Artificial intelligence for multiscale subsurface flow in the energy transition
Abstract
Artificial intelligence is changing how we observe and predict subsurface flow. Pore‑scale research made this shift visible first: segmentation, generative reconstruction, super‑resolution and rapid property prediction are making the analysis and physical interpretation of large datasets easier, faster and more accurate. In parallel, rapid predictions for CO₂ storage, underground hydrogen storage and hydrocarbon recovery continue to advance. However, understanding flow in the subsurface is inherently a multi scale process both in space ‑ from micron‑sized pores that control displacement processes to carbon dioxide storage sites extending over tens of kilometres ‑ and time ‑ from millisecond displacement to secure storage over millennia. To address this challenge with machine learning tools demands transferability, physical consistency and transparent provenance. In this editorial we explore the current state‑of‑the‑art and provide a personal perspective for the future of artificial intelligence for flow in porous media for the energy transition.
Document Type: Editorial
Cited as: Zhu, L., Cai, J., Blunt, M. J. Artificial intelligence for multiscale subsurface flow in the energy transition. Advances in Geo‑Energy Research, 2026, 21(1): 98‑100. https://doi.org/10.46690/ager.2026.07.10
DOI:
https://doi.org/10.46690/ager.2026.07.10Keywords:
Artiffcial intelligence, pore-scale imaging, digital rock physics, multiscale subsurface ffow, geological carbon storage, underground hydrogen storageReferences
Andrew, M., Bijeljic, B., Blunt, M. J. Pore‑scale imaging of trapped supercritical carbon dioxide in sandstones and carbonates. International Journal of Greenhouse Gas Control, 2014, 22: 1‑14.
Andrew, M., Menke, H., Blunt, M. J., et al. The imaging of dynamic multiphase fluid flow using synchrotron‑based X‑ray microtomography at reservoir conditions. Transport in Porous Media, 2015, 110(1): 1‑24.
Blunt, M. J. Multiphase Flow in Permeable Media: A Pore‑Scale Perspective. Cambridge University Press, Cambridge, 2017.
Bultreys, T., Ellman, S., Schleputz, C. M., et al. 4D micro‑velocity reveals multiphase flow field perturbations in porous media. Proceedings of the National Academy of Sciences of the United States of America, 2024, 121(12): e2316723121.
Fu, S., Mao, S., Carbonero, A., et al. Deep learning‑based surrogate modeling for underground hydrogen storage. Advances in Water Resources, 2025, 203: 105014.
Gu, Y., Spurin, C., Wen, G. Learning pore‑scale multi‑phase flow from experimental data with graph neural network. arXiv:2411.14192, 2024.
Jiang, F., Guo, Y., Tsuji, T., et al. Upscaling permeability using multiscale X‑ray‑CT images with digital rock modeling and deep learning techniques. Water Resources Research, 2023, 59(3): e2022WR033267.
Ma, Z., Bijeljic, B., Wen, G., et al. Super‑resolution imaging of multiphase fluid distributions in porous media using deep learning. Transport in Porous Media, 2025, 152(10): 85.
Mosser, L., Dubrule, O., Blunt, M. J. Reconstruction of three‑dimensional porous media using generative adversarial neural networks. Physical Review E, 2017, 96(4): 043309.
Santos, J. E., Xu, D., Jo, H., et al. PoreFlow‑Net: A 3D convolutional neural network to predict fluid flow through porous media. Advances in Water Resources, 2020, 138: 103539.
Spurin, C., Bultreys, T., Ricker, M., et al. Real‑time imaging reveals distinct pore‑scale dynamics during transient and equilibrium subsurface multiphase flow. Water Resources Research, 2020, 56(12): e2020WR028287.
Tang, K., Armstrong, R. T., Mostaghimi, P., et al. Scaling deep learning for material imaging with a pseudo 3D model for domain transfer. Nature Communications, 2025, 16(1): 11293.
Thaysen, E. M., Butler, I. B., Hassanpouryouzband, A., et al. Time‑resolved 2D and 3D imaging of hydrogen and brine displacement processes in porous Clashach sandstone. Journal of Colloid and Interface Science, 2025, 694: 137704.
Wang, Y., Chu, H., Lyu, X. Deep learning in CO₂ geological utilization and storage: Recent advances and perspectives. Advances in Geo‑Energy Research, 2024, 13(3): 161‑165.
Wang, Y. D., Shabaninejad, M., Armstrong, R. T., et al. Deep neural networks for improving physical accuracy of 2D and 3D multi‑mineral segmentation of rock micro‑CT images. Applied Soft Computing, 2021, 104: 107185.
Wang, J., Yang, Y., Zhang, Q., et al. Pore scale modeling of wettability impact on CO₂ capillary and dissolution trapping in natural porous media. Advances in Water Resources, 2025, 201: 104982.
Wang, C., Zhu, L., Gu, Y., et al. Learning pore‑scale multiphase flow from 4D velocimetry. arXiv:2603.12516, 2026.
Wen, G., Li, Z., Long, Q., et al. Real‑time high‑resolution CO₂ geological storage prediction using nested Fourier neural operators. Energy & Environmental Science, 2023, 16(4): 1732‑1741.
Xie, C., Zhu, J., Wang, J., et al. Direct prediction of relative permeability curve from 3D digital rock images based on deep learning approaches. International Journal of Rock Mechanics and Mining Sciences, 2023, 170: 105544.
Ye, S., Song, X., Ma, Z., et al. A noise‑resistant and annotation‑free supervoxel‑based algorithm for rapid segmentation of multiphase X‑ray images. Advances in Geo‑Energy Research, 2025, 16(1): 50‑59.
Zhang, R., Song, X., Zhu, L., et al. SAMamba3D: Adapting segment anything for generalizable three‑dimensional segmentation of multiphase pore‑scale images. Advances in Geo‑Energy Research, 2026, 21(2): 109‑124.
Zhao, J., Wang, F., Cai, J. 3D tight sandstone digital rock reconstruction with deep learning. Journal of Petroleum Science and Engineering, 2021, 207: 109020.
Zhu, L., Bijeljic, B., Blunt, M. J. Generation of pore‑space images using improved pyramid Wasserstein generative adversarial networks. Advances in Water Resources, 2024, 190: 104748.
Zhu, L., Bijeljic, B., Blunt, M. J. Diffusion model‑based generation of three‑dimensional multiphase pore‑scale images. Transport in Porous Media, 2025, 152(3): 22.
Downloads
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Linqi Zhu, Jianchao Cai, Martin J. Blunt
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.