H040
AI Advances in Subsurface Hydrology and Energy
Session ID#: 248537
Session Description:
The integration of artificial intelligence (AI) and data science into subsurface hydrology and energy is transforming our capabilities to model and manage complex processes. This session invites innovative applications of AI, including machine learning, deep learning, and large language models, to groundwater flow, contaminant transport, geothermal energy, carbon sequestration, underground hydrogen storage, and hydrocarbon recovery. This session aims to address challenges related to data scarcity, data diversity, AI-based model uncertainty, surrogate models, and multi-scale heterogeneity, as well as interdisciplinary efforts that combine AI with physics-based models, laboratory measurements, and field data. We encourage contributions that discuss interpretability, robustness, and the deployment of AI in decision-making for sustainable energy and subsurface water resources. This session offers a transformative opportunity for hydrologists, hydrogeologists, geoscientists, energy researchers, and data scientists to foster transdisciplinary collaborations, facilitate knowledge exchange, and advance the frontiers of AI in subsurface studies.
Co-Sponsor(s):
Index Terms:
1402 - Critical Zone [CRITICAL ZONE]
1832 Groundwater transport [HYDROLOGY]
5104 Fracture and flow [PHYSICAL PROPERTIES OF ROCKS]
5139 Transport properties [PHYSICAL PROPERTIES OF ROCKS]
Primary Convener: Dr. Behzad Ghanbarian, University of Texas at Arlington, Arlington, TX, United States
Conveners: Maruti K. Mudunuru, Pacific Northwest National Laboratory, Subsurface Science Group, Richland, WA, United States and Alexandre M Tartakovsky, University of Illinois at Urbana Champaign, Civil and Environmental Engineering, Champaign, United States
Student/Early Career Convener: David A Barajas-Solano, Pacific Northwest National Laboratory, Richland, WA, United States
Sensitivity-Constrained Neural Operators for Forward and Inverse Problems in High-Dimensional PDEs (Invited) (1851568)
Chaopeng Shen, Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, United States, Abdolmehdi Behroozi, Pennsylvania State University Main Campus, Department of Civil and Environmental Engineering, University Park, PA, United States and Daniel Kifer, Pennsylvania State University, Department of Computer Science & Engineering, University Park, United States
Simulator-in-the-Loop: Differentiable, Simulation-Driven Learning for Heterogeneous Storage Reservoir Characterization (1857004)
Harun Ur Rashid, Los Alamos National Laboratory, Los Alamos, NM, United States, Aleksandra Pachalieva, Los Alamos National Laboratory, Computational Earth Science, Los Alamos, NM, United States and Daniel O'Malley, Los Alamos National Laboratory, Earth and Environmental Sciences Division, Los Alamos, United States
Uncertainty-Aware Area of Review Assessment via Probabilistic Committee Machine Learning: Application to Large-Scale Carbon Storage Reservoirs (Invited) (1873686)
Alexander Y Sun, National Energy Technology Laboratory Pittsburgh, Pittsburgh, PA, United States, Chung Yan Shih, National Energy Technology Laboratory Support Contractor / Leidos, Pittsburgh, United States and Hema Siriwardane, National Energy Technology Laboratory Morgantown, Morgantown, United States
The development of an AI-based tool for tracer selection and test design in geothermal energy applications (1878069)
Luka Vucinic1, Maria Isabel Freitas da Silva Vucinic1,2, David O'Connell3, Hakim Mezali1, Sofie de Volder4, Amael Poulain4, Fatima Ajia1, Catherine Coxon2, Laurence Gill3 and Dragan Savic5,6, (1)Glasgow Caledonian University London, London, United Kingdom, (2)Trinity College Dublin, Department of Geology, Dublin, Ireland, (3)Trinity College Dublin, Department of Civil, Structural and Environmental Engineering, Dublin, Ireland, (4)Traqua, Namur, Belgium, (5)University of Exeter, Centre for Water Systems, Exeter, United Kingdom, (6)KWR Water Research Institute, Nieuwegein, Netherlands
