5:15 PM - 7:15 PM
[SCG45-P07] Introduction of continuous AE measurement system to laboratory experiments using a sub-meter scale specimen
Keywords:Induced Seismicity, Fluid Injection, Acoustic Emission, Laboratory Experiment
Induced seismicity in resource engineering is mainly caused by fault slip triggered by fluid injection. Fault slip can occur when shear stress on the fault plane exceeds static friction, which depends on normal stress and friction coefficient. Fluid injection reduces effective normal stress, static friction and making fault slip more likely to occur. In fact, there have been cases where the injection of fluids during unconventional resource development and geothermal development, and CO2 injection in CCS have increased the seismicity of the target regions (Zoback and Gorelick, 2012; Schultz et al., 2020). Since some induced seismicity became felt earthquakes and posed risks to economic activity, we need to investigate the mechanism of fault slip associated with fluid injection. In Ito et al. (2024), they conducted laboratory experiments applying biaxial compressive stress to a rock specimen and injecting fluid through a borehole to the fault plane, and they measured shear strains along the fault plane. However, measurement and analysis of shear strain alone does not capture the entire picture of fault slip behavior because it did not include a viewpoint from the elastic waves generated by fault slip. This study introduces a continuous Acoustic Emission (AE) measurement system to capture AE, enabling spatiotemporal analysis of fault slip along with shear strain.
The employed AE system is capable of continuous recording for over an hour at a high sampling rate of 2 MHz. Elastic waves are detected by AE sensors (AE-900-B, NF Corp.) , amplified by amplifiers (2/4/6C, MISTRAS Group Inc.), then digitized by A/D converters (PXIe-6396, National Instruments Corp.). Data is transferred via a Thunderbolt 3 cable to an SSD (Extreme Pro Portable SSD, SanDisk Corp.).
Before the introduction of this system to the apparatus, a measurement stress test ensured system stability. Then, a drop-ball test verified AE measurement process and a series of AE analysis. In this test, we dropped a metal ball onto the rock specimen multiple times, detected AE events caused by the dropped ball using STA/LTA method, picked P-wave arrivals manually, and located hypocenters using Single Event Determination (SED) method. As a result, we were able to locate the hypocenter for all events, with an average spatial error of about 2-5 mm.
The rock specimen used in this study is the 600 x 600 x 600 mm3 Mogami andesite used by Ito et al. (2024). The specimen includes the saw cut fault at an angle of 45° to simulate a fault plane, and a fluid injection hole perpendicular to the fault plane was drilled to the center of the fault plane. There are three grooves on the fault plane, and 39 strain gauges (KFGS-2-120-D31-11, KYOWA Electronics Co., Ltd.) were buried with epoxy resin inside them. In addition, we attached 12 AE sensors to the surface of the specimen. The rock specimens are loaded with biaxial compressive stress using flat jacks, and we measure the shear strain at 10 Hz and AE at 2 MHz. We analyze AE data by STA/LTA method for AE event detection and P-wave arrivals detection, and SED method for hypocenter location. We also plan to analyze AE data by advanced seismic wave analysis methods such as hypocenter relocation using the Double Difference (DD) method (Waldhauser and Ellsworth, 2020) and machine learning (Naoi et al., 2022).
The employed AE system is capable of continuous recording for over an hour at a high sampling rate of 2 MHz. Elastic waves are detected by AE sensors (AE-900-B, NF Corp.) , amplified by amplifiers (2/4/6C, MISTRAS Group Inc.), then digitized by A/D converters (PXIe-6396, National Instruments Corp.). Data is transferred via a Thunderbolt 3 cable to an SSD (Extreme Pro Portable SSD, SanDisk Corp.).
Before the introduction of this system to the apparatus, a measurement stress test ensured system stability. Then, a drop-ball test verified AE measurement process and a series of AE analysis. In this test, we dropped a metal ball onto the rock specimen multiple times, detected AE events caused by the dropped ball using STA/LTA method, picked P-wave arrivals manually, and located hypocenters using Single Event Determination (SED) method. As a result, we were able to locate the hypocenter for all events, with an average spatial error of about 2-5 mm.
The rock specimen used in this study is the 600 x 600 x 600 mm3 Mogami andesite used by Ito et al. (2024). The specimen includes the saw cut fault at an angle of 45° to simulate a fault plane, and a fluid injection hole perpendicular to the fault plane was drilled to the center of the fault plane. There are three grooves on the fault plane, and 39 strain gauges (KFGS-2-120-D31-11, KYOWA Electronics Co., Ltd.) were buried with epoxy resin inside them. In addition, we attached 12 AE sensors to the surface of the specimen. The rock specimens are loaded with biaxial compressive stress using flat jacks, and we measure the shear strain at 10 Hz and AE at 2 MHz. We analyze AE data by STA/LTA method for AE event detection and P-wave arrivals detection, and SED method for hypocenter location. We also plan to analyze AE data by advanced seismic wave analysis methods such as hypocenter relocation using the Double Difference (DD) method (Waldhauser and Ellsworth, 2020) and machine learning (Naoi et al., 2022).