Improving geothermal resource assessment: a data-driven approach to chemical geothermometry using deep learning.

AlGaiar, Mahmoud

Abstract

This study presents a deep learning model trained on a dataset of 674 water samples from Nevada to predict geothermal reservoir temperatures. The model outperforms traditional geothermometers and other machine learning models, achieving high accuracy and demonstrating global applicability when tested on samples from different geothermal fields around the world.

Citation

ALGAIAR, M. 2025. Improving geothermal resource assessment: a data-driven approach to chemical geothermometry using deep learning. Presented at the 2025 SEG (Society of Exploration Geophysicists) Net-zero emissions workshop, 23-25 June 2025, [virtual event].

• Presentation Conference Type: Presentation / Talk
• Conference Name: 2025 SEG (Society of Exploration Geophysicists) Net-zero emissions workshop
• Start Date: Jun 23, 2025
• End Date: Jun 25, 2025
• Deposit Date: Jul 7, 2025
• Publicly Available Date: Jul 7, 2025
• Peer Reviewed: Peer Reviewed
• Keywords: Geothermal resource exploration; Hidden/blind geothermal resources; Artificial intelligence; Machine learning; Deep learning
• Public URL: https://rgu-repository.worktribe.com/output/2922507
• Additional Information: The file accompanying this record contains the extended abstract and slides presented at the conference which have been incorporated into a single file on this repository.