LLNL, partners open access to CO2 storage simulator
GEOSX is an open-source, high-performance simulator to help researchers better understand large-scale geological carbon dioxide (CO2) storage. The figure shows a simulation of the fluid pressure distribution in a faulted reservoir due to CO2 injection through multiple wells. The field in the image is an offshore field in the Gulf of Mexico, near Texas. It is not actively injecting CO2, but is being used as analog site to study what CO2 storage in the gulf would look like, and what risks need to be accounted for in a future project design. Credit: Geologic data courtesy Gulf Coast Carbon Center.
After more than two years of joint research, Lawrence Livermore National Laboratory (LLNL), Total and Stanford University are releasing an open-source, high-performance simulator for large-scale geological carbon dioxide (CO2) storage.
The GEOSX simulator will enable researchers around the world to build on the work of the three partners, providing an open framework to accelerate the development of carbon capture, utilization and storage (CCUS) technologies. It also has application to a variety of other subsurface energy technologies.
The simulator uses field data to predict the behavior and impact of CO2 stored in deep geological repositories. Its resolution and speed were made possible by new technological developments in the fields of algorithmics and high performance computing. It will be used to improve the management and security of geological CO2 repositories and plan for the widespread implementation of these projects at an industrial scale.
“We are proud of the development of GEOSX, the first tool able to simulate the geological storage of CO2 at this scale. Making its source code openly available is intended to offer researchers around the world the opportunity to participate in the development of CCUS, for which Total has been mobilizing significant resources for several years,” said Marie-Noëlle Semeria, Total's chief technology officer. “GEOSX brings us closer to the position of technological leadership that we are seeking to achieve throughout the CCUS chain, in line with our ambition to get to net zero by 2050.”
“GEOSX combines breakthrough scientific computing and high-performance technologies that will allow Total and the community to accelerate the implementation of future large-scale storage projects, necessary to achieve carbon neutrality objectives on a global scale,” said Hamdi Tchelepi, chairman of the Energy Resources Engineering Department and the project’s lead research scientist for Stanford. “The innovations resulting from our joint research program open the door to a new era for the modeling of geological storage of CO2.”
“We are excited to be working with Stanford and Total to support widespread deployment of carbon capture, utilization and storage projects,” said Pat Falcone, LLNL’s deputy director for Science and Technology. “Climate change is a global challenge, and we view CCUS as a critical component in the transition to a low-carbon economy. It also is thrilling to observe the cutting-edge capabilities being enabled by our next-generation supercomputing platforms and gratifying to have created this new capability as a part of such an effective industry, university and national laboratory partnership.”
GEOSX is a simulation platform focused on multiphysics and hardware portability. Funding for its development has been provided by the DOE Exascale Computing Project through the Exascale Subsurface Simulator Project, and by Total through the FC-MAELSTROM Project. It draws on Total, Stanford and LLNL’s 20-plus years of expertise in simulation and HPC research. GEOSX also benefits from significant investments in HPC tools through LLNL’s RADIUSS Project (Rapid Application Development via an Institutional Universal Software Stack), notably the LvArray, RAJA, CHAI, Umpire and Caliper packages.
GEOSX source code
RADIUSS software portal
TagsEarth and Atmospheric Science
Atmospheric, Earth, and Energy
HPC, Simulation, and Data Science
Physical and Life Sciences