5:15 PM - 6:45 PM
[SVC28-P04] Estimation of subsurface structure using autocorrelation function of teleseismic earthquakes in the Kuju geothermal area
Keywords:geothermal, seismic interferometry, autocorrelation, Kuju
A supercritical geothermal system, which is considered to have several times greater production capacity than conventional geothermal power generation, may exist in the Kuju geothermal area, and various surveys have been conducted. Reflection seismic surveys are one of the geophysical methods with the highest resolution, but even with expensive and large-scale equipment, the complex subsurface structure of geothermal areas often makes it impossible to obtain a clear formation boundary. This study aims to apply passive and low-cost seismic data analysis to extract reflection response based on seismic interferometry to natural earthquakes observed in geothermal areas to clarify the deep subsurface structure in the Kuju area.
In this study, we used teleseismic data with predominant low-frequency components to estimate the structure of the deep subsurface. Low-frequency data were extracted by applying a bandpass filter in the frequency range of 0.05 Hz to 1.2 Hz to seismic data. By applying an autocorrelation process based on seismic interferometry, we obtained an autocorrelation function for each seismometer location, including reflection signals beneath the seismometers. In the estimated autocorrelation function, the clear reflection signals were obtained in the two-way travel time of 2 sec to 4 sec, and the arrival time of each seismometer was picked considering the consistency among the nearby seismometers. The distance from the seismometer to the depth of the reflection surface was obtained by assuming the P-wave velocity down to the reflection surface to be 3.3 km/sec. The depth of the reflection surface below sea level was obtained by subtracting the distance from the surface to the estimated depth of the reflection surface from the elevation of each seismometer. In the southern part of the study area, the reflection surface is relatively shallow, ranging from ~1.5 km to 3 km in depth, and the depth increases toward the north, reaching ~5 km in depth at the deepest part. Comparing our results with the results of the basement rock depths estimated by previous gravity surveys, the overall trend is consistent as the basement rock depths also show a relatively shallow depth on the southwest side and a gradual deepening toward the north. This basement depth can be considered to correspond to the depression caused by the Shishimuta caldera inferred from the gravity survey. Since the reflection surface depth determined by this study is generally consistent with the basement depth estimated by the gravity survey, the reflection waves extracted by autocorrelation of teleseismic earthquakes are considered to be from the upper surface of the basement.
The results of this study suggest that autocorrelation analysis of teleseismic earthquakes based on seismic interferometry is effective for low-cost estimation of a wide range of basement structures in geothermal regions. Since long-term seismic observations are often conducted for microseismic monitoring in geothermal areas, this analysis can be applied to data for the purpose of microseismic monitoring to bring new added value.
Acknowledgement: We use seismic data acquired for the project “Potential survey and estimation of power generation of supercritical geothermal resources in Kuju area”, funded by New Energy and Industrial Technology Development Organization (NEDO).
In this study, we used teleseismic data with predominant low-frequency components to estimate the structure of the deep subsurface. Low-frequency data were extracted by applying a bandpass filter in the frequency range of 0.05 Hz to 1.2 Hz to seismic data. By applying an autocorrelation process based on seismic interferometry, we obtained an autocorrelation function for each seismometer location, including reflection signals beneath the seismometers. In the estimated autocorrelation function, the clear reflection signals were obtained in the two-way travel time of 2 sec to 4 sec, and the arrival time of each seismometer was picked considering the consistency among the nearby seismometers. The distance from the seismometer to the depth of the reflection surface was obtained by assuming the P-wave velocity down to the reflection surface to be 3.3 km/sec. The depth of the reflection surface below sea level was obtained by subtracting the distance from the surface to the estimated depth of the reflection surface from the elevation of each seismometer. In the southern part of the study area, the reflection surface is relatively shallow, ranging from ~1.5 km to 3 km in depth, and the depth increases toward the north, reaching ~5 km in depth at the deepest part. Comparing our results with the results of the basement rock depths estimated by previous gravity surveys, the overall trend is consistent as the basement rock depths also show a relatively shallow depth on the southwest side and a gradual deepening toward the north. This basement depth can be considered to correspond to the depression caused by the Shishimuta caldera inferred from the gravity survey. Since the reflection surface depth determined by this study is generally consistent with the basement depth estimated by the gravity survey, the reflection waves extracted by autocorrelation of teleseismic earthquakes are considered to be from the upper surface of the basement.
The results of this study suggest that autocorrelation analysis of teleseismic earthquakes based on seismic interferometry is effective for low-cost estimation of a wide range of basement structures in geothermal regions. Since long-term seismic observations are often conducted for microseismic monitoring in geothermal areas, this analysis can be applied to data for the purpose of microseismic monitoring to bring new added value.
Acknowledgement: We use seismic data acquired for the project “Potential survey and estimation of power generation of supercritical geothermal resources in Kuju area”, funded by New Energy and Industrial Technology Development Organization (NEDO).