Local Earthquake Response on the Submarine Communication Cable in The Northern South China Sea

Local Earthquake Response on the Submarine Communication Cable in The Northern South China Sea | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Local Earthquake Response on the Submarine Communication Cable in The Northern South China Sea Hao Li, Xuzhang Shen, Zhaohui Li, Shaoyi Chen This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4811181/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 15 Apr, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract As a new seismological monitoring technology, distributed acoustic sensing (DAS) not only provides a high-precision seismic observation method, but also uses undersea communication fiber or dark fiber to reduce the layout cost, and supports long-term observation for several years, and the sensing resolution reaches the scale of meters. This technique provides a broad prospect for seismic observation in oceanic area which is difficult to be covered by traditional seismic observation instruments. In this paper, we analyzed the DAS waveforms of 10 local-earthquake events waveforms recorded by offshore submarine communication cable in the northern South China Sea, and discussed the response ability of DAS system based on shallow sea communication cable to near-earthquake. It is found that the signals are affected by surface gravitational waves and micro seismic noise (Scholte Wave) generated by solid-liquid coupling motion. Then, waveform stacking, discrete wavelet transform, and empirical mode decomposition are used to denoise the signal in shallow sea area for extracting earthquake waveform. Submarine optical cable is the most widely distributed submarine signal monitoring system in the world. Long-term and cheap deep-sea seismic wave observation based on such the large-scale global ocean observation network can provide important data for the development of seismology. Earth and environmental sciences/Solid earth sciences/Seismology Physical sciences/Optics and photonics/Applied optics/Optical sensors Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Introduction Distributed acoustic sensing (DAS) is a newly developed technique that transforms telecommunication fiber optic cables into linear arrays of ground-motion sensors. Compared with the conventional seismometer, DAS not only make great cost saving, but also have an advantage in dense spatial sampling. In the last three decades, DAS has continually been using in gas industry as well as vertical seismic profiling. Meanwhile, the seismology was making progress towards the ocean study. There are still huge gaps in ocean study nevertheless on account of the aesthetic maintenance cost and recovery difficulty in OBS layout in deep ocean. However, it became effortless to capture the ground-motion signals when the DAS technique by using communication optical fiber cable under the sea came into sight. Above all, the extensive coverage of optical fiber around the Earth surface provides a highly operational observation platform for seismology researchers (Yu et al., 2019). Without seabed burying operation and observation duration limit, the DAS system have a long observation duration up to several months or even one year when the general geophones can just last for the boundary of couple of months. On the other hand, dense array of meter scale and more than 100km long-distance effective observation distance highlight its advantages with respect to traditional techniques. The application of DAS in passive earthquake seismology, however, is still at its early stage. Last few years, several ocean DAS experiments have been carried out with the high degree of correlation with a seismograph waveform. For example, Ethan et al (2019), analyzed the waveforms of 2018-08-19 Mw8.2 Fiji teleseismic deep earthquake recorded by the DAS array in the Southern Bight of the North shallow Sea offshore Zeebrugge, Belgium. They found that DAS waveforms show a high degree of correlation with a seismometer record (Ethan et al., 2019). In the MEUST-NUMerEnv project in Toulon, France, Sladen et al. (2019) also found coherent earthquake waveforms recorded at a dense DAS array in deep ocean showing the high sensitivity as a dense geophone array from a local M1.9 event (Sladen et al., 2019). Two experiments show the capacity of DAS system of detecting the earthquake and even the micro earthquake. However, some shortages as a newcomer were exposed in most of the shallow ocean DAS experiments: (i). Coupling condition and seabed incline determine the quality of the DAS data. (ii). The response of DAS is weak in capturing the ground motion of P-wave due to its different vibration characteristic with S-wave. At the same time, the biggest limitation of DAS in submarine is that the shallow sea area is adverse to observe due to the noise of surface gravity waves within 0.03~0.5 Hz and Scholte waves generated by solid-liquid interface coupling motion within 0.3~2 Hz (Sladen et al., 2019). It not only causes the waste of communication fiber optic cables resources in shallow sea, but also leads to the low utilization of natural local seismic waveforms. So it is necessary to introduce some denoising methods to process the observation data of submarine cable DAS in shallow ocean. In this article, we compared DAS records of 10 local earthquakes based on submarine fiber-optic cables with the records from broadband seismometers. The optical cable was early used in optical communication and its coupling with the submarine was unknown. In the case, some signals enhancement methods have been taken in the data processing, such as filtering, stacking and Discrete wavelets transform (DWT) as well as Ensemble Empirical Mode Decomposition (EEMD) Transform. Four of ten records show the high-similarity to the broadband seismic waveforms with clear p and s seismic phases, and the seismic signals in another six records are ambiguous. Then, we analyze the ability of local earthquake monitoring by communication cable in submarine, as well as the potential for a long-term and inexpensive observation way in deep ocean. Methods Observation System. The observation platform of this study is based on ISAC (Integrated Sensing and Communication) technology, in which the communication modules and sensing modules can be existed in one fiber without any interference (Chen et al., 2020). In this way, any optical fiber could be used for sensing especially the ones under the sea whether in using for communication or not. It immensely filled a gap in the field of undersea observation in seismology. Since the 1990s, the research on fiber sensing increased yearly (Zhan et al., 2019), such as earthquake detection (Marra et al., 2018; Williams et al., 2019), Illuminating seafloor faults (Nathaniel et al., 2019), etc. The distributed sensing technology has numerous advantages in seismology observation, but still need more effort to improve the signal-to-noise ratio (SNR) and application result for foundations building. We used a section of the communication cable in Pearl River Estuary for monitoring test and select one Mw3.2 event in DongYuan, Guangdong province of China for raw signal quality verification. Fig.1 shows the DAS observation system, which was set in Pearl River Estuary, the northern South China Sea area with the starting point on Sanjiao Island and the ending point on Guishan Island, with the submarine fiber-optic cable about 10km. The probe pulse was sent from the side of Sanjiao Island, of which the Rayleigh scattering occurred while it travels along the fiber. The backscattering light eventually would be received by the sensing modules installed on the island (Chen et al., 2020). There would be a phase difference of the backscattering light when vibration generated with the strain of the fiber changed. Hence, the accurate location where the fiber warped as well as the amplitude of the vibration can be easily calculated according to the travelling time of the backscattering light and the phase difference. For comparative verification, the records of ZhuHai broadband Seismic Station are used for waveform comparison. In the test, the spatial sampling rate is adjustable in the range of 2m, 4m and 8m, So the gauge length is also floating. Gauge length determines the measuring unit length, and long gauge length measures more effectively than short gauge length, but it is pointless for measuring of longer gauge length because of monitoring accuracy. Results of different Methods Raw signal. The volume of the data per day is an immense number with the sampling rate up to 1000Hz. It is formidable to process all the data, so we adopted the scheme that selecting DAS records according to the Earthquake catalogue. Here, we collected the records of ten earthquake events for research. This section will introduce the characteristic of raw waveform, and the comparison between it after processing and Zhuhai seismic station record. The representative one of the events is a 2021.9.4T07:51:52 Mw3.2 