Comparing Physics-Based and Machine Learning Approaches for Sustainable Aquifer Management in Emilia-Romagna (Italy) (1912581)
Ilaria Delfini1, Daniel Zamrsky2,3 and Alberto Montanari1, (1)University of Bologna, Department of Civil, Chemical, Environmental, and Materials Engineering, Bologna, Italy, (2)Utrecht University, Department of Physical Geography, Utrecht, Netherlands, (3)IHE Delft Institute for Water Education, Water Resources and Ecology Department, Delft, Netherlands
Discrete Spatial Diffusion Framework and Assessment Metrics for Rock Image Generation (1934429)
Qianqian Zhou1, Javier E. Santos2, Agnese Marcato3, Masa Prodanovic1 and Michael Pyrcz1, (1)University of Texas at Austin, Austin, TX, United States, (2)Los Alamos National Laboratory, Los Alamos, United States, (3)Los Alamos National Laboratory, Earth and Environmental Sciences, Los Alamos, United States
Physics-informed machine learning for estimating permeability and dispersivity distributions in three-dimensional sandbox experiments (1939482)
Hongkyu Yoon, Sandia National Laboratories, Geomechanics Department, Albuquerque, NM, United States and Jonghyun Harry Lee, University of Hawai‘i at Mānoa, Civil, Environmental and Construction Engineering/ Water Resources Research Center, Honolulu, United States
Differentiable Physics-constrained subsurface inversion with Latent Diffusion model (1982205)
Honghui Du, University of Minnesota Twin Cities, Minneapolis, MN, United States and Qizhi He, University of Minnesota, Department of Civil, Environmental, and Geo- Engineerin, Minneapolis, United States
Latent-Space Surrogate Model for Fast Time-Lapse ERT Inversion (1982388)
Yuanzhe Wang, University of Illinois at Urbana Champaign, Civil and Environmental Engineering, Urbana, United States, Glenn E Hammond, Sandia National Laboratories, Applied Systems Analysis & Research, Albuquerque, United States, Tim C Johnson, Pacific Northwest National Laboratory, Richland, United States and Alexandre M Tartakovsky, University of Illinois at Urbana Champaign, Civil and Environmental Engineering, Champaign, United States
Assessing shallow geothermal potential by spatiotemporal geostatistics: a case study in Hsinchu, Taiwan (1990516)
Hwa-Lung Yu1, Sheng-Che Hsu2, Jia-Lun Wang2 and Huating Tseng2, (1)National Taiwan University, Department of Bioenvironmental Systems Engineering, Taipei, Taiwan, (2)National Taiwan University, Bioenvironmental Systems Engineering, Taipei, Taiwan
VAE-DNN: Energy-Efficient Trainable-by-Parts Surrogate Model for Groundwater Flow Modeling and Parameter Estimation (1994511)
Alexandre M Tartakovsky1, Yifei Zong1,2 and Yuanzhe Wang3, (1)University of Illinois at Urbana Champaign, Civil and Environmental Engineering, Champaign, United States, (2)Zhengzhou University, School of Water Conservancy and Transportation, Zhengzhou, China, (3)University of Illinois at Urbana Champaign, Civil and Environmental Engineering, Urbana, United States
Multiscale Inversion of Subsurface Heterogeneity Using Hierarchical Geostatistics and Deep Learning (1996237)
Farzad Moeini, University of Cincinnati Main Campus, Cincinnati, OH, United States, Chuanjun Zhan, Jilin University, Changchun, China, Zhenxue Dai, Jilin University, College of Construction Engineering, Changchun, China, Kenneth C Carroll, New Mexico State University, Department of Plant and Environmental Sciences, Las Cruces, United States and Reza Soltanian, University of Cincinnati Main Campus, Department of Geosciences, Cincinnati, United States
