Machine listening to geothermal micro-and laboratory nano-seismicity: can we improve heat mining safety and efficacy?
Ben Holtzman
Lamont-Doherty Earth Observatory
- Date & Time
- Location
- Online-only seminar via Microsoft Teams
- Host
- Leah Langer
- Summary
Geothermal heat mining has the potential to become a significant contributor to a clean energy transition. However, significant problems must be overcome to safely and effectively engineer fluid pathways through fracture networks. Multiple approaches to unsupervised machine learning in seismology have the potential to help us discover complex and subtle patterns in acoustic signals. Physics-constrained learning, though not discussed here in detail, has the potential to associate patterns in multiple data types in time, micro- and macroscopic. Our aim is to be able to associate changes in a reservoir's acoustic signals with changes in its thermal-mechanical state and fracture processes. Towards this, we are building sets of analyses on active reservoirs and laboratory experiments. I will show results from two sets of (unpublished, ongoing) studies- first on a local section in the northwestern corner of The Geysers geothermal field in Northern CA, and second on a set of deformation experiments on basalt.
Closed captions are typically available a few days after the seminar. To turn them on, press the ‘CC’ button on the video player. For older seminars that don’t have closed captions, please email us, and we will do our best to accommodate your request.