Earthquake in DongYuan city, Guangdong province, China, whose focal depth is 9km, as shown in Fig.1, in which the sampling rate is 100Hz and the spatial resolution is 2 m. So there are around 5000 observation points were arranged in 10km. The raw data recorded by the DAS undersea is in low quality, with the reason of the high sensitivity of the sensing fiber and the complex condition in the sea as well as the weak coupling between the fiber and the ground. Because the record of the ocean surface gravity waves is significant and periodic, it must be taken into consideration in our study. As the frequency of the surface waves is between 0.05Hz and 0.3Hz, and their amplitude decays rapidly with depth by the linear theory of gravity waves. In a DAS test, Sladen found that the distributed cable recorded the dynamic pressure produced by the surface gravity wave down to a depth of 100m, closed to the minimum with the amplitudes as low as 1 nstrain/s (nano-strain per second) were measured (Sladen et al., 2019). As shown in Fig.1, the average depth of the cable in our test is lower than 10m, so the effect of the surface wave is in an influential range and the signals of the earthquake is buried in the surface-wave acoustic noise with the frequency band below 0.3Hz. The continuous waveform transform (CWT) spectrum shows the noise level as shown in Fig.2 , in which wave noise with frequency less than 1 Hz, take the main part of the signals and the Pseudo Color(P-Color) figure is mainly recorded the sea wave. In Fig.2, waveforms show no signal about the earthquake events. In the frequency band of 0.02Hz-0.5Hz, the amplitude of the surface gravity wave is the main component. It is hard to observe the signals of earthquake in the other two one-channel waveform with the reason of the weak coupling and high-level noise in submarine. As the seismic signal is weak in the records of single channel, the stacking profile with P-Color image has often been used to enhance the signal to noise ratio. The frequency-wavenumber (f-k) domain analysis is used to separate coherent signals by using two-dimensional(2D) Fast Fourier Transform. Fig.3 shows that the ocean waves with a speed of less than 10m/s have the strongest energy, and the noise amplitude shows that during these 10 minutes of recording, the energy of waves in the westward direction is stronger, indicating that the waves are propagating towards the west in this time. The energy of Scholte waves with wave speeds of 100 m/s and 1000 m/s is slightly weaker than that of ocean waves, but it also completely covers the seismic signal, and as a surface wave, Scholte wave energy exhibits a certain dispersion propagation effect. Fig.4 shows waveforms with different frequency band filters, Fig 4b and 4c indicate that the first 2Km cable successfully receive the S wave, because the part is in a slope with the supporting force of gravity to get the better coupling. This test suggests the normal communication cable have the potential in undertaking the mission for submarine micro-earthquake monitoring, although it has the shortage of the unidentified coupling with the ground. On the other hand, it is easier to monitor a local earthquake with high frequency wave which will not be cover by the surface gravity wave without difficulty by high-frequency filtering. In the contrary, the frequency of the teleseism is usually in 0.02Hz-1Hz, in which the surface gravity is in dominating position. DAS responds to the external vibration or acoustic signal around the sensing fiber, whose phase change is found linear to the external vibration at a certain location. In the course of vibration recording by sensing fiber, Earthquake, acoustic noise, ocean waves, temperature variation and so on, would cause fiber strain change. In order to highlight the earthquake signals, some denoising method need to be proposed in data processing. This section mainly introduce three common methods, Stacking, Discrete Wavelet Transform(DWT) and Ensemble Empirical Mode Decomposition (EEMD) Transform. Fig.5 shows the map of 10 earthquakes in this study, all of which are local events with magnitudes around 3.0 MW (see Figure S1). In the fiber optic interception data of ten earthquake events, the sampling rate is 100Hz, and the sampling point intervals include 2m, 4m, and 8m, which correspond to 4756, 2378, and 1189 sampling points. Here we still take the event of 2021.9.4T07:51:52 Mw3.2 Earthquake in DongYuan as an example to demonstrate the data processing and the other events analysis result see in figure S2 in supplementary. Channel Stacking. In the test, every 2 meters a sensing unit is along the 10km optical cable. By the values superposition of several channels, which is also called simple stacking, the time-varying noise and optical noise produced by temperature fluctuation will be removed and the effect of noise suppression will be achieved. But the simple stack cannot be allow to adapt different earthquakes, so the overlap stack should be complied in the waveform processing. Overlap stack is usually used in the undersized samples for repetitive and efficient utilization as shown in Fig.6. In passive earthquake seismology, seismic phase analyzing is to deduce the origin of the specific seismic phases, and then determine the seismic source characteristics and the structure and properties of the earth's internal structures, So the seismic phases are the base of other intensive study. In the case that the signals of the DAS record are not so clear, stacking will suppress noise while eliminating the influence of temperature by gathering multiple traces of data. All the DAS stacking waveforms is arranged by fiber distance, is used for seismic phase analyzing. In the CWT spectrum of 30-traces stacking, 0.01Hz-2Hz is surface wave and Scholte oriented conforming to the test conclusion showed in Fig.3, so we use band-pass filtering of 2Hz-10Hz to keep the high frequency so that the part is shown about the event with P and S wave in Fig.7. And the P-Color show the clear signals alongside the fiber, but the middle part is not so clear because of the weak coupling possibly. In Fig.8, the DAS trace shows the high-similarity in the graph by comparing with the ZhuHai seismic station. However, there are still three shortages of DAS technique: (i). The wavefield recorded by the cable is dominated by S waves, which is inconsistent with the seismographic station record in which the P-waves are visible. Hence, the very small amplitude of the P-wave on the DAS which may be covered by noise is owing the lack of sensitivity ( Ethan et al.,2019 ); (ii). The frequency band of Earthquake signals of DAS is not as wide as the seismographic station records; (iii). The other seismic phases are fuzzy in DAS record. At the same time, No-signal events still show no waveform after stacking, while the SNR of other weak-signal events got enhanced along the fiber (see Figure S3-S5). Discrete Wavelets Transform. The shortages of the traditional FFT(Fourier Fast Transform) is obvious that fitting the time window is rigid and hard to setup for a accurate figure. In contrast, Discrete Wavelet Transform(DWT) was proposed to match a large range of the frequency band for separating variable back-scattering light frequency. Fig.9 shows the DWT result, and both are after stacking and 2Hz~10Hz filtering, and in both graph the P and S wave are obvious, but the DWT one wipe off the most of the jamming signals and retain the seismic phases signals，of which the wavelet is db8 and the method is soft threshold of 0.2. Through experimentation, discrete wavelet transform can greatly improve the SNR while preserving important seismic signals, which can greatly improve the utilization of data for shallow ocean fiber optic cable observations. But for weak signal P-waves, this method has a certain degree of weakening effect on P-waves, which is also the limitation of this method. Other DWT-processing events result see Figure S6. Ensemble Empirical Mode Decomposition (EEMD) Transform. Due to the limitations of discrete wavelet transform, this research attempts another method for deep denoising research, namely ensemble empirical mode decomposition (EEMD). EEMD is a data-analysis signal decomposition method used to decompose complex nonlinear signals into a set of intrinsic mode functions (IMFs), with each IMF representing an inherent vibration mode in the signal. The advantage of EEMD lies in its ability to reduce modal aliasing in EMD, enhance the ability to extract local features of signals, and have better robustness to noise compared to traditional EMD. The analysis results shown in Fig.10 show that for the seismic waveform after channels stacking, a total of 12 IMFs are decomposed. It can be seen that the vibration modes of each IMF have their unique properties, indicating that EEMD has good robustness for the data in this test. By analyzing the waveform characteristics and amplitude size of each IMF, and also combining the PSD spectrum characteristics of each decomposed IMF for frequency