Estimation of shallow structure along the Hinagu Fault by applying seismic interferometry to DAS observations conducted along National Route 3

doi:10.21203/rs.3.rs-4480554/v1

Download PDF

Research Article

Estimation of shallow structure along the Hinagu Fault by applying seismic interferometry to DAS observations conducted along National Route 3

https://doi.org/10.21203/rs.3.rs-4480554/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Distributed acoustic sensing (DAS) is a newly developed geophysical observation method and has attracted wide attention in seismology for realizing ultra-high-density observations. DAS uses fiber-optic cables and measures the strain at every point along the cable. This advantage renders DAS an effective tool for investigating near-surface geotechnical properties. Near fault zones, it is important to obtain detailed geotechnical information in advance because of the potential for significant damage in an earthquake. In this study, we recorded continuous ground motion for approximately one month using a 40 km-long fiber-optic communication cable running under National Route 3 in Kumamoto Prefecture. The cross-correlation function (CCF) was calculated using ambient noise, and three-station interferometry was applied to improve the signal-to-noise ratio of the CCF. Using the reconstructed CCF between channels, we calculated the dispersion curves by conducting multichannel surface wave analysis and estimated the one-dimensional velocity structure of each section from the fundamental modes of the dispersion curves. We obtained the detailed shallow S-wave velocity structure to a depth of 180 m along the Hinagu Fault for approximately 2.5 km. The obtained velocity structure showed that the low-velocity region increased abruptly with depth from the center to the latter half of the analyzed section. This velocity change occurs when the national highway running parallel to the fault gradually leaves the fault, suggesting a structural change from solid volcanic layers to thick shallow sedimentary layers derived from the Yatsushiro Plain.

Distributed acoustic sensing

Hinagu Fault

Shallow structure

Ambient noise

Three-station interferometry

Imaging shallow subsurface structures is important for urban development, including building design, use of underground spaces, and seismic simulation. In fault zone areas, ground rupture and the amplification of seismic motion can cause significant damage, and information on shallow structures is necessary to evaluate soil properties. In addition, the shallow subsurface S-wave velocity structure of a fault zone can be significantly heterogeneous (Chimoto et al., 2015). Taylor et al. (2019) used ambient noise to investigate the near-surface structure of the North Anatolian Fault Zone. The authors observed a strong seismic velocity gradient across the fault along the distributed acoustic sensing (DAS) line. Such velocity variations are also observed in the Xiaojiang fault zone (Liang et al., 2023).

On April 14, 2016, an earthquake of magnitude 6.5 occurred in Kumamoto Prefecture, Japan, at a depth of 11 km, according to Japan Meteorological Agency (JMA). Twenty-eight hours later, in the early morning of April 16, 2016, a large earthquake occurred at a depth of 12 km with a JMA magnitude of 7.3. Both earthquakes measured a maximum intensity of seven in the Kumamoto Prefecture and caused extensive damage to residents, housing, and infrastructure. The two earthquakes were caused by different faults (Fig. 1): the April 14 foreshock occurred at the northern end of the Hinagu Fault, which extends southwest from Mashiki City to the southern Yatsushiro Sea, and the April 16 mainshock occurred on the Futagawa Fault, which extends westward from the Aso region through Mashiki City to the Uto Peninsula. The foreshock of the Kumamoto earthquake ruptured mainly the northern part of the Hinagu fault. Laboratory experiments (Scholz, 2015) show that stress accumulation decreases the slope of the straight line in the Gutenberg-Richter relation (Gutenberg and Richter 1944). Nanjo et al. (2019) found that the southern Hinagu and Yatsushiro Sea sections, which showed lower b-values, were not ruptured by the two large Kumamoto earthquakes. Therefore, these areas have the potential for large earthquakes in the future. In particular, the Hinagu and Yatsushiro Sea sections are among the most active fault zones in Japan, with a 6% and 16% probability of earthquakes within the next 30 years, respectively (The Headquarters for Earthquake Research Promotion, 2024).

The Japan Seismic Hazard Information Stations (J-SHIS) were established in 2005 by the National Research Institute for Earth Science and Disaster Resilience (NIED) and has been operational since then (Fujiwara et al., 2006). The J-SHIS provides average S-wave velocities of the upper 30 m (AVS30) and the ground amplification factor is calculated using the microtopography classification data with a resolution of 250 m mesh. To determine the detailed structure around a fault, it is necessary to obtain actual observations and a more detailed velocity model.

