Summary of Research
A fundamental challenge in Earth and Energy Sciences is accurately modeling, mapping and monitoring the subsurface structures that control fluid flow through fractured media. Subsurface fluid flow depends on connectivity, geometry, and properties of fractures (e.g., aperture), as well as the interaction with the surrounding matrix. Current approaches (such as discrete fracture networks) for fluid flow and transport modeling require robust knowledge of the subsurface to create accurate models, therefore, high-resolution characterization of fractures and their evolution can increase model accuracy and is essential to understanding fracture-flow dynamics.
I aim to tackle the challenge of identifying and modeling subsurface fracture networks, their evolution and their influence in fluid flow by integrating high-resolution distributed acoustic sensing (DAS) data with machine learning. I hypothesize that fractures, and consequently fracture networks, can be characterized by using their unique seismo-acoustic signatures. My approach involves the integration of laboratory and field-scale experiments, which will be integrated using a machine learning approach and simulations of using discrete fracture network and graph-based methods.
Addressing this challenge can provide a way to characterize fluid flow in fractured systems. The integration of acoustic signatures with machine learning and subsequently, fluid flow modeling, can improve the accuracy of fluid flow and transport models. The applications include, among others, enhanced geothermal energy, superhot rock, water management and CO2 storage.
Publications
Caraccioli P.D., Mordensky, S.P., DeAngelo, J., Burns, E.R., Lipor, J.J. (In press - Geothermics). Improving Machine Learning Models for Hydrothermal Resource Favorability by Selecting Representative Negatives and Relevant Features.
Caraccioli P.D., Mordensky, S.P., DeAngelo, J., Burns, E.R., Lipor, J.J. (2024) Separating Signals in Elevation Data Improves Supervised Machine Learning Predictions for Hydrothermal Favorability. Geothermal Rising Conference Transactions, 48, pp 2217-2236.
Caraccioli, P.D., Mordensky, S.P., Lindsey, C.R., DeAngelo, J., Burns, E.R., Lipor, J.J. (2023) Don't let negatives hold you back: accounting for underlying physics and natural distributions when selecting negative training sites leads to better machine learning predictions. Geothermal Rising Conference Transactions, 47, 1672-1693, Reno, Nevada, 1-5 October 2023.
Caraccioli P.D., Mordensky, S.P., DeAngelo, J., Burns, E.R., Lipor, J.J. Separating Signals in Elevation Data Improves Supervised Machine Learning Predictions for Hydrothermal Favorability. Geothermal Rising Conference. Waikoloa, HI, USA, 27-30 October, 2024 [Oral Presentation]
Caraccioli, P.D., Mordensky, S.P., Lindsey, C.R., DeAngelo, J., Burns, E.R., Lipor, J.J. (2023). Don't let negatives hold you back: accounting for underlying physics and natural distributions when selecting negative training sites leads to better machine learning predictions. Geothermal Rising Conference, Reno, Nevada, 1-5 October 2023 [Oral Presentation]
Caraccioli P.D., Mordensky, S.P., DeAngelo, J., Burns, E.R., Lipor, J.J. Improving Supervised Machine Learning Predictions for Hydrothermal Favorability by Separating Signals in Elevation Data. Geological Society of America Abstracts with Programs. Spokane, WA, USA, 15-17 May 2024 [Oral Presentation]
Caraccioli, P.D., Mordensky, S.P., Lindsey, C.R., DeAngelo, J., Burns, E.R, Lipor, J.J. Improving data-driven resource assessment by accounting for the expected natural distribution of hydrothermal systems. Geological Society of America Abstracts with Programs. Pittsburgh, PA, USA, 15-18 October 2023 [Oral Presentation]
Caraccioli Salinas, P.D, Mordensky, S.P., DeAngelo, J., Burns, E.R., Lipor, J.J. 2024, Separating Elevation Signals Improves Model Predictions. Portland State University-Louis Stokes Alliance For Minority Participation Symposium, June 2024 [Poster Session]
Caraccioli Salinas, P.D, Mordensky, S.P., Lindsey, C.R., DeAngelo, J., Burns, E.R., Lipor, J.J., 2023, Impartial Selection of Negative Sites Improves Model Predictions. Portland State University-Louis Stokes Alliance For Minority Participation Symposium, 9 June 2023 [Poster Session]
Chhun, Chanmaly, Rebecca Pearce, Pascal Caraccioli Salinas, Seth Saltiel, and Carolina Munoz Saez. Survey of Methods, Challenges, and Pathways Forward for Superhot Rock Characterization. PROCEEDINGS, 50th Workshop on Geothermal Reservoir Engineering Stanford University, Stanford, California, February 10-12, 2025.