band selection, as shown in Fig.11, it can be seen that IMF2, IMF3, and IMF4 all have similar characteristics of seismic waves. However, due to the small amplitude of IMF2 and the excessive noise signal of IMF4, and the fact that the PSD peaks are mainly concentrated in the 2-10 Hz IMF, IMF2 and IMF2+IMF3 were selected as the seismic wave IMFs for this batch of data. Fig.12 show the waveforms after denoising with multiple EEMD channels. It can be observed that this method removes most of the noise, improves the signal-to-noise ratio better than the discrete wavelet transform, and leaves strong seismic waveforms, demonstrating the feasibility of the EEMD method. In addition, by selecting IMF [3] as the signal reference term, it can be seen that in the pseudo color map of IMF [2]+IMF [3], there is more preservation for P-waves, and the signal-to-noise ratio is similar. By using EEMD to denoise 5 event waveforms (see Figure S7), it was found that EEMD outperformed discrete wavelet transform in waveform fidelity, but did not significantly improve noise elimination and SNR. Furthermore, seismic signals could not be extracted for corresponding fuzzy events. Comparing two methods, the similarity lies in the fact that both can significantly improve the SNR and to some extent enhance the signal of the weak coupling section. The difference is that discrete wavelet transform is more effective in eliminating noise from strong signals, while EEMD is more effective in signal extraction. Selecting corresponding methods for processing observation data from different shallow ocean areas can greatly improve the SNR. Discussion Possible Causes of Undetected Earthquakes. By comparing the 10 earthquakes selected in the earthquake catalog, it can be found that 5 of them occurred around the Dongyuan Reservoir in Dongyuan County, Heyuan City, Guangdong Province. Therefore, the distance between their epicenters is almost equal. However, it has been proven that only three of these earthquakes can be detected, so the distance between the epicenters is clearly not the most important influencing factor. Additionally, it can also be proved that the characteristics of the medium and the terrain structure cannot be the main considerations for incomplete seismic detection in this test. In addition, due to the lack of data on each seismic source mechanism in this article, the source mechanism can lead to differences in the directionality, wavefront shape, energy distribution, and frequency characteristics of seismic waves, that is, different radiation patterns of different source mechanisms. Therefore, there is currently no evidence to rule out the influence of seismic source mechanisms on fiber optic sensing signals. For the characteristics of the station itself, optical fibers only have a single radial component, while fixed stations have three directional components for recording. This may also have a significant impact on the recording of seismic waves by optical fibers. Therefore, We compare 8 detected seismic events with the waveforms of ZhuHai seismic station to explore the reasons for weak reponse. Fig.13 show the waveforms comparison between broadband seismometers and DAS. The clear seismic phases can be detected by the first four optical cables, the amplitude recorded by the seismic station is relatively large. Correspondingly, the DAS also records the clear seismic signals. In the last four records, the amplitude values of the waves at seismic station are relatively weak, and the seismic signals in fiber optic sensors are also buried behind the noise signals. Therefore, through comparative analysis, it can be concluded that the clear waveform recorded by DAS signals is directly related to the waveform amplitude recorded by seismic station. The reason for the weak waveform recording at seismic stations is also the reason for the weak fiber optic signal recording. Based on the above analysis, it is possible that the unclear seismic signals of DAS might be caused by the different focal mechanisms and the combined effects of the above factors. Meanwhile, due to the single radial component of the optical fiber itself, it is difficult to capture weak seismic energy. The Future Application and Limitation. A great advantage of DAS is that it only requires an interrogation unit at one end of a telecommunication fiber optic cable, which makes it cost effective and easy to implement. There is already a vast installed base of telecommunication cables around the world; opportunistic DAS deployments utilizing this resource provide a low‐ cost approach for dense array acquisition. In some harsh environment, such as offshore, where conventional seismic deployment is difficult, DAS may be of more value in terms of seismic monitoring, subsurface imaging, and hazard assessment. One of the current limitations of DAS is its higher noise level than conventional broadband seismometers, The other one is the dependence of DAS to the ground coupling. In most cases the submarine cable are direct burial with multi-fibers because of the usage of communication, while the illustration might be casual with weak coupling with the submarine ground. In view of the difficulty of submarine construction, it is burdensome for reburial and hopeless to ask for communication construction crew burying with the demand for coupling standard. So, as a tentative submarine DAS test, we make the research according to the data quality, and what we can try is to make the best use of high-quality data and avoid the debased ones. Conclusions In 2020 and 2021, a new DAS field test was carried out in Pearl River Rstuary, Guangdong province, China. Using 10-events waveform data from the 4 September 2021 to 7 June 2022, we explore four Earthquake events with the high-similarity to the seismographic stations data successfully. There are three conclusions: (1) Fiber optic cable in shallow ocean is affected by surface gravitational waves of 0.03 Hz-0.5 Hz and primary and secondary microseismic noise (Scholte waves) generated by solid-liquid coupling motion at 0.3-2 Hz, resulting in seismic signals hidden in two types of higher energy noise signals. (2) By the processing and deep denoising method proposed in this article for optical cable signal processing in shallow sea areas, weak seismic signals can be extracted from high noise environments. Seismic phases include P-waves and S-waves, and seismic signals can be enhanced through waveform stacking, discrete wavelet transform, and ensemble empirical mode decomposition methods while recovering as much seismic phase information as possible. (3) This study processed ten near epicenters in the earthquake directory, and attempted to extract signals through deep denoising. Four clear seismic waveforms were observed. For other earthquakes, fiber optic seismic response was dim. By comparing the waveform with Zhuhai Seismic Station, it can be seen that the seismic station can record clear waveforms, and the submarine optical cable can also record them. For seismic events with low energy recorded by seismic stations, due to radiation patterns and single component optical fibers, the energy is covered by noise and cannot be observed. For fiber optic sensing technology as a new technology, in addition to using high signal-to-noise ratio land and deep ocean optical cables for observation, communication optical cables in shallow submarine can also be used for earthquake observation and analysis. For the widely distributed shallow-ocean optical cables are expected to be further developed and used. Declarations Acknowledgments This research is supported by the National Natural Science Foundation of China (Grant Number 42230305), the Second Tibetan Plateau Scientific Expedition and Research Program (STEP) (Grant Number 2019QZKK0701), Guangdong Province Introduced Innovative R&D Team (Grant Number 2017ZT07Z066), and Guangdong Science and Technology program (Grant Number 2023A1111120028). Data availability DAS datasets acquired during on-site expeditions conducted by the ISAC and are open to the public at the Figshare repository. The datasets used in the present study are available from the corresponding author on reasonable request. The ISAC data is available from the Figshare at https://doi.org/10.6084/m9.figshare.26390881.v1. References Abdalla FYO, Zhao YQ, Wu LW. Denoising ECG signal by complete EEMD adaptive noise[C]. 2017 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), 2017: 337-342. Bao, X and L Chen. Recent Progress in Distributed Fiber Optic Sensors[J]. Sensors, 2012, 12(7): p. 8601-8639. Chen, D, Q Liu and Z He. 108-km Distributed Acoustic Sensor With 220-pε/√Hz Strain Resolution and 5-m Spatial Resolution[J]. Journal of Lightwave Technology, 2019, 37(18): p. 4462-4468. Cheng, F, et al. Utilizing distributed acoustic sensing and ocean bottom fiber optic cables for submarine structural characterization[J]. Scientific Reports, 2021, 11(1). Chen SY, et al.Photonic Integrated Sensing and Communication System Harnessing Submarine Fiber-Optic Cables for Coastal Event Monitoring[J]. APL Photon. 