance the broader project.
Contribution to a US Exascale Readiness Report
Earth system model’s exascale readiness has been of great interest not only to DOE, but also to US and international science community. An Exascale Readiness Report was recently produced by the High Performance Computing (HPC) Implementation Team as part of the Earth System Modeling and Prediction (ESM&P) subcommittee of The Interagency Council for Advancing Meteorological Services (ICAMS)’ Committee on Research and Innovation (CORI). The report authors surveyed 12 of the major U.S. modeling efforts and outlined their strategies for utilizing Exascale supercomputer architectures. E3SM’s Chief Computational Scientist Dr. Mark Taylor and Performance Coordinator Dr. Sarat Sreepathi were key contributors to this effort together with scientists from other national laboratories supported by multiple US agencies.
Improving Model Physics Remains Essential
While the team prioritizes the high resolution, exascale and ML/AI frontiers, improving model physics remains the central focus of the project for model fidelity, credibility, and as the foundation for trustworthy training data for AI emulations. Major progress in this aspect includes the documentation and evaluation of E3SMv2.1 with its improved representations of many key properties such as the ocean surface layer temperature, North Atlantic sea ice extent and many atmospheric variables.
Another key endeavor in improving model physics is bug fixes. In a model development project like E3SM, our team scientists spend enormous time finding and fixing bugs, resulting in improvements that can rival major parameterization changes. Starting from this issue, E3SM newsletter Floating Points is proud to introduce Heroic Bug Fixes, a recurring column that recognizes the critical yet often overlooked work of debugging.
Resources Newly Available: Data, Software, Tutorials and All Hands Material
E3SM project is excited to announce 1) the release of E3SM-Unified 1.11.1, the latest version of the unified Conda (and Spack) environment that brings together all E3SM-supported analysis tools into one easy-to-install package; 2) the availability of the biogeochemistry Human-Earth System Feedbacks group’s v2.1 data, the updates to the E3SM data availability and data publication policies moving forward; 3) a summary of the recent E3SM All Hands Meeting including meeting presentation material; and 4) the E3SM EAMxx and Offline Land Model Testbed (OLMT) Tutorials.
Research Highlights
Selected research highlights in this issue focus on the advanced understanding of the complex interplay between Earth system changes, human activities like irrigation and dam construction, and advancements in modeling techniques. Specifically, they include how 1) Earth system changes affect future agricultural productivity; 2) irrigation expansion impact air humidity and moist-heat; 3) a new method to estimate plant productivity may improve environmental predictions; and 4) high-resolution modeling reveal detailed impacts of dam construction and deforestation.
E3SM Awards and Special Recognition
In the February All Hands Meeting, the project continued its annual tradition by recognizing its team members for their exceptional contributions, vital to E3SM’s advancement but often overlooked in traditional evaluations. The Executive Committee chose the awardees from Council member nominations: Drs. Alan Di Vittorio, Ziming Ke, Ben Hillman, Erin Thomas, Carolyn Begeman, Shixuan Zhang and Tom Vo. The team also took the opportunity to express their appreciation of Dr. Elizabeth Hunke, who retired recently, for “her seminal contributions to, and longstanding leadership of, sea-ice modeling for E3SM and the global community”.
In the All Hands meeting, Dr. Dave Bader reflected on the project’s first decade. As founding Council Chair, he presented Founder’s Awards to five individuals for their critical early contributions during what he called the “Dark Days,” when DOE’s ability to build a world-class modeling system was doubted. Each awardee – Drs. Chris Golaz, Renata McCoy, Jim Foucar, Jon Wolfe and Kate Calvin – received an inscribed crystal paperweight, symbolizing their lasting impact.
Congratulations to all with my deepest appreciation to the awardees for your exceptional contributions in establishing, shaping and continued advancement of E3SM!
Concluding Remarks
At the end of the message, I want to thank Peter Caldwell, Ruby Leung, Mark Taylor, Renata McCoy and the full E3SM Leadership Team for their extraordinary effort during the past few months in developing the new decadal strategic plan which aligns the project priorities to best support the evolving DOE priority mission needs. I also thank the full E3SM team and E3SM ecosystem science community for your continued dedication and contribution to E3SM during this time!
Sincerely,
Xujing