Chhun, C., Pearce R., Caraccioli, P.D., Saltiel S., Munoz-Saez C., (2024). Bridging the Gaps: A Survey of Methods, Challenges, and Pathways Forward for Superhot Rock Siting and Characterization. Clean Air Task Force. https://cdn.catf.us/wp-content/uploads/2024/10/23204011/shr-siting-characterization-1.pdf
Burns, E.R., Mordensky, S.P., DeAngelo, J., Caraccioli, P.D., Lipor, J.J., and Williams, C.F., 2024, New Strategies to Assess the Electric-grade Geothermal Energy Resources of the Great Basin, USA, American Geophysical Union Annual Meeting (AGU24), 9-13 December 2024, Washington, DC, USA.
Burns, E.R., Mordensky, S.P., DeAngelo, J., Caraccioli, P.D., Lipor, J.J., and Williams, C.F., 2024, New Strategies to Assess the Electric-grade Geothermal Energy Resources of the Great Basin, USA, Geological Society of America Connects Annual Meeting, 22-25 September 2024, Anaheim, CA, USA.
Mordensky, S.P., Burns, E.R., Lipor, J.J., DeAngelo, J., Caraccioli, P.D., 2024, Modernizing the Workflow for the Hydrothermal Resource Assessment Favorability Maps for the Great Basin, USA. Geological Society of America Joint Cordilleran and Rocky Mountain Section – 120th and 74th Annual Meeting, Spokane, WA, 15-17 May 2024.
Mordensky, S.P., Burns, E.R., DeAngelo, J., Lipor, J.J., Caraccioli, P.D., 2024, A Data-Driven Approach for Assessing the Location and Scale of Geothermal Resources in the Great Basin, USA. Subsurface Geothermal Symposium: New Technology Drilling and Subsurface Characterization. American Association of Petroleum Geologists and Geothermal Rising, Reno, NV, 31 July – 2 August June 2024. [Oral/Invited]
Mordensky, S.P., Burns, E.R., Lipor, J.J., DeAngelo, J., Caraccioli, P.D. Machine Learning for Hydrothermal Prediction. U.S. Geological Survey Mineral Resource Assessment Training Seminar Series. 1 May 2024. [Oral/Invited/Virtual]
Mordensky, S.P., Burns, E. R., Lipor J. J., DeAngelo, J., Caraccioli P. D. Machine Learning Strategies for Mapping Favorability of Sparse Resources. Insights from the Geothermal Resources Investigations Project. USGS Mineral Resource Symposium. Denver, CO, USA, March 19-21, 2024
Mordensky, S.P., Burns, E. R., Lipor J. J., DeAngelo, J., Caraccioli P. D. Adapting Supervised Machine Learning Approaches for Hydrothermal Resource Assessments. Geological Society of America Abstracts with Programs. Pittsburgh, PA, USA, 15-18 October 2023.
Chhun, C., Pearce R., Caraccioli, P.D., Saltiel S., Munoz-Saez C., Bridging the Gap for Superhot Rock Geothermal Characterization and Siting: Insights from Mt. Kuju, American Geophysical Union Annual Meeting (AGU24), 9-13 December 2024, Washington, DC, USA.
Chhun, C., Pearce R., Caraccioli, P.D., Saltiel S., Munoz-Saez C., Identifying Technology and Knowledge Gaps for Characterization and Siting of Superhot Rock Geothermal Reservoirs to better inform policy pathways towards new research, development and demonstration. Geothermal Rising Conference. Waikoloa, HI, USA, 27-30 October, 2024.
Awards
Cornell Engineering Colman Fellowship, 2024.
Latin Honors Bachelors Degree: Summa Cum Laude.
GSA Student Travel Grant, Cordilleran Section, 2024.
Geothermal Rising Student Contest Paper Winner (Undergraduate), 2023.
Geothermal Rising Undergraduate Scholarship Award, 2023
WING Scholarship, 2023
Expanding Representation in Geosciences Scholarship, Geological Society of America, 2023.
Christina and David Vernier Endowed Scholars Program, 2023 - 2024
Barbara McDougal Endowed Scholarship, 2023 - 2024
Cliff and Deborah Burns Scholarship, 2023 - 2024.
McNair Scholars Program, 2023
President's List, Portland State University, Spring 2022, Fall 2022, Spring 2023, Fall 2023