2020,5, 030901. Calvert AJ. Ray tracing based prediction and subtraction of water-layer multiples [J]. Geophysics, 1990, 55: 443~445. Dou S, et al. Distributed Acoustic Sensing for Seismic Monitoring of The Near Surface: A TrafficNoise Interferometry Case Study[J]. Scientific Reports, 2017,7(1). Dong X, et al. Random and Coherent Noise Suppression in DAS-VSP Data by Using a Supervised Deep Learning Method[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: p. 1-5. F. Williams, et al. Distributed sensing of microseisms and teleseisms with submarine dark fibers[J]. Nat. Comm. vol. 10, 2019, Art. no. 5778. Fang G, et al. Urban Near‐Surface Seismic Monitoring Using Distributed Acoustic Sensing[J]. Geophysical Research Letters, 2020, 47(6). Fernandez-Ruiz M.R, et al. Seismic Monitoring With Distributed Acoustic Sensing From the Near-Surface to the Deep Oceans[J]. Journal of Lightwave Technology, 2022, 40(5): p. 1453-1463. Fernández-Ruiz M.R, et al. Distributed acoustic sensing for seismic activity monitoring[J]. APL Photonics, 2020, 5(3): p. 030901. Fabelinskii IL. Molecular scattering of light [M]. Springer Science & Business Media, 2012. Marra, et al. Ultrastable laser interferometry for earthquake detection with terrestrial and submarine cables[J], Science, 2018, vol. 361, pp. 486-490, . Huang N E, Shen Z, Long S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear nonstationary time series analysis[C]. Proceedings: Mathematical,Physical and Engineering Sciences. London The Royal Society Press, 1998, 454(1971):903-995. JB Ajo-Franklin, N.J. T.C. Dawe and J.B. Ajo-Franklin. Illuminating seafloor faults and ocean dynamics with dark fiber distributed acoustic sensing[J]. Science (American Association for the Advancement of Science), 2019. 366(6469): p. 1103-1107. Jousset P, et al. Dynamic strain determination using fibre-optic cables allows imaging of seismological and structural features[J]. Nature Communications, 2018. 9(1). Lindsey N.J, et al. Fiber‐Optic Network Observations of Earthquake Wavefields[J]. Geophysical Research Letters, 2017. 44(23). Lindsey N.J, H. Rademacher and J.B. Ajo Franklin. On the Broadband Instrument Response of Fiber‐Optic DAS Arrays[J]. Journal of Geophysical Research: Solid Earth, 2020. 125(2). Lior I, et al. Imaging an underwater basin and its resonance modes using optical fiber distributed acoustic sensing[J]. Seismological research letters, 2022, 93(3): p. 1573-1584. Lior I, et al. On the Detection Capabilities of Underwater Distributed Acoustic Sensing. Journal of Geophysical Research: Solid Earth, 2021. 126(3). Lamare, P. The MEUST deep sea infrastructure in the Toulon site. EPJ Web Conf. 2016, 116, 09001. Lu X. Coherent Rayleigh time domain reflectometry:novel applications for optical fibre sensing[D]. EPFL Swiss Federal Institute of Technology, 2016. Li X L, Li Z H, Wang E Y, et al. Analysis of natural mineral earthquake and blast based on Hilbert–Huang transform (HHT). Science Direct[J]. Journal of Applied Geophysics,2016, 128: 79-86. Martin, E.R. et al. A Seismic Shift in Scalable Acquisition Demands New Processing: Fiber-Optic Seismic Signal Retrieval in Urban Areas with Unsupervised Learning for Coherent Noise Removal[J]. IEEE Signal Processing Magazine, 2018. 35(2): p. 31-40. Matsumoto, H, et al. Detection of hydroacoustic signals on a fiber-optic submarine cable[J]. Scientific Reports, 2021, 11(1). Masoudi A, Newson TP. Contributed Review:Distributed optical fibre dynamic strain sensing[J]. Review of Scientific Instruments, 2016, 87: 011501. Mayne W H. Common-reflection-point horizontal data-stacking technique[J]. Geophysics, 1962,32,225-229. Nathaniel J. LindseyT. Craig DaweJonathan B. Ajo-Franklin. Illuminating seafloor faults and ocean dynamics with dark fiber distributed acoustic sensing[J]. Science, 2019, 366, 1103-1107. Olshansky R. Propagation in glass optical wave guides[J]. Reviews of Modern Physics. 1979, 51(2): 341. Roy S K, Molla M, Hirose K, et al. Pitch estimation of noisy speech signals using EMD fourier based hybrid algorithm[C]. IEEE International Symposium on Circuits & Systems.IEEE, 2010: 2658-2661. Rilling G, Flandrin P, Goncalves P. On empirical mode decomposition and its algorithms[J]. Journal of the Acoustical Society of America, 2003, 265(04): 785- 788. Sladen A, et al. Distributed sensing of earthquakes and ocean-solid Earth interactions on seafloor telecom cables[J]. Nature Communications, 2019. 10(1). Satoshi Ide, E.A.H.M. Very broadband strain-rate measurements along a submarine ber-optic cable off Cape Muroto, Nankai subduction zone, Japan[J]. Earth, Planets and Space, 2021. Schroeder J, Mohr R,Macedo PB. Rayleigh and Brillouin scattering in K2OSiO2 glasses [J] . Journal of the American Ceramic Society, 1973, 56(10): 510-514. Schoenberger M. Optimum weighted stack for multiple suppression [J]. Geophysics, 1996, 61: 891-901. Tyapkin Yu K. Optimized estimates of a complicated model of the multichannel seismic record with statistic and deterministic regularization[J]. Russian Geology and Geophysics, 1994, 35(1): 109-115. Verschuur DJ, Berkhout AJ, Wapenaarc PA. Adaptive surface related multiple elimination [J]. Geophysics, 1992, 57:1167-1177. Walter F, et al. Distributed acoustic sensing of microseismic sources and wave propagation in glaciated terrain[J]. Nature Communications, 2020. 11(1). Widess MB, et al. Quanfifying resolving power of seismic systems[J]. Geophysics, 1982, 47(8): 1160-1173. White R E. The performance of optimum stacking filters in suppressing uncorrelated noise [J].Geophysical Prospecting,1977,25:165-178. Yu, C, Zhan, Z. Lindsey, N. J. Ajo‐ Franklin, J. B. & Robertson, M. The potential of DAS in teleseismic studies: Insights from the Goldstone experiment[J]. Geophysical Research Letters, 2019,46, 1320-1328. Zhao A B, Shen G N, Chen Y, et al. The application of HHT and neural network in feature extraction of ship targets[J]. Technical Acoustics, 2012, 31(03): 272-276. Zhan, Z. Distributed acoustic sensing turns fiber-optic cables into sensitive seismic antennas[J]. Seismological research letters, 2019. 91(1): p. 1-15. Additional Declarations No competing interests reported. Supplementary Files Supplementary.pdf Supplementary.pdf Cite Share Download PDF Status: Published Journal Publication published 15 Apr, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 20 Sep, 2024 Reviews received at journal 17 Sep, 2024 Reviews received at journal 15 Sep, 2024 Reviewers agreed at journal 11 Sep, 2024 Reviewers agreed at journal 29 Aug, 2024 Reviewers agreed at journal 26 Aug, 2024 Reviewers invited by journal 21 Aug, 2024 Editor assigned by journal 19 Aug, 2024 Editor invited by journal 13 Aug, 2024 Submission checks completed at journal 10 Aug, 2024 First submitted to journal 27 Jul, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4811181","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":349831936,"identity":"2b4bb822-a5ad-4401-b3db-281125615ebd","order_by":0,"name":"Hao Li","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Hao","middleName":"","lastName":"Li","suffix":""},{"id":349831938,"identity":"31d8cbd3-04df-498d-9417-c3bde829401e","order_by":1,"name":"Xuzhang Shen","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBUlEQVRIiWNgGAWjYDACCTB5oJ4fRD6AifIQoSVBsgFEkqTF4ACQIkqL/OzmY495/tzJM752+CHQlrrE+TMSGB+8bWOQN8ehxeDOsXTDGTzPis1upxkAtRxO3HAjgdlwbhuD4c4GHFokcswkPkgcZtx2OwGk5UDiBokENmneNgaIU7E5bEb+N4kEg8OMm2enf4A5jP03Pi0MN3LYJD4kAN0jnQOyhTmx4UYCGzM+LQY30swkZxw4bCxxO6cAGG6HjTecedgsOeechOEGnA5LfibN8+ewHP/s9M0fPlTUyc5vTz744U2ZjTxOh6FZyuDYwMDYwACLL6KAPfFKR8EoGAWjYKQAABq7Y9RLB/rTAAAAAElFTkSuQmCC","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Xuzhang","middleName":"","lastName":"Shen","suffix":""},{"id":349831940,"identity":"573b00e5-4d31-413b-aa79-efe61db45268","order_by":2,"name":"Zhaohui Li","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Zhaohui","middleName":"","lastName":"Li","suffix":""},{"id":349831943,"identity":"a524a463-8895-49b9-b296-d11c37ea0bb5","order_by":3,"name":"Shaoyi Chen","email":"","orcid":"","institution":"Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Shaoyi","middleName":"","lastName":"Chen","suffix":""}],"badges":[],"createdAt":"2024-07-27 05:36:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4811181/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4811181/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-93682-2","type":"published","date":"2025-04-15T15:56:57+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":64069494,"identity":"db4c0f9a-bc36-4c36-9dca-e74220adcf06","added_by":"auto","created_at":"2024-09-06 06:32:15","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":4897563,"visible":true,"origin":"","legend":"\u003cp\u003eMap of the observed system\u003c/p\u003e\n\u003cp\u003e(a) Map of the earthquake events (red star), optical cable and broadband seismic station(triangle); (b) Map of the Pearl River Estuary area and submarine cable area; (c) details of the test cable from San Jiao Island and GuiShan Island with the length of 10 kilometers.