Spatiotemporal observations using seismic networks have been common since the 2000s. Areal observations using a large number of spatially dense seismometers, such as the large-N array, have been developed later (e.g., Schmandt et al., 2013; Matsumoto et al., 2020). Microtremor surveys, which use arrays with several seismometers, have been conducted to estimate shallow structures. However, this method is labor-intensive and difficult to perform in urban areas. Recently, DAS, which uses fiber optic cables as sensors to measure the strain or strain rate in time series, has attracted attention as an observational method in geosciences (see Zhan, 2019 for a review). The DAS technique involves installing a measurement device, called an interrogator, at the end of a fiber-optic cable and injecting optical pulses into the cable from the device. The optical pulse is then scattered by the impurities in the cable (Rayleigh backscattering), and the backscattered wave returns to the interrogator. When the cable is subjected to vibrations, the phase of the scattered wave changes as the cable expands or contracts. The strain in this section is determined by measuring this change between two points along the cable. DAS has the advantage of using existing cables. This makes it easy to connect the interrogator to a fiber-optic cable and set the parameters, allowing observations to be conducted in urban areas and densely populated areas along national highways where observations using conventional seismometers are difficult. In addition, DAS enables ultra-high-density multi-point observations because the observation points are spaced several meters apart along several tens of kilometers of the cable. Owing to these advantages, DAS has been used to estimate inland fault locations (Atterholt et al., 2022), epicenters (Lentaset al., 2023), and subsurface structures using various microtremors (e.g., Song et al., 2021; Shao et al., 2022; Jiang et al., 2023; Cheng et al., 2023). Additionally, it has been used for fault zone imaging (Yang et al., 2022) and submarine cable studies (e.g., Lior et al., 2022; Shinohara et al., 2022). In volcanic regions, DAS has been used for the source studies of volcanic earthquakes (Klaasen et al., 2021), site characterization (Nishimura et al., 2021), and tomography studies of the volcanic basement (Biondi et al., 2023).

Several studies have applied seismic interferometry to DAS data to obtain velocity structures. Song et al. (2021) applied seismic interferometry to DAS observations conducted using fiber-optic cables running under two roads in China. The results showed that one road had a shallow velocity structure, whereas, for the other road, which had heavy vehicle traffic, the noise sources were distributed in a non-uniform and non-isotropic manner, making it impossible to extract surface waves from seismic interferometry. Song et al., (2022) applied three-station interferometry (TSI) in the above section to solve this problem. The TSI method could extract surface waves, and the shallow velocity structure at this road was clarified, although the noise source distribution was complex. Stehly et al. (2008) reconstructed the surface wave signal between two stations by calculating the cross-correlation between the coda waves of the noise cross-correlation function (NCF) obtained from the two stations and a third station (C3: Correlation of Coda of Correlation). By analyzing a continuous record of 150 stations, Froment et al. (2011) demonstrated that C3 can sufficiently suppress the effects caused by the distribution of anisotropic noise sources. TSI for direct waves was proposed by Curtis et al. (2010), which uses the entire NCF instead of the coda waves of the NCF, and the cross-correlation and convolution are calculated based on the locations of the virtual sources and receiver points, as described above. Qiu et al. (2021) improved the signal-to-noise ratio of surface waves extracted from the NCFs of a linear seismometer array using the TSI method.

In this study, DAS observations were conducted using fiber-optic cables laid underground along National Route 3 in Kumamoto. Seismic interferometry and the TSI methods were then applied to the observed data. Finally, surface wave and inversion analyses were performed to determine the shallow S-wave velocity structure of the national route along the Hinagu Fault.

2.1. Location and Data

DAS observations were conducted along Route 3 in Kumamoto Prefecture for approximately one month, from February 15 to March 11, 2023. Figure 1 shows the section covered by the observation. We used a fiber optic cable owned by the Ministry of Land, Infrastructure, Transport, and Tourism (MLIT), which was installed underground, and conducted DAS observations on the section from the MLIT Kumamoto Maintenance Branch Office to the Yatsushiro Maintenance Branch Office, located approximately 40 km to the south. As shown in Fig. 1, a part of National Route 3 runs along the Hinagu Fault.

The interrogator was an iDAS (manufactured by Silixa), which was installed at the Kumamoto Maintenance Subbranch of the MLIT. The parameters were set to 4 m channel spacing, 10 m gauge length, and 400 Hz sampling frequency, resulting in a total of 9,984 observation points.

