Back to Top
Skip to main content
Deep Learning Tool To Ensure Safe Carbon Storage Developed in NETL-Sponsored Project
Animated depiction of an underground carbon storage example.

Ensuring safe carbon storage operations is key to achieving a carbon emissions-free economy by 2050.

Zanskar Geothermal and Minerals Inc. (Zanskar), with NETL support, recently concluded a project that developed a deep learning tool for subsurface monitoring that could help ensure safe storage of carbon dioxide (CO2) at geologic sites, which is critical for meeting the nation’s decarbonization goals.

“Before making major investments, carbon storage site operators are looking for assurances from the scientific community that their operations will be safe,” said NETL’s James Gardiner, project manager for NETL’s Carbon Transport and Storage Team. “By monitoring the subsurface with sensors, researchers can detect changes underground. In the case of this project, Zanskar is monitoring for seismic activity — vibrations in the earth.”

Seismicity could cause instability at a geologic storage site, so it’s important for researchers to better characterize these locations and determine what, if any, seismic activity is occurring before, during or after CO2 injection. A type of fiber-optic sensing technology called distributed acoustic sensing (DAS) shows great promise as a seismic monitoring tool, but current data management and processing methods associated with DAS are not sufficient to fully realize the technology’s potential.

“Zanskar worked to overcome these obstacles by adapting and optimizing novel deep-learning techniques to improve the detectability of certain signals of interest, including local, regional and distant earthquakes,” Gardiner said. “Deep learning is a type of machine learning that uses artificial neural networks.”

In this project, the tool was trained on DAS measurements using convolutional neural networks. This deep-learning approach resulted in a data filter that has better computational efficiency and an improved signal-to-noise ratio relative to other commonly applied filters. The benefits of this deep-learning tool include faster and more accurate real-time monitoring, identification of more seismic events and improved subsurface imaging.

This project was funded with a Phase I award through the Small Business Innovation Research (SBIR) program, which encourages domestic small businesses to engage in federal research and research and development with the potential for commercialization. The technology could be further developed into a real-time monitoring tool, and this project was recently selected for an SBIR Phase II award, where researchers will develop a seismic monitoring workflow that could be deployed at carbon storage sites.

NETL is a U.S. Department of Energy national laboratory that drives innovation and delivers technological solutions for an environmentally sustainable and prosperous energy future. By using its world-class talent and research facilities, NETL is ensuring affordable, abundant, and reliable energy that drives a robust economy and national security, while developing technologies to manage carbon across the full life cycle, enabling environmental sustainability for all Americans.