MULTI-STAGE ALTERATION IN COSO GEOTHERMAL SYSTEM AS REVEALED BY MACHINE LEARNING
Loading...
Date
2025-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
The Ohio State University
Abstract
The U.S. Dept. of Energy (DOE) initiated a program to explore the potential for Enhanced Geothermal Systems (EGS) wherein hydraulic fracturing creates new fractures and re-opens existing ones to increase porosity and permeability. The DOE designated several active geothermal areas in the U.S. to test this concept such as the Coso Geothermal Area in California. This research product focused on the modal mineralogy, the alteration sequences, and late-stage calcite fill and veins in well 42A-16. Nine thin sections from depths ranging from 1000 ft to 8930 ft were imaged using Scanning Electron Microscopy (SEM) for mineral chemical composition. Qualitative Evaluation of Minerals by Scanning Electron Microscopy (QEMSCAN) was used for the initial creation of modal mineralogy and colored scanning. Image processing through ImageJ and machine learning tool, ilastik, were used to better reconstruct modal mineralogy, and understand the relationship between calcite alongside other minerals as a product of temperature, pressure, and depth, within the system. The calcite shows strong correlations with major rock forming minerals quartz, plagioclase, and potassium feldspar throughout the well, and alteration phases such as chlorite and amphiboles at depth. The order of the alteration sequence in Coso lay the framework for an environment conducive to late-stage calcite vein precipitation. Results also suggest calcite vein fill decreased modern permeability and porosity by a magnitude of 10. This study could serve as a reason to further investigate the effects of EGS on carbonates and other minerals within a system that is out of equilibrium.
Description
Keywords
Geothermal Energy, Machine Learning