2.2. Seismic events during DAS observation

The DAS observations recorded the vibrations caused by a variety of factors. Figure 2 shows an example of a seismic event. The traffic noises appeared as near-vertical lines in the figure are generated by cars passing on national roads. The noise increases after channel 8000 because the intensity of the optical pulses decreases with the propagation distance, and the backscattered light becomes weaker. Even under these circumstances, we observed the seismic event shown in Fig. 2. We confirmed using the JMA unified earthquake catalog that this event was a JMA magnitude 1.6 earthquake at a depth of 10 km. This indicates that the DAS observations can detect even very small earthquakes. Similarly, a comparison of the DAS records with the JMA unified earthquake catalog confirmed 22 seismic events during the observation period.

2.3. Power Spectral Density (PSD)

The obtained DAS recordings were spectrally analyzed to determine the PSD of each channel for the frequency bands of 1–5 Hz, 5–15 Hz, and 15–30 Hz (Fig. 3). During several periods, the DAS interrogator was discontinued. The frequencies of 5–15 Hz were dominant, which is similar to the results of previous studies, such as Lindsey et al. (2020) and Shao et al. (2022), which used traffic noise. Figure 3 shows that the overall PSD values are higher during the daytime than during the late-night hours, and the PSD varies significantly between channels 4000 and 8000. This may be due to the influence of traffic volume, which is higher during daytime when there is more human activity. In addition, after channel 4000, the bypass changes from a two-lane road to a one-lane road, which is likely to cause variations in traffic volume. Interestingly, the Sunday PSD in the time series is generally low in all frequency bands. This is because of the usually low Sunday traffic. There were some channels where the PSD was always high and the oscillations constantly continued. This is attributed to bridges and intersections. In the case of bridges, not only vehicles but wind and other factors also can contribute to constant shaking. At intersections, vehicles can drive over the fiber while the other side stops, causing constant shaking.

The DAS records are rich in ambient noise caused by vehicles and human activities. Therefore, we performed a cross-correlation analysis using these records to retrieve surface waves and estimate the shallow structure of the Hinagu Fault. The flow of the analysis is illustrated in Fig. 4. For preprocessing, we followed the seismic interferometry method (Bensen et al., 2007). For each day, we used 3 h of DAS recordings (13:00–15:00) in the channel section along the Hinagu Fault. First, the data were detrended, bandpass filtered at 1–30 Hz, and resampled to 100 Hz. Next, the data were divided into 10-minute time windows, waveforms were converted to 1-bit, spectral whitening was performed, cross-correlation functions (CCFs) were computed, and phase-weighted stacking (PWS) of the CCFs was performed (Schimmel and Paulssen 1997).

3.1. Three-Station Interferometry (TSI)

Heavy vehicle crossings and strong seismic waves from oblique directions resulted in asymmetric CCFs. In addition, the signal-to-noise ratio was low, and clear CCFs could not be obtained for channels far from the virtual source. Therefore, TSI was used in addition to conventional seismic interferometry to enhance the surface wave signal propagating between two channels.

Unlike the usual seismic interferometry method, which uses seismic ambient noise as the input, the TSI uses three points: a virtual seismic source and two receiver points. The TSI for the DAS configuration was proposed by Song et al. (2022). In TSI, the CCF is calculated between channels $i$ and $j$ and the common channel $k$ as the virtual source in the frequency domain as follows:

$${G}_{ij}^{k,t+1}=\left\{\begin{array}{c}\overline{{G}_{ik}^{t}\left(\omega \right)} \bullet {G}_{jk}^{t}\left(\omega \right) (k<i)\\ {G}_{ik}^{t}\left(\omega \right) \bullet {G}_{jk}^{t}\left(\omega \right) (i<k<j)\\ {G}_{ik}^{t}\left(\omega \right) \bullet \overline{{G}_{jk}^{t}\left(\omega \right)} (j<k)\end{array}\right.. \left(1\right)$$

where ${{G}^{t}}_{ik}$ is the CCF between channels $i$ and $k$ in the $t$-th iteration of the TSI; the bars above the letters represent complex conjugates. The CCF for a single-channel pair can be reconstructed by stacking the CCFs for all possible virtual sources.

$${G}_{ij}^{t+1}= \sum _{k=1}^{N}{G}_{ij}^{k,t+1}\left(\omega \right). \left(2\right)$$

The TSI method can be repeatedly applied to the CCF reconstructed using the TSI method, and ${G}_{ij}^{0}$ is the original CCF. This iteration enhances the wave signal propagating between two points compared to the original CCF and is expected to improve the signal-to-noise ratio (Figure S1).

