11:30 AM - 11:45 AM
[SSS09-07] Utilization of S-wave splitting to monitor the underground heterogeneity in the Oku-Aizu geothermal field
Keywords:S-wave splitting, geothermal
To provide measures against global warming, geothermal power generation has attracted attention in Japan due to its rich geothermal resources. It is commonly known that a decline in geothermal power output occurs because of a reduction in underground hot water and steam during long-term power plant operations. Thus, continuous subsurface monitoring will help us to estimate the spatiotemporal changes in underground hot fluids and fractures. This study examines the potential of micro-seismic waveforms in geothermal areas as a low-cost and long-term subsurface monitoring tool. S-wave splitting is a phenomenon that seismic S-waves divide into two components while passing through an anisotropic medium: a faster and slower polarization direction. A faster component relates to the dominant fracture direction or mineral anisotropic axis, and a time lag between two splitting waves relates to the degree of fracture closure. We apply S-wave splitting analysis to multiple seismic waves to evaluate spatiotemporal variations in the local heterogeneity of the geothermal reservoir.
The performance of Yanaizu-Nishiyama Geothermal Power Plant in Fukushima, Japan, decreased from 65,000 kW in 1995 to 30,000 kW by 2017. To mitigate the reduction in steam production, a gravity-driven water injection test was started in 2015 and continues to date. We focus on 267 micro-seismic events that occurred after November 2017. There are nine three-component velocity seismometers with a sampling frequency of 1000 Hz placed around the power plant: five stations are on the surface (YAE-1 to YAE-5) and four in the boreholes (YAE-6 to YAE-9). We select the waveforms that pass through geothermal reservoirs. The analysis window is 0.30 seconds, starting 0.05 seconds before the S-wave arrival. We calculate the cross-correlation between N-S and E-W components with a bandpass filter from 3 to 60 Hz. Rotating the coordinate axis in five-degree increments from 0 to 90 degrees, we calculate the rotation angle and cross-correlation value for each degree at 0.001-second intervals. For each event, we record the rotation angle, the time lag, and the cross-correlation value at the highest cross-correlation value.
We compare the results before (November 2017 to July 2018) and after (July 2019 to March 2020) the major water recharge. We classify stations into three groups: reservoir areas (YAE-2, 5, 8, 9), southern areas on the reinjection well area (YAE-1, 7), and eastern areas on the reinjection well (YAE-3, 4, 6). The faster S wave directions have changed at multiple stations in the reservoir area and the water injection area before and after the water injection. Significant faster S wave direction changes are observed at borehole stations YAE-8 and 9. In addition, the faster S wave directions are mainly divided into two directions: NW-SE and NE-SW. The direction of the faster S-wave is generally considered to be the long axis of the heterogeneity and the direction of the maximum principal stress. Thus, the faster S wave direction reflects the predominant heterogeneous structure, surrounding faults, and fracturing systems in the geothermal area. NW-SE direction closely aligns with the direction of the maximum principal stress axis from the Japanese crustal stress database (AIST, 2024). This direction is also consistent with the orientation of faults near the recharge well (Dian et al., 2024). NE-SW direction corresponds to the major fault orientation located northwest of the reinjection well. Due to recharging, the time lag between two S waves is confirmed to increase at the conventional reservoir area (YAE-2, 8, 9). The results suggest the recharged water filled in cracks changes the crack shape and shows the potential of S-wave splitting as a monitoring tool for geothermal areas. Our results would also help with research to utilize the underground structure for CO2 storage and geological disposal of radioactive areas.
The performance of Yanaizu-Nishiyama Geothermal Power Plant in Fukushima, Japan, decreased from 65,000 kW in 1995 to 30,000 kW by 2017. To mitigate the reduction in steam production, a gravity-driven water injection test was started in 2015 and continues to date. We focus on 267 micro-seismic events that occurred after November 2017. There are nine three-component velocity seismometers with a sampling frequency of 1000 Hz placed around the power plant: five stations are on the surface (YAE-1 to YAE-5) and four in the boreholes (YAE-6 to YAE-9). We select the waveforms that pass through geothermal reservoirs. The analysis window is 0.30 seconds, starting 0.05 seconds before the S-wave arrival. We calculate the cross-correlation between N-S and E-W components with a bandpass filter from 3 to 60 Hz. Rotating the coordinate axis in five-degree increments from 0 to 90 degrees, we calculate the rotation angle and cross-correlation value for each degree at 0.001-second intervals. For each event, we record the rotation angle, the time lag, and the cross-correlation value at the highest cross-correlation value.
We compare the results before (November 2017 to July 2018) and after (July 2019 to March 2020) the major water recharge. We classify stations into three groups: reservoir areas (YAE-2, 5, 8, 9), southern areas on the reinjection well area (YAE-1, 7), and eastern areas on the reinjection well (YAE-3, 4, 6). The faster S wave directions have changed at multiple stations in the reservoir area and the water injection area before and after the water injection. Significant faster S wave direction changes are observed at borehole stations YAE-8 and 9. In addition, the faster S wave directions are mainly divided into two directions: NW-SE and NE-SW. The direction of the faster S-wave is generally considered to be the long axis of the heterogeneity and the direction of the maximum principal stress. Thus, the faster S wave direction reflects the predominant heterogeneous structure, surrounding faults, and fracturing systems in the geothermal area. NW-SE direction closely aligns with the direction of the maximum principal stress axis from the Japanese crustal stress database (AIST, 2024). This direction is also consistent with the orientation of faults near the recharge well (Dian et al., 2024). NE-SW direction corresponds to the major fault orientation located northwest of the reinjection well. Due to recharging, the time lag between two S waves is confirmed to increase at the conventional reservoir area (YAE-2, 8, 9). The results suggest the recharged water filled in cracks changes the crack shape and shows the potential of S-wave splitting as a monitoring tool for geothermal areas. Our results would also help with research to utilize the underground structure for CO2 storage and geological disposal of radioactive areas.