| Title | 3D Seismic Imaging and Fault Detection at the Lightning Dock Geothermal Field |
|---|---|
| Authors | Lianjie HUANG, Kai GAO, David LI, Chenglong DUAN, Trenton CLADOUHOS, Yingcai ZHENG, Boming WU, and Michael SWYER |
| Year | 2024 |
| Conference | Stanford Geothermal Workshop |
| Keywords | Fault detection, first-arrival traveltime tomography, full-waveform inversion, Lightning Dock geothermal field, machine learning, reverse-time migration, siting geothermal wells |
| Abstract | New wells are often needed to improve energy production in geothermal power plants. Integrating 3D seismic imaging and fault detection with geological and geochemical information can help site new geothermal injection wells to reduce drilling risk and ensure the sustainability of geothermal power production. Lightning Dock Geothermal (LDG) LLC conducted a 3D active-source surface seismic survey in 2011 using accelerated weight drop sources for subsurface characterization. We use an open-source data analysis package called Madagascar to process the raw seismic data, update the velocity model in the shallow region using first-arrival traveltime tomography, improve the entire 3D surface velocity model using full-waveform inversion, and produce a 3D subsurface image of the Lightning Dock geothermal field using reverse-time migration. We then detect faults on the 3D seismic image using a machine learning algorithm based on nested residual U-Net. Our 3D seismic imaging and fault detection results provide valuable information for siting new geothermal wells at the Lightning Dock geothermal field. |