4.1. Reconstructed CCF using TSI method

We calculated the CCFs using conventional seismic interferometry for every 400 channel sections along the cable. However, the signal-to-noise ratio (SNR) of CCF was poor in almost all sections. The two sections shown in Fig. 5, channels 6000 − 6400 and channels 6470 − 6850, were close to the Hinagu Fault and showed relatively high SNR; therefore, we focused our analysis on these sections. A total of 286 time windows per channel per observation period were used, excluding the DAS outage period.

Next, TSI was performed on the CCFs stack using the PWS. We used only the positive lag time portion of the CCF for the TSI iterations. Figure 6 shows a comparison of shot gathers between the original CCF with a virtual source at channel 6000 and the CCF reconstructed by TSI. The lag time was set between − 20 and 20 s to calculate the cross-correlation. The CCFs reconstructed using the TSI method exhibited a higher SNR than the original CCFs. Propagating surface waves at distant channels appeared after applying the TSI method. More TSI iterations enhanced the SNR. The retrieved surface waves included both Love and Rayleigh waves. According to Nakahara et al. (2021) and Fukushima et al. (2022), the Rayleigh wave obtained by the correlation in a channel pair decays in the order of ${d}^{-\frac{1}{2}}$, where $d$ is the distance between two points, whereas Love waves decay in the order of ${d}^{-\frac{3}{2}}$. Therefore, we assume that the waves retrieved from the CCFs are Rayleigh waves, because of a relatively low contribution of Love waves.

4.2. Dispersion Curve

The CCFs reconstructed using the TSI were used to obtain the dispersion spectra by conducting the multichannel analysis of surface waves (MASW), as proposed by Park et al. (1999). MASW can be expressed as follows:

$$E\left(f,c\right)= \sum _{i=1}^{N}\frac{{G}_{ij}({x}_{i}-{x}_{j},f)}{\left|{G}_{ij}({x}_{i}-{x}_{j},f)\right|}{e}^{\frac{i2\pi f{(x}_{i}-{x}_{j})}{c}} \left(3\right)$$

where $E\left(f,c\right)$ is the frequency and phase-velocity domain dispersion spectrum, and ${G}_{ij}({x}_{i},f)$ is the i-th channel record in the frequency domain for the virtual shot at the j-th channel. Figure 7 shows the dispersion curves obtained from the original CCF and reconstructed CCFs by TSI for the virtual shot at channel 6000 as in Fig. 6. For MASW, we used data with a positive lag time of the CCF and up to 79 channels (316 m away from the virtual shot). A comparison of the results in Fig. 7 showed that the dispersion curves were clear when TSI was performed. The dispersion curves were also extended to the high-frequency side after TSI iterations. Overall, the dispersion curves for the fundamental mode showed phase velocities of 400–700 m/s at a frequency range of 3–8 Hz. The higher-order mode was more visible than the fundamental mode at frequencies between 10 and 15 Hz. Similarly, as shown in Fig. 7, the dispersion curve stabilized after approximately three TSI iterations (Figure S2). Therefore, we proceeded with the inversion analysis using only the dispersion curves of the fundamental mode with three TSI iterations.

4.3. 1D inversion

An initial S-wave velocity model was constructed prior to inversion. The dispersion spectra were calculated for each of the 10 channels of the virtual source, and their peak values were obtained. The peaks corresponding to the fundamental mode were selected manually. From the generated frequency-phase-velocity profile, the S-wave velocity and the corresponding depth were calculated using the phase velocity information and frequency. Here, the empirical equation by Xia et al., (1999) was used to calculate the S-wave velocity and depth. the empirical equation for the S-wave velocity ${V}_{{S}_{i}}$ and depth ${d}_{i}$ is expressed as follows:

$${V}_{{S}_{i}}= \frac{{c}_{r\left({f}_{i}\right)}}{0.88} \left(i=1\cdots m\right) , \left(4\right)$$

$${d}_{i}=\frac{{V}_{{S}_{i}}}{{f}_{i}}\times 0.63 \left(5\right)$$

where ${c}_{r\left({f}_{i}\right)}$ is the phase velocity corresponding to the frequency ${f}_{i}$ of the dispersion curve, and $m$ is the pick number. Using the above equations, the S-wave velocity and the corresponding depth were determined; a range of depths was specified every 30 m, and the average S-wave velocity within that range was calculated. These calculations were conducted for all 33 profiles to create an initial model of the 2D S-wave velocity structure (Figure S3).