\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/dd3eca385aa1355717c00d03.png"},{"id":64069496,"identity":"970888ba-88ef-438c-947d-9cfecf6f3fb1","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1499718,"visible":true,"origin":"","legend":"\u003cp\u003eRaw data analysis figures in the 3.2Mw event. (a) P-Color of the DAS signals among 100 seconds after the origin time; (b) The image of the continuous waveform transform spectrum as well as three waveform with different frequency band, 0.02Hz-0.5Hz, 0.5Hz-2Hz, 2Hz-10Hz.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/b4b71a44814dd34dda66dff3.png"},{"id":64069509,"identity":"d09a8468-82d3-4596-8749-fc75c260c773","added_by":"auto","created_at":"2024-09-06 06:32:17","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":2870816,"visible":true,"origin":"","legend":"\u003cp\u003eFrequency wavenumber spectrum (FK Analysis) of the 10 minute recorded Mw3.2 magnitude earthquake on September 4, 2021 at 07:51:52 in Dongyuan\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/13223ec7a3637a5f0531535e.png"},{"id":64069675,"identity":"f02c741e-8afa-444c-bb47-604563a39136","added_by":"auto","created_at":"2024-09-06 06:40:16","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":4461395,"visible":true,"origin":"","legend":"\u003cp\u003eOne-channel P-Color of the DAS signals among 100 seconds after the origin time and the clear detail views. (a) 0.02Hz-0.5Hz band-pass filtering p-color shows the information of the surface gravity wave which occupy the main amplitude. (b) 0.5Hz-2Hz band-pass filtering p-color contains the waveform of the surface gravity wave and weak S wave. (c) 2Hz-10Hz band-pass filtering p-color in the front part of the cable go down to submarine with tilt support force display kind of S wave.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/b9c40d67a253a47858f2bf47.png"},{"id":64069506,"identity":"8f31a55c-677a-461f-a6de-39322c9b58f0","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":788518,"visible":true,"origin":"","legend":"\u003cp\u003eMap of Earthquakes and stations . Red star represents the DAS observation platform, red triangle is Zhuhai seismic station; The circles are Earthquake events, green for clear waveform events detected by DAS and blue for no-signal events.\u003c/p\u003e","description":"","filename":"image5.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/759f0ff41e94f577e59f31c2.png"},{"id":64069505,"identity":"f55abecc-b382-4e31-9652-c3828e0f86e3","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":20409556,"visible":true,"origin":"","legend":"\u003cp\u003eAbout 10km DAS array in two different stacking methods. Simple stack is used to universal application for waveform detection with the shortage of incapablity of reverting seismic phase. Overlap stack shows the ability of expanding capacity of DAS traces.\u003c/p\u003e","description":"","filename":"image6.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/2c71d6f78309db40250e9848.png"},{"id":64069510,"identity":"7bbde39d-0c78-4798-92af-f5844f7fc3a9","added_by":"auto","created_at":"2024-09-06 06:32:17","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":720313,"visible":true,"origin":"","legend":"\u003cp\u003eThe waveform and its analysis after 30-traces stacking. (a) P-color reveals the clearer waveform of P wave and S wave. (b) The cwt spectrogram of single trace after stack. (c) Single trace waveform comparison (red : after-stack; grey : original single trace) with the band-pass filter of 2Hz - 10Hz.\u003c/p\u003e","description":"","filename":"image7.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/d2dd882f740ee841463b7ac3.png"},{"id":64069502,"identity":"53c14002-2045-4a37-9dc4-fad318ec9576","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":248366,"visible":true,"origin":"","legend":"\u003cp\u003eThe waveform comparison between DAS fiber and ZhuHai seismographic station record. The cwt spectrogram and the waveform show high-similarity, but high-level noise exist in DAS record.\u003c/p\u003e","description":"","filename":"image8.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/d29dc47117f51e60902ccca5.png"},{"id":64069501,"identity":"c7393b8b-0fd4-4c15-9ee7-bfaa2152d90f","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":1376492,"visible":true,"origin":"","legend":"\u003cp\u003e(a) P-color map after stack. (b) P-color map after stack and DWT. Discrete wavelet transform eliminate most of the noise and retain the P and S waveform according to the different frequency bands and threhold parameter. The DWT result shows the malpractice of P wave elimination in the case that the wave amplitude is equivalent to the noise.\u003c/p\u003e","description":"","filename":"image9.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/f9b82f70323e49a61ecd689c.png"},{"id":64069834,"identity":"f5118016-4c7b-4eba-ae4a-e7b0c8f8b2da","added_by":"auto","created_at":"2024-09-06 06:48:16","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":929820,"visible":true,"origin":"","legend":"\u003cp\u003eSeismic waveforms after single channel stacking and 12 different vibration waveforms generated after EEMD analysis.IMFs(5-11) are ignored due to micro waveforms.\u003c/p\u003e","description":"","filename":"image10.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/c10666126f3806441c032c75.png"},{"id":64069503,"identity":"e83dc91d-d92b-4cd4-8ceb-b45892ca7a2a","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":618108,"visible":true,"origin":"","legend":"\u003cp\u003eSeismic Waveform after Single Channel Stacking and Power Spectral Density of the First 5 IMFs after EEMD Analysis\u003c/p\u003e","description":"","filename":"image11.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/43d3e83364a786dd2589f20d.png"},{"id":64069674,"identity":"074a3041-52dc-4d68-a1c0-c2cec24d98e9","added_by":"auto","created_at":"2024-09-06 06:40:16","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":4188562,"visible":true,"origin":"","legend":"\u003cp\u003eComparison between Multi channel Stacked Waveform and EEMD Denoising Waveform. (a) The figure shows the results of 30 stacked waveforms after EEMD of IMF [2] as seismic waves, including p-color map and the SNR along the fiber as well as the max SNR waveform compared with Zhuhai station. (b) the results of after EEMD of IMF [2]+IMF [3] as seismic waves. The second one contains more details but more noise.\u003c/p\u003e","description":"","filename":"image12.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/cc9d5bfd233bef88d8d6d7ca.png"},{"id":64069504,"identity":"491baaea-6e20-4177-9acd-88193a63d603","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":596993,"visible":true,"origin":"","legend":"\u003cp\u003eThe comparison of eight DAS records and ZhuHai seismographic station records. (red ：DAS array waveform; green ： Zhuhai seismographic station record), broadband records are in global-max normalization for amplitude comparison. The first four DAS records show the clear Earthquake waveform in the case that the amplitude of the broadband seismographic waveform is in high level. On the contrary, the other DAS records about S wave are coverd by noise, when the broadband records are in low level.\u003c/p\u003e","description":"","filename":"image13.png","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/06a39df50fb2ee75690844f8.png"},{"id":81051510,"identity":"e1112527-8fb6-4545-94cb-75e11e4e6763","added_by":"auto","created_at":"2025-04-21 16:10:53","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":46023843,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/67a6541d-ea4d-4c6a-85d0-e7004c307d47.pdf"},{"id":64069495,"identity":"a762ac76-dba9-4a2e-9f2e-46f5135746a3","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":4394088,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/d769ccf9dd752fdb2a5b6a20.pdf"},{"id":64069498,"identity":"727e210c-5b40-48f8-8da6-fffa4737f678","added_by":"auto","created_at":"2024-09-06 06:32:16","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":4394088,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4811181/v1/a716fb4abdd3648d707d4346.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Local Earthquake Response on the Submarine Communication Cable in The Northern South China Sea","fulltext":[{"header":"Introduction","content":"\u003cp\u003eDistributed acoustic sensing (DAS) is a newly developed technique that transforms telecommunication fiber optic cables into linear arrays of ground-motion sensors. Compared with the conventional seismometer, DAS not only make great cost saving, but also have an advantage in dense spatial sampling. In the last three decades, DAS has continually been using in gas industry as well as vertical seismic profiling. Meanwhile, the seismology was making progress towards the ocean study. There are still huge gaps in ocean study nevertheless on account of the aesthetic maintenance cost and recovery difficulty in OBS layout in deep ocean. However, it became effortless to capture the ground-motion signals when the DAS technique by using communication optical fiber cable under the sea came into sight. Above all, the extensive coverage of optical fiber around the Earth surface provides a highly operational observation platform for seismology researchers (Yu et al., 2019). Without seabed burying operation and observation duration limit, the DAS system have a long observation duration up to several months or even one year when the general geophones can just last for the boundary of couple of months. On the other hand, dense array of meter scale and more than 100km long-distance effective observation distance highlight its advantages with respect to traditional techniques.