View description

Session: ESC-SSA Joint Session: Advanced Methods for Harnessing Seismic Noise Analysis in Applied Seismology - I

Type: Oral

Room: Ballroom B

Date: 4/15/2026

Presentation Time: 08:30 AM Pacific

Presenting Author: Peter Malin

Description

In 2023 and 2025 we deployed 982 and 1320 nodal seismographs in the area surrounding the Newberry Volcano Geothermal Field’s 3 km deep 55-29 test well. These arrays were used as ambient seismic detectors for potential non-earthquake seismic signals from permeable, fluid-filled, fractures. The 2023 survey covered ~27 km² and focused on a ~1.6x1.6x1.8 km³ imaging volume around 55-29. It was designed to image down to ~2.2 km. The 2025 survey covered ~37 km² and was aimed at extending this coverage to ~3.2 km. Both networks were operated at 500 Hz sampling for several days (5 and 11). The data were processed using the 1-way travel time, fracture-source to surface-receiver, PFI^TM steaming-migration method. A critical step was the development of an accurate velocity model for use in focusing the PFI stack. This was accomplished in 2023 by improving previous velocity models. In 2025 we used a DAS-based calibration method that gave accurate travel times from near the bottom of 55-29.

The 2023 survey was conducted during environmentally and culturally quite time. The time-lapse imaging detected near-and-far well signals (~0 to ~1.2 km offset) and seismically dead zones. Based on previous verified interpretations of PFI maps, we identify these signals as coming from fluid-filled fractures, originating from local microstrain induced slip events. The dead zones coincide with the clay cap. Signal orientations in the active zone are consistent with the regional geology. Our interpretation in terms of fractures is also supported by a low-pressure stimulation in 55-29 (~22 bars), which activated signals up to distances of 0.4 km. The 2025 survey was conducted during a period of high industrial activity with only a score of quiet hours, significantly interferring with shallow signal processing. Consequently, our efforts were focused on the deeper volume around 55-29, were we detected stimulation generated signals. In addition to geothermal resource mapping, we suggest that PFI can also be used to directly profile the pluming of volcanic systems.

Additional Authors

Peter Malin

Presenting Author Corresponding Author

malin@duke.edu

Duke University, Durham, North Carolina, United States

Jeff Eppink

jeppink@enegis.com