Next, a one-dimensional inversion analysis was performed on the initial model. Rix (1991) showed that a reliable depth is approximately half the length of the maximum-minimum wavelength. The minimum wavelength was 31 m, and it was sensitive to depths as shallow as 15 m. The maximum wavelength was 361 m and was sensitive to a depth of 180 m. Therefore, we narrowed the depth range to be estimated in this area. The model consists of six layers: five 30 m layers and a half-space, with unknown S-wave velocities in each layer. The inversion was performed using the initial model for each section. The blank cells in the initial model, where S-wave velocities were not obtained, were filled with the average S-wave velocity in all sections of the layers at that depth. We used the inversion program of the Computer Program in Seismology (CPS: Herrmann 2013), and the final velocity structure was obtained after 10 iterations. The forward-calculated dispersion curves from the obtained velocity structure model are shown in Fig. 7, as an example. The final inversion results for all channels and the corresponding aerial photographs are shown in Figs. 8 and 9. As shown in Fig. 7, the dispersion curves show good agreement with the observations. Figure 8 shows that the sedimentary layers are uniformly distributed in the shallow part up to a depth of approximately 30 m. However, in the deeper part, a large velocity change was observed with depth on the northern side (the first half of the channel). Furthermore, on the southern side, a low-velocity zone, which appeared to be a sedimentary layer, existed to a depth of approximately 120 m. At the section between channels 6130–6150, National Route 3, which runs almost parallel to the fault, gradually moves away from the fault.

In this study, we determined the S-wave velocity structure to a depth of 180 m along the Hinagu Fault over a distance of approximately 2.5 km. Figure 8 shows the inversion results for the channels 6000–6320 section. In this section, mountains are located on the eastern side (upper part) of the Hinagu Fault, and rural areas are located on the western side (lower part). This area is called the “Yatsushiro Plain” and is formed by the north–south elongated fan-shaped delta and reclaimed land created by rivers such as the Kuma River, which flows from the Kyushu Mountains in the east to the Yatsushiro Sea in the west. More than two-thirds of this plain is reclaimed land created during the Edo–early Showa period. In addition, the mountainous areas in this section contain volcanic strata such as tuff and limestone deposited by the eruptive activity of Mount Aso. In other words, the volcanic strata along the national route in the northern part of the section bounded by channel 6150 in Fig. 8 were formed in relatively shallow areas, forming stable soil. On the other hand, along the national route south of the channel 6150, as the route moves away from the fault, the shallow structure changes from strong volcanic strata to thick sedimentary layers, such as fans and deltas in the Yatsushiro Plain.

For the section from channels 6470 to 6770 (Fig. 9), a large S-wave velocity change was observed around channel 6650. Around channel 6650, the section approaches the eastern mountain face again, and the same tendency as at channel 6150, as shown in Fig. 8, was observed. In addition, the low-velocity zone extends to a depth of 130 m in the section before this point, suggesting the presence of a thick sedimentary layer. In the section from channel 6150 to 6400 (Fig. 8), a thick sedimentary layer was also observed to a depth of 120 m. Between these two sections, the Hikawa River is located, which is approximately 80 m wide. Therefore, the thick sedimentary layers observed in these two sections were most likely created by the river flowing between them, as shown in Fig. 10. In the section after channel 6650, the S-wave velocity was relatively low, approximately 300 m/s near the ground surface.

To estimate the detailed shallow structure near the Hinagu Fault, we used a 40-km-long fiber-optic cable laid under National Route 3 in Kumamoto Prefecture and conducted DAS observations. Part of the cable runs along the southern part of the Hinagu Fault, which has been identified as a possible source of future large earthquakes. The ultra-high-density strain waveforms at 4-m intervals along the cable with a total of 9,984 channels were recorded for one month. National Route 3 has heavy traffic, and most of the data were obscured by traffic-induced vibrations. Seismic interferometry was applied to ambient noise recordings, including traffic noise, to obtain a cross-correlation function; however, this method was unable to obtain a stable cross-correlation function owing to the non-uniform and non-stationary noise source distribution along the national highway. Therefore, we applied the TSI method and reconstructed a cross-correlation function with an improved signal-to-noise ratio for the surface waves extracted for a 2.5 km section along the Hinagu Fault. Rayleigh wave dispersion curves were extracted from the reconstructed cross-correlation functions. One-dimensional inversion analysis using the fundamental mode of the obtained dispersion curves revealed a detailed shallow S-wave velocity structure along the Hinagu Fault to a depth of 180 m over a distance of approximately 2.5 km. The obtained velocity structure indicates that soft sedimentary layers are generally widespread near the ground surface; however, the low-velocity region increased rapidly with depth from the center to the south of the analyzed section. This change in velocity reflects the difference in geological structure caused by the gradual separation of the national route, which runs parallel to the fault. These subsurface heterogeneities suggest that local amplification of seismic waves occurs in this area.

This study showed that the combination of DAS observations and the TSI method can be used to estimate detailed subsurface velocity structure even in an environment, with non-uniform and non-stationary noise sources distributed, such as a national highway. This indicates that the proposed method can be applied to urban areas and other national routes. In addition to geoscientific knowledge, this method is expected to provide useful information for disaster prevention measures in densely populated areas.

DAS, distributed acoustic sensing; MLIT, Ministry of Land, Infrastructure, Transport, and Tourism; PSD, power spectral density; CCF, cross-correlation function; PWS, phase-weighted stack; TSI, three-station interferometry.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Availability of data and materials

Records of the DAS observation are available upon reasonable request. For the inversion of the velocity structure, we used Computer Program in Seismology (Herrmann 2013).

Competing interests

The authors declare that they have no competing interests.

Funding

This work was partly supported by JSPS KAKENHI Grant Numbers JP20K04097 and JP23K03553.

Authors’ contributions

SH and KE designed the observations, contributed to data acquisition, and analyzed the DAS data. SH drafted the manuscript.

Acknowledgements

The DAS observations were supported by Kumamoto River and National Highway Office, Kyushu Regional Development Bureau, Ministry of Land, Infrastructure, Transport and Tourism (MLIT).

Atterholt, J., Zhan, Z., & Yang, Y. (2022). Fault Zone Imaging With Distributed Acoustic Sensing: Body-To-Surface Wave Scattering. Journal of Geophysical Research: Solid Earth, 127(11), e2022JB025052. https://doi.org/10.1029/2022JB025052
Bensen, G. D., Ritzwoller, M. H., Barmin, M. P., Levshin, A. L., Lin, F., Moschetti, M. P., Shapiro, N. M., & Yang, Y. (2007). Processing seismic ambient noise data to obtain reliable broad-band surface wave dispersion measurements. Geophysical journal international, 169(3), 1239-1260.https://doi.org/10.1111/j.1365-246X.2007.03374.x
Biondi, E., Zhu, W., Li, J., Williams, E. F., & Zhan, Z. (2023). An upper-crust lid over the Long Valley magma chamber. Science Advances, 9(42), eadi9878.https://doi.org/10.1126/sciadv.adi9878
Cheng, F., Ajo-Franklin, J. B., Nayak, A., Tribaldos, V. R., Mellors, R., Dobson, P., & Imperial Valley Dark Fiber Team. (2023). Using Dark Fiber and Distributed Acoustic Sensing to Characterize a Geothermal System in the Imperial Valley, Southern California. Journal of Geophysical Research: Solid Earth, 128(3), e2022JB025240.https://doi.org/10.1029/2022JB025240
Chimoto, K., Saguchi, K., Yamanaka, H. (2015) Estimation of shallow S-wave velocity structures using microtremor array exploration in the Tachikawa-fault zone: Journal of Japan Association for Earthquake Engineering, 2015, 15(1), 1_81-1_100 (in Japanese) https://doi.org/10.5610/jaee.15.1_81
Curtis, A., & Halliday, D. (2010). Source-receiver wave field interferometry. Physical Review E, 81(4), 046601. https://doi.org/10.1103/PhysRevE.81.046601
Froment, B., Campillo, M., & Roux, P. (2011). Reconstructing the Green's function through iteration of correlations. Comptes Rendus Geoscience, 343(8-9), 623-632. https://doi.org/10.1016/j.crte.2011.03.001
Fujiwara, H., Kawai, S., Aoi, S., Ishii, T., Okumura, T., Hayakawa, Y., Morikawa, N., Senna, S., Kobayashi, K., & Hao, K. X. S. (2006, October). Japan seismic hazard information station, J-SHIS. In Proc 4th Int Conf Earthq Eng (Vol. 274). https://www.academia.edu/download/46408294/4icee-0274.pdf Accessed 9 March 2024.
Fukushima, S., Shinohara, M., Nishida, K., Takeo, A., Yamada, T., & Yomogida, K. (2022). Detailed S-wave velocity structure of sediment and crust off Sanriku, Japan by a new analysis method for distributed acoustic sensing data using a seafloor cable and seismic interferometry. Earth, Planets and Space, 74(1), 1-11. https://doi.org/10.1186/s40623-022-01652-z
Gutenberg, B., & Richter, C. F. (1944) Frequency of earthquakes in California. Bulletin of the Seismological society of America, 34(4), 185-188.
Herrmann, R. B. (2013). Computer programs in seismology: An evolving tool for instruction and research. Seismological Research Letters, 84(6), 1081-1088. https://doi.org/10.1785/0220110096
Jiang, W., Lin, J., Liu, B., Zhang, R., Zhang, B., Yang, Z., & Gu, Y. (2023). Distributed acoustic sensing for shallow structure imaging using mechanical noise: A case study in Guangzhou, China. Journal of Applied Geophysics, 215, 105139. https://doi.org/10.1016/j.jappgeo.2023.105139
Klaasen, S., Paitz, P., Lindner, N., Dettmer, J., & Fichtner, A. (2021). Distributed acoustic sensing in volcano-glacial environments—Mount Meager, British Columbia. Journal of Geophysical Research: Solid Earth, 126(11), e2021JB022358. https://doi.org/10.1029/2021JB022358
Lentas, K., Bowden, D., Melis, N. S., Fichtner, A., Koroni, M., Smolinski, K., Bogris, A., Nikas, T., Simos, C., & Simos, I. (2023). Earthquake location based on Distributed Acoustic Sensing (DAS) as a seismic array. Physics of the Earth and Planetary Interiors, 344, 107109. https://doi.org/10.1016/j.pepi.2023.107109
Liang, Y., Xie, T., Lu, Z., Li, Q., Zhang, X., Wang, G., Wang, G., Chen, S., & Chang, C. (2023) Shear wave velocity structure in the middle segment of the Xiaojiang fault zone using ambient noise tomography. Front. Earth Sci. 11:1161989. https://doi.org/10.3389/feart.2023.1161989
Lindsey, N. J., Yuan, S., Lellouch, A., Gualtieri, L., Lecocq, T., & Biondi, B. (2020). City-scale dark fiber DAS measurements of infrastructure use during the COVID-19 pandemic. Geophy Res Lett, 47(16). https://doi.org/10.1029/2020GL089931
Lior, I., Mercerat, E. D., Rivet, D., Sladen, A., & Ampuero, J. P. (2022). Imaging an underwater basin and its resonance modes using optical fiber distributed acoustic sensing. Seismological Society of America, 93(3), 1573-1584. https://doi.org/10.1785/0220210349
Matsumoto, S., Iio, Y., Sakai, S., & Kato, A. (2020). Inelastic strain in the hypocentral region of the 2000 Western Tottori earthquake (M 7.3) inferred from aftershock seismic moment tensors. Earth, Planets and Space, 72(1), 1-11. https://doi.org/10.1186/s40623-020-01186-2
Nakahara, H., Emoto, K., & Nishimura, T. (2021). Extending the formulation of the spatial autocorrelation (SPAC) method to strain, rotation and tilt. Geophysical Journal International, 227(1), 287-302. https://doi.org/10.1093/gji/ggab217
Nanjo, K. Z., Izutsu, J., Orihara, Y., Kamogawa, M., & Nagao, T. (2019) Changes in seismicity pattern due to the 2016 Kumamoto earthquakes identify a highly stressed area on the Hinagu fault zone. Geophysical Research Letters, 46(16), 9489-9496. https://doi.org/10.1029/2019GL083463
Nishimura, T., Emoto, K., Nakahara, H., Miura, S., Yamamoto, M., Sugimura, S., Ishikawa, A., & Kimura, T. (2021). Source location of volcanic earthquakes and subsurface characterization using fiber-optic cable and distributed acoustic sensing system. Scientific reports, 11(1), 6319. https://doi.org/10.1038/s41598-021-85621-8
Park, C. B., Miller, R. D., & Xia, J. (1999). Multichannel analysis of surface waves. Geophysics, 64(3), 800-808. https://doi.org/10.1190/1.1444590
Qiu, H., Niu, F., & Qin, L. (2021). Denoising Surface Waves Extracted From Ambient Noise Recorded by 1-D Linear Array Using Three-Station Interferometry of Direct Waves. Journal of Geophysical Research: Solid Earth, 126(8), e2021JB021712. https://doi.org/10.1029/2021JB021712
Rix, G. J., & Leipski, E. A. (1991). Accuracy and resolution of surface wave inversion. Recent advances in instrumentation, data acquisition and testing in soil dynamics: Proceedings of Sessions Sponsored by the Geotechnical Engineering Division of the American Society of Civil Engineers Inc., Publication of American Society of Civil Engineers.
Schmandt, B., & Clayton, R. W. (2013). Analysis of teleseismic P waves with a 5200-station array in Long Beach, California: Evidence for an abrupt boundary to Inner Borderland rifting. Journal of Geophysical Research: Solid Earth, 118(10), 5320-5338. https://doi.org/10.1002/jgrb.50370
Schimmel, M., & Paulssen, H. (1997). Noise reduction and detection of weak, coherent signals through phase-weighted stacks. Geophysical Journal International, 130(2), 497-505. https://doi.org/10.1111/j.1365-246X.1997.tb05664.x
Scholz, C. H. (2015) On the stress dependence of the earthquake b value. Geophys. Res. Lett., 42: 1399–1402. https://doi.org/10.1002/2014GL062863
Shao, J., Wang, Y., & Chen, L. (2022). Near-surface characterization using high-speed train seismic data recorded by a distributed acoustic sensing array. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-11. https://doi.org/10.1109/TGRS.2022.3153831
Shao, J., Wang, Y., Zheng, Y., Yao, Y., Wu, S., Yang, Z., & Xue, Q. (2022). Near-surface characterization using urban traffic noise recorded by fiber-optic distributed acoustic sensing. Frontiers in Earth Science, 10, 943424. https://doi.org/10.3389/feart.2022.943424
Shinohara, M., Yamada, T., Akuhara, T., Mochizuki, K., & Sakai, S. I. (2022). Performance of seismic observation by distributed acoustic sensing technology using a seafloor cable off Sanriku, Japan. Frontiers in Marine Science, 9, 844506. https://doi.org/10.3389/fmars.2022.844506
Song, Z., Zeng, X., Chi, B., Bao, F., & Osotuyi, A. G. (2022). Using the three-station interferometry method to improve urban DAS ambient noise tomography. Frontiers in Earth Science, 10, 952410. https://doi.org/10.3389/feart.2022.952410
Song, Z., Zeng, X., Xie, J., Bao, F., & Zhang, G. (2021). Sensing shallow structure and traffic noise with fiber-optic internet cables in an urban area. Surveys in Geophysics, 42, 1401-1423. https://doi.org/10.1007/s10712-021-09678-w
Stehly, L., Campillo, M., Froment, B., & Weaver, R. L. (2008). Reconstructing Green's function by correlation of the coda of the correlation (C3) of ambient seismic noise. Journal of Geophysical Research: Solid Earth, 113(B11). https://doi.org/10.1029/2008JB005693
Taylor, G., Rost, S., Houseman, G. A., & Hillers, G. (2019) Near-surface structure of the North Anatolian Fault zone from Rayleigh and Love wave tomography using ambient seismic noise. Solid Earth, 10(2), 363-378. https://doi.org/10.5194/se-10-363-2019
The Headquarters for Earthquake Research Promotion (2024) List of long-term evaluation results of major active faults. https://www.jishin.go.jp/main/choukihyoka/ichiran_pref.pdf Accessed 9 March 2024. (in Japanese)
Xia, J., Miller, R. D., & Park, C. B. (1999). Estimation of near-surface shear-wave velocity by inversion of Rayleigh waves. Geophysics, 64(3), 691-700. https://doi.org/10.1190/1.1444578
Yang, Y., Atterholt, J. W., Shen, Z., Muir, J. B., Williams, E. F., & Zhan, Z. (2022). Sub-kilometer correlation between near‐surface structure and ground motion measured with distributed acoustic sensing. Geophysical Research Letters, 49(1), e2021GL096503. https://doi.org/10.1029/2021GL096503
Zhan, Z. (2019). Distributed Acoustic Sensing Turns Fiber‐Optic Cables into Sensitive Seismic Antennas. Seismological Research Letters 2019, 91 (1), 1–15.

Download PDF

Editorial decision: Minor Revision
05 Aug, 2024
Reviewers agreed at journal
07 Jul, 2024
Editor assigned by journal
01 Jun, 2024
First submitted to journal
26 May, 2024

You are reading this latest preprint version

Estimation of shallow structure along the Hinagu Fault by applying seismic interferometry to DAS observations conducted along National Route 3

Status:

Version 1

Abstract

Figures

1. Introduction

2. DAS observation in Kumamoto

2.1. Location and Data

2.2. Seismic events during DAS observation

2.3. Power Spectral Density (PSD)

3. Method

3.1. Three-Station Interferometry (TSI)

4. Results

4.1. Reconstructed CCF using TSI method

4.2. Dispersion Curve

4.3. 1D inversion

5. Discussion

6. Conclusion

Abbreviations

Declarations

References

Supplementary Files

Status:

Version 1