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe application of DAS in passive earthquake seismology, however, is still at its early stage. Last few years, several ocean DAS experiments have been carried out with the high degree of correlation with a seismograph waveform. For example, Ethan et al (2019), analyzed the waveforms of 2018-08-19 Mw8.2 Fiji teleseismic deep earthquake recorded by the DAS array in the Southern Bight of the North shallow Sea offshore Zeebrugge, Belgium. They found that DAS waveforms show a high degree of correlation with a seismometer record (Ethan et al., 2019). In the MEUST-NUMerEnv project in Toulon, France, Sladen et al. (2019) also found coherent earthquake waveforms recorded at a dense DAS array in deep ocean showing the high sensitivity as a dense geophone array from a local M1.9 event (Sladen et al., 2019). Two experiments show the capacity of DAS system of detecting the earthquake and even the micro earthquake. However, some shortages as a newcomer were exposed in most of the shallow ocean DAS experiments: (i). Coupling condition and seabed incline determine the quality of the DAS data. (ii). The response of DAS is weak in capturing the ground motion of P-wave due to its different vibration characteristic with S-wave. At the same time, the biggest limitation of DAS in submarine is that the shallow sea area is adverse to observe due to the noise of surface gravity waves within 0.03~0.5 Hz and Scholte waves generated by solid-liquid interface coupling motion within 0.3~2 Hz (Sladen et al., 2019). It not only causes the waste of communication fiber optic cables resources in shallow sea, but also leads to the low utilization of natural local seismic waveforms. So it is necessary to introduce some denoising methods to process the observation data of submarine cable DAS in shallow ocean.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn this article, we compared DAS records of 10 local earthquakes based on submarine fiber-optic cables with the records from broadband seismometers. The optical cable was early used in optical communication and its coupling with the submarine was unknown. In the case, some signals enhancement methods have been taken in the data processing, such as filtering, stacking and Discrete wavelets transform (DWT) as well as Ensemble Empirical Mode Decomposition (EEMD) Transform. Four of ten records show the high-similarity to the broadband seismic waveforms with clear p and s seismic phases, and the seismic signals in another six records are ambiguous. Then, we analyze the ability of local earthquake monitoring by communication cable in submarine, as well as the potential for a long-term and inexpensive observation way in deep ocean.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eObservation System.\u003c/strong\u003e The observation platform of this study is based on ISAC (Integrated Sensing and Communication) technology, in which the communication modules and sensing modules can be existed in one fiber without any interference (Chen et al., 2020). In this way, any optical fiber could be used for sensing especially the ones under the sea whether in using for communication or not. It immensely filled a gap in the field of undersea observation in seismology. Since the 1990s, the research on fiber sensing increased yearly (Zhan et al., 2019), such as earthquake detection (Marra et al., 2018; Williams et al., 2019), Illuminating seafloor faults (Nathaniel et al., 2019), etc. The distributed sensing technology has numerous advantages in seismology observation, but still need more effort to improve the signal-to-noise ratio (SNR) and application result for foundations building. We used a section of the communication cable in Pearl River Estuary for monitoring test and select one Mw3.2 event in DongYuan, Guangdong province of China for raw signal quality verification.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFig.1 shows the DAS observation system, which was set in Pearl River Estuary, the northern South China Sea area with the starting point on Sanjiao Island and the ending point on Guishan Island, with the submarine fiber-optic cable about 10km. The probe pulse was sent from the side of Sanjiao Island, of which the Rayleigh scattering occurred while it travels along the fiber. The backscattering light eventually would be received by the sensing modules installed on the island (Chen et al., 2020). There would be a phase difference of the backscattering light when vibration generated with the strain of the fiber changed. Hence, the accurate location where the fiber warped as well as the amplitude of the vibration can be easily calculated according to the travelling time of the backscattering light and the phase difference. For comparative verification, the records of ZhuHai broadband Seismic Station are used for waveform comparison. In the test, the spatial sampling rate is adjustable in the range of 2m, 4m and 8m, So the gauge length is also floating. Gauge length determines the measuring unit length, and long gauge length measures more effectively than short gauge length, but it is pointless for measuring of longer gauge length because of monitoring accuracy.\u003c/p\u003e\n\u003ch3\u003eResults of different Methods\u003c/h3\u003e\n\u003cp\u003e\u003cstrong\u003eRaw signal. \u003c/strong\u003eThe volume of the data per day is an immense number with the sampling rate up to 1000Hz. It is formidable to process all the data, so we adopted the scheme that selecting DAS records according to the Earthquake catalogue. Here, we collected the records of ten earthquake events for research. This section will introduce the characteristic of raw waveform, and the comparison between it after processing and Zhuhai seismic station record. The representative one of the events is a 2021.9.4T07:51:52 Mw3.2 Earthquake in DongYuan city, Guangdong province, China, whose focal depth is 9km, as shown in Fig.1, in which the sampling rate is 100Hz and the spatial resolution is 2 m. So there are around 5000 observation points were arranged in 10km.\u003c/p\u003e\n\u003cp\u003eThe raw data recorded by the DAS undersea is in low quality, with the reason of the high sensitivity of the sensing fiber and the complex condition in the sea as well as the weak coupling between the fiber and the ground.\u003c/p\u003e\n\u003cp\u003eBecause the record of the ocean surface gravity waves is significant and periodic, it must be taken into consideration in our study. As the frequency of the surface waves is between 0.05Hz and 0.3Hz, and their amplitude decays rapidly with depth by the linear theory of gravity waves. In a DAS test, Sladen found that the distributed cable recorded the dynamic pressure produced by the surface gravity wave down to a depth of 100m, closed to the minimum with the amplitudes as low as 1 nstrain/s (nano-strain per second) were measured (Sladen et al., 2019). As shown in Fig.1, the average depth of the cable in our test is lower than 10m, so the effect of the surface wave is in an influential range and the signals of the earthquake is buried in the surface-wave acoustic noise with the frequency band below 0.3Hz. The continuous waveform transform (CWT) spectrum shows the noise level as shown in Fig.2 , in which wave noise with frequency less than 1 Hz, take the main part of the signals and the Pseudo Color(P-Color) figure is mainly recorded the sea wave.\u003c/p\u003e\n\u003cp\u003eIn Fig.2, waveforms show no signal about the earthquake events. In the frequency band of 0.02Hz-0.5Hz, the amplitude of the surface gravity wave is the main component. It is hard to observe the signals of earthquake in the other two one-channel waveform with the reason of the weak coupling and high-level noise in submarine. As the seismic signal is weak in the records of single channel, the stacking profile with P-Color image has often been used to enhance the signal to noise ratio. The frequency-wavenumber (f-k) domain analysis is used to separate coherent signals by using two-dimensional(2D) Fast Fourier Transform. Fig.3 shows that the ocean waves with a speed of less than 10m/s have the strongest energy, and the noise amplitude shows that during these 10 minutes of recording, the energy of waves in the westward direction is stronger, indicating that the waves are propagating towards the west in this time. The energy of Scholte waves with wave speeds of 100 m/s and 1000 m/s is slightly weaker than that of ocean waves, but it also completely covers the seismic signal, and as a surface wave, Scholte wave energy exhibits a certain dispersion propagation effect.