9:30 AM - 9:45 AM
[SVC28-03] Geosolution-Based Optimization of Geothermal Energy Resources from field to development scale: Case study of Hydrothermal-Volcanic Systems on Mt. Kuju
Keywords:Ambient noise imaging , S-wave velocity , Seismic anisotropy, Hydrothermal-volcanic system, Machine learning, Temperature model
The Japanese Islands are a long volcanic arc along active subduction zone (~120 active volcanoes). This arc system offers abundant geothermal energy resources throughout Japan, including both volcanic and non-volcanic systems. Our research mainly used ambient noise data in various volcanic settings in Japan, Kenya, and New Zealand, contributing to the formulation of optimal geothermal exploration and development strategies.
We conducted a comprehensive study from field to development scale of Mt. Kuju to understand the Kuju geothermal-volcanic system and focused on this volcano in Japan as a case study.
First, we analyzed a large array seismometer network (~35km x 35km) to investigate the heat source (or deep supercritical geothermal reservoir) based on direct surface wave (ambient noise) tomography. Phase velocity through zero-crossing cross-correlation waveform analysis was employed. Our findings revealed a large supercritical reservoir or shallow heat source at a depth of > 3.5km from the surface that feeds the shallow geothermal system beneath the volcanic group of Kuju. Additionally, the results illustrated plumbing-like fractures connecting to the surface, where numerous local hot springs are widespread in this region.
Second, we deployed dense seismometers (~10km x 10km) above the heat source locations to examine reservoir-scale details. We identified potential geothermal structures (fractured zones, fluid flow pathways) based on seismic velocity, anisotropic structure, and temperature models of existing potential and new geothermal systems of Mt. Kuju, using the azimuthal anisotropic ambient noise imaging and machine learning models. Our results indicate low S-wave velocity structures (<2 km/s) related to faulted or fractured zones at 0–2 km depth below sea level, particularly near power stations and along Mts. Kuroiwa and Hossho, volcanic systems associated with dense microseismicity. Anisotropic fast polarization of 3–7 % within the low-velocity zone is oriented N30°W, parallel to a NW–SE-trending fault beneath the Hatchobaru and Otake geothermal power stations.
Furthermore, we enhanced the 3D temperature model estimation based on machine learning (e.g., Ensemble Boosted Trees), incorporating data from 8 temperature logs, 3D S-wave velocity, 3D resistivity, and 3D seismic anisotropy (Azimuth, Amplitude). We obtained a 3D temperature variation of up to 250 oC (R-Square ~0.9) just below sea level in our identified geothermal reservoirs. These results should be combined/overlaid with other geophysical findings, including geological and geochemical maps (e.g., down to a few meter scales for optimally sitting new geothermal injection and production areas associated with minimal seismicity accumulation).
In summary, our approach presented here can offer an exploration imaging model applicable to the other ~119 active volcanic systems and geosolutions for optimization of geothermal resources at different scales, as well as advancing/promoting geothermal energy development in Japan.
This study is supported by Japan Science and Technology Agency (JST: Grant Number JPMJSA1905) and Japan International Cooperation Agency through "Comprehensive Solution for Optimum Development of Geothermal Systems in East African Rift Valley" (LENGO Project) by the Science and Technology Research Partnership for Sustainable Development (SATREPS) program and the New Energy and Industrial Technology Development Organization (NEDO), Japan. This work is partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant Numbers JP20H01997, JP20K04133, JP21H05202, and JP22H05108), and Shikoku Electric Power Co., Inc.
We conducted a comprehensive study from field to development scale of Mt. Kuju to understand the Kuju geothermal-volcanic system and focused on this volcano in Japan as a case study.
First, we analyzed a large array seismometer network (~35km x 35km) to investigate the heat source (or deep supercritical geothermal reservoir) based on direct surface wave (ambient noise) tomography. Phase velocity through zero-crossing cross-correlation waveform analysis was employed. Our findings revealed a large supercritical reservoir or shallow heat source at a depth of > 3.5km from the surface that feeds the shallow geothermal system beneath the volcanic group of Kuju. Additionally, the results illustrated plumbing-like fractures connecting to the surface, where numerous local hot springs are widespread in this region.
Second, we deployed dense seismometers (~10km x 10km) above the heat source locations to examine reservoir-scale details. We identified potential geothermal structures (fractured zones, fluid flow pathways) based on seismic velocity, anisotropic structure, and temperature models of existing potential and new geothermal systems of Mt. Kuju, using the azimuthal anisotropic ambient noise imaging and machine learning models. Our results indicate low S-wave velocity structures (<2 km/s) related to faulted or fractured zones at 0–2 km depth below sea level, particularly near power stations and along Mts. Kuroiwa and Hossho, volcanic systems associated with dense microseismicity. Anisotropic fast polarization of 3–7 % within the low-velocity zone is oriented N30°W, parallel to a NW–SE-trending fault beneath the Hatchobaru and Otake geothermal power stations.
Furthermore, we enhanced the 3D temperature model estimation based on machine learning (e.g., Ensemble Boosted Trees), incorporating data from 8 temperature logs, 3D S-wave velocity, 3D resistivity, and 3D seismic anisotropy (Azimuth, Amplitude). We obtained a 3D temperature variation of up to 250 oC (R-Square ~0.9) just below sea level in our identified geothermal reservoirs. These results should be combined/overlaid with other geophysical findings, including geological and geochemical maps (e.g., down to a few meter scales for optimally sitting new geothermal injection and production areas associated with minimal seismicity accumulation).
In summary, our approach presented here can offer an exploration imaging model applicable to the other ~119 active volcanic systems and geosolutions for optimization of geothermal resources at different scales, as well as advancing/promoting geothermal energy development in Japan.
This study is supported by Japan Science and Technology Agency (JST: Grant Number JPMJSA1905) and Japan International Cooperation Agency through "Comprehensive Solution for Optimum Development of Geothermal Systems in East African Rift Valley" (LENGO Project) by the Science and Technology Research Partnership for Sustainable Development (SATREPS) program and the New Energy and Industrial Technology Development Organization (NEDO), Japan. This work is partially supported by Japan Society for the Promotion of Science (JSPS) KAKENHI (Grant Numbers JP20H01997, JP20K04133, JP21H05202, and JP22H05108), and Shikoku Electric Power Co., Inc.