\u003c/p\u003e\n\u003cp\u003eFig.4 shows waveforms with different frequency band filters, Fig 4b and 4c indicate that the first 2Km cable successfully receive the S wave, because the part is in a slope with the supporting force of gravity to get the better coupling. This test suggests the normal communication cable have the potential in undertaking the mission for submarine micro-earthquake monitoring, although it has the shortage of the unidentified coupling with the ground. On the other hand, it is easier to monitor a local earthquake with high frequency wave which will not be cover by the surface gravity wave without difficulty by high-frequency filtering. In the contrary, the frequency of the teleseism is usually in 0.02Hz-1Hz, in which the surface gravity is in dominating position.\u003c/p\u003e\n\u003cp\u003eDAS responds to the external vibration or acoustic signal around the sensing fiber, whose phase change is found linear to the external vibration at a certain location. In the course of vibration recording by sensing fiber, Earthquake, acoustic noise, ocean waves, temperature variation and so on, would cause fiber strain change. In order to highlight the earthquake signals, some denoising method need to be proposed in data processing. This section mainly introduce three common methods, Stacking, Discrete Wavelet Transform(DWT) and Ensemble Empirical Mode Decomposition (EEMD) Transform.\u003c/p\u003e\n\u003cp\u003eFig.5 shows the map of 10 earthquakes in this study, all of which are local events with magnitudes around 3.0 MW (see Figure S1). In the fiber optic interception data of ten earthquake events, the sampling rate is 100Hz, and the sampling point intervals include 2m, 4m, and 8m, which correspond to 4756, 2378, and 1189 sampling points.\u003c/p\u003e\n\u003cp\u003eHere we still take the event of 2021.9.4T07:51:52 Mw3.2 Earthquake in DongYuan as an example to demonstrate the data processing and the other events analysis result see in figure S2 in supplementary.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eChannel Stacking.\u003c/strong\u003e In the test, every 2 meters a sensing unit is along the 10km optical cable. By the values superposition of several channels, which is also called simple stacking, the time-varying noise and optical noise produced by temperature fluctuation will be removed and the effect of noise suppression will be achieved. But the simple stack cannot be allow to adapt different earthquakes, so the overlap stack should be complied in the waveform processing. Overlap stack is usually used in the undersized samples for repetitive and efficient utilization as shown in Fig.6.\u003c/p\u003e\n\u003cp\u003eIn passive earthquake seismology, seismic phase analyzing is to deduce the origin of the specific seismic phases, and then determine the seismic source characteristics and the structure and properties of the earth's internal structures, So the seismic phases are the base of other intensive study. In the case that the signals of the DAS record are not so clear, stacking will suppress noise while eliminating the influence of temperature by gathering multiple traces of data. All the DAS stacking waveforms is arranged by fiber distance, is used for seismic phase analyzing. In the CWT spectrum of 30-traces stacking, 0.01Hz-2Hz is surface wave and Scholte oriented conforming to the test conclusion showed in Fig.3, so we use band-pass filtering of 2Hz-10Hz to keep the high frequency so that the part is shown about the event with P and S wave in Fig.7. And the P-Color show the clear signals alongside the fiber, but the middle part is not so clear because of the weak coupling possibly.\u003c/p\u003e\n\u003cp\u003eIn Fig.8, the DAS trace shows the high-similarity in the graph by comparing with the ZhuHai seismic station. However, there are still three shortages of DAS technique: \u003cem\u003e(i). The wavefield recorded by the cable is dominated by S waves, which is inconsistent with the seismographic station record in which the P-waves are visible. Hence, the very small amplitude of the P-wave on the DAS which may be covered by noise is owing the lack of sensitivity (\u003c/em\u003eEthan et al.,2019\u003cem\u003e); (ii). The frequency band of Earthquake signals of DAS is not as wide as the seismographic station records; (iii). The other seismic phases are fuzzy in DAS record.\u003c/em\u003e At the same time, No-signal events still show no waveform after stacking, while the SNR of other weak-signal events got enhanced along the fiber (see Figure S3-S5).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscrete Wavelets Transform. \u003c/strong\u003eThe shortages of the traditional FFT(Fourier Fast Transform) is obvious that fitting the time window is rigid and hard to setup for a accurate figure. In contrast, Discrete Wavelet Transform(DWT) was proposed to match a large range of the frequency band for separating variable back-scattering light frequency. Fig.9 shows the DWT result, and both are after stacking and 2Hz~10Hz filtering, and in both graph the P and S wave are obvious, but the DWT one wipe off the most of the jamming signals and retain the seismic phases signals，of which the wavelet is db8 and the method is soft threshold of 0.2.\u003c/p\u003e\n\u003cp\u003eThrough experimentation, discrete wavelet transform can greatly improve the SNR while preserving important seismic signals, which can greatly improve the utilization of data for shallow ocean fiber optic cable observations. But for weak signal P-waves, this method has a certain degree of weakening effect on P-waves, which is also the limitation of this method. Other DWT-processing events result see Figure S6.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEnsemble Empirical Mode Decomposition (EEMD) Transform. \u003c/strong\u003eDue to the limitations of discrete wavelet transform, this research attempts another method for deep denoising research, namely ensemble empirical mode decomposition (EEMD). EEMD is a data-analysis signal decomposition method used to decompose complex nonlinear signals into a set of intrinsic mode functions (IMFs), with each IMF representing an inherent vibration mode in the signal. The advantage of EEMD lies in its ability to reduce modal aliasing in EMD, enhance the ability to extract local features of signals, and have better robustness to noise compared to traditional EMD.\u003c/p\u003e\n\u003cp\u003eThe analysis results shown in Fig.10 show that for the seismic waveform after channels stacking, a total of 12 IMFs are decomposed. It can be seen that the vibration modes of each IMF have their unique properties, indicating that EEMD has good robustness for the data in this test.\u003c/p\u003e\n\u003cp\u003eBy analyzing the waveform characteristics and amplitude size of each IMF, and also combining the PSD spectrum characteristics of each decomposed IMF for frequency band selection, as shown in Fig.11, it can be seen that IMF2, IMF3, and IMF4 all have similar characteristics of seismic waves. However, due to the small amplitude of IMF2 and the excessive noise signal of IMF4, and the fact that the PSD peaks are mainly concentrated in the 2-10 Hz IMF, IMF2 and IMF2+IMF3 were selected as the seismic wave IMFs for this batch of data.\u003c/p\u003e\n\u003cp\u003eFig.12 show the waveforms after denoising with multiple EEMD channels. It can be observed that this method removes most of the noise, improves the signal-to-noise ratio better than the discrete wavelet transform, and leaves strong seismic waveforms, demonstrating the feasibility of the EEMD method. In addition, by selecting IMF [3] as the signal reference term, it can be seen that in the pseudo color map of IMF [2]+IMF [3], there is more preservation for P-waves, and the signal-to-noise ratio is similar.\u003c/p\u003e\n\u003cp\u003eBy using EEMD to denoise 5 event waveforms (see Figure S7), it was found that EEMD outperformed discrete wavelet transform in waveform fidelity, but did not significantly improve noise elimination and SNR. Furthermore, seismic signals could not be extracted for corresponding fuzzy events.\u003c/p\u003e\n\u003cp\u003eComparing two methods, the similarity lies in the fact that both can significantly improve the SNR and to some extent enhance the signal of the weak coupling section. The difference is that discrete wavelet transform is more effective in eliminating noise from strong signals, while EEMD is more effective in signal extraction. Selecting corresponding methods for processing observation data from different shallow ocean areas can greatly improve the SNR.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003e\u003cstrong\u003ePossible Causes of Undetected Earthquakes.\u0026nbsp;\u003c/strong\u003eBy comparing the 10 earthquakes selected in the earthquake catalog, it can be found that 5 of them occurred around the Dongyuan Reservoir in Dongyuan County, Heyuan City, Guangdong Province. Therefore, the distance between their epicenters is almost equal. However, it has been proven that only three of these earthquakes can be detected, so the distance between the epicenters is clearly not the most important influencing factor. Additionally, it can also be proved that the characteristics of the medium and the terrain structure cannot be the main considerations for incomplete seismic detection in this test. In addition, due to the lack of data on each seismic source mechanism in this article, the source mechanism can lead to differences in the directionality, wavefront shape, energy distribution, and frequency characteristics of seismic waves, that is, different radiation patterns of different source mechanisms. Therefore, there is currently no evidence to rule out the influence of seismic source mechanisms on fiber optic sensing signals. For the characteristics of the station itself, optical fibers only have a single radial component, while fixed stations have three directional components for recording. This may also have a significant impact on the recording of seismic waves by optical fibers. Therefore, \u0026nbsp;We compare 8 detected seismic events with the waveforms of ZhuHai seismic station to explore the reasons for weak reponse.\u003c/p\u003e\n\u003cp\u003eFig.13 show the waveforms comparison between broadband seismometers and DAS. The clear seismic phases can be detected by the first four optical cables, the amplitude recorded by the seismic station is relatively large. Correspondingly, the DAS also records the clear seismic signals. In the last four records, the amplitude values of the waves at seismic station are relatively weak, and the seismic signals in fiber optic sensors are also buried behind the noise signals. Therefore, through comparative analysis, it can be concluded that the clear waveform recorded by DAS signals is directly related to the waveform amplitude recorded by seismic station. The reason for the weak waveform recording at seismic stations is also the reason for the weak fiber optic signal recording.\u003c/p\u003e\n\u003cp\u003eBased on the above analysis, it is possible that the unclear seismic signals of DAS might be caused by the different focal mechanisms and the combined effects of the above factors. Meanwhile, due to the single radial component of the optical fiber itself, it is difficult to capture weak seismic energy.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe Future Application and Limitation.\u0026nbsp;\u003c/strong\u003eA great advantage of DAS is that it only requires an interrogation unit at one end of a telecommunication fiber optic cable, which makes it cost effective and easy to implement. There is already a vast installed base of telecommunication cables around the world; opportunistic DAS deployments utilizing this resource provide a low‐ cost approach for dense array acquisition. In some harsh environment, such as offshore, where conventional seismic deployment is difficult, DAS may be of more value in terms of seismic monitoring, subsurface imaging, and hazard assessment.\u003c/p\u003e\n\u003cp\u003eOne of the current limitations of DAS is its higher noise level than conventional broadband seismometers, The other one is the dependence of DAS to the ground coupling. In most cases the submarine cable are direct burial with multi-fibers because of the usage of communication, while the illustration might be casual with weak coupling with the submarine ground. In view of the difficulty of submarine construction, it is burdensome for reburial and hopeless to ask for communication construction crew burying with the demand for coupling standard. So, as a tentative submarine DAS test, we make the research according to the data quality, and what we can try is to make the best use of high-quality data and avoid the debased ones.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003ch6\u003eConclusions\u003c/h6\u003e\n\u003cp\u003eIn 2020 and 2021, a new DAS field test was carried out in Pearl River Rstuary, Guangdong province, China. Using 10-events waveform data from the 4 September 2021 to 7 June 2022, we explore four Earthquake events with the high-similarity to the seismographic stations data successfully. There are three conclusions:\u003c/p\u003e\n\u003cp\u003e(1) Fiber optic cable in shallow ocean is affected by surface gravitational waves of 0.03 Hz-0.5 Hz and primary and secondary microseismic noise (Scholte waves) generated by solid-liquid coupling motion at 0.3-2 Hz, resulting in seismic signals hidden in two types of higher energy noise signals.\u003c/p\u003e\n\u003cp\u003e(2) By the processing and deep denoising method proposed in this article for optical cable signal processing in shallow sea areas, weak seismic signals can be extracted from high noise environments. Seismic phases include P-waves and S-waves, and seismic signals can be enhanced through waveform stacking, discrete wavelet transform, and ensemble empirical mode decomposition methods while recovering as much seismic phase information as possible.\u003c/p\u003e\n\u003cp\u003e(3) This study processed ten near epicenters in the earthquake directory, and attempted to extract signals through deep denoising. Four clear seismic waveforms were observed. For other earthquakes, fiber optic seismic response was dim. By comparing the waveform with Zhuhai Seismic Station, it can be seen that the seismic station can record clear waveforms, and the submarine optical cable can also record them. For seismic events with low energy recorded by seismic stations, due to radiation patterns and single component optical fibers, the energy is covered by noise and cannot be observed.\u003c/p\u003e\n\u003cp\u003eFor fiber optic sensing technology as a new technology, in addition to using high signal-to-noise ratio land and deep ocean optical cables for observation, communication optical cables in shallow submarine can also be used for earthquake observation and analysis. For the widely distributed shallow-ocean optical cables are expected to be further developed and used.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch6\u003eAcknowledgments\u003c/h6\u003e\n\u003cp\u003eThis research is supported by the National Natural Science Foundation of China (Grant Number 42230305), the Second Tibetan Plateau Scientific Expedition and Research Program (STEP) (Grant Number 2019QZKK0701), Guangdong Province Introduced Innovative R\u0026amp;D Team (Grant Number 2017ZT07Z066), and Guangdong Science and Technology program (Grant Number 2023A1111120028).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDAS datasets acquired during on-site expeditions conducted by the ISAC and are open to the public at the Figshare repository. The datasets used in the present study are available from the corresponding author on reasonable request. The ISAC data is available from the Figshare at https://doi.org/10.6084/m9.figshare.26390881.v1.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAbdalla FYO, Zhao YQ, Wu LW. Denoising ECG signal by complete EEMD adaptive noise[C]. 2017 IEEE International Symposium on Signal Processing and Information Technology (ISSPIT), 2017: 337-342.\u003c/li\u003e\n\u003cli\u003eBao, X and L Chen. Recent Progress in Distributed Fiber Optic Sensors[J]. Sensors, 2012, 12(7): p. 8601-8639.\u003c/li\u003e\n\u003cli\u003eChen, D, Q Liu and Z He. 108-km Distributed Acoustic Sensor With 220-p\u0026epsilon;/\u0026radic;Hz Strain Resolution and 5-m Spatial Resolution[J]. Journal of Lightwave Technology, 2019, 37(18): p. 4462-4468.\u003c/li\u003e\n\u003cli\u003eCheng, F, et al. Utilizing distributed acoustic sensing and ocean bottom fiber optic cables for submarine structural characterization[J]. Scientific Reports, 2021, 11(1).\u003c/li\u003e\n\u003cli\u003eChen SY, et al.Photonic Integrated Sensing and Communication System Harnessing Submarine Fiber-Optic Cables for Coastal Event Monitoring[J]. APL Photon. 2020,5, 030901.\u003c/li\u003e\n\u003cli\u003eCalvert AJ. Ray tracing based prediction and subtraction of water-layer multiples [J]. Geophysics, 1990, 55: 443~445.\u003c/li\u003e\n\u003cli\u003eDou S, et al. Distributed Acoustic Sensing for Seismic Monitoring of The Near Surface: A TrafficNoise Interferometry Case Study[J]. Scientific Reports, 2017,7(1).\u003c/li\u003e\n\u003cli\u003eDong X, et al. Random and Coherent Noise Suppression in DAS-VSP Data by Using a Supervised Deep Learning Method[J]. IEEE Geoscience and Remote Sensing Letters, 2022, 19: p. 1-5.\u003c/li\u003e\n\u003cli\u003eF. Williams, et al. Distributed sensing of microseisms and teleseisms with submarine dark fibers[J]. Nat. Comm. vol. 10, 2019, Art. no. 5778.\u003c/li\u003e\n\u003cli\u003eFang G, et al. Urban Near‐Surface Seismic Monitoring Using Distributed Acoustic Sensing[J]. Geophysical Research Letters, 2020, 47(6).\u003c/li\u003e\n\u003cli\u003eFernandez-Ruiz M.R, et al. Seismic Monitoring With Distributed Acoustic Sensing From the Near-Surface to the Deep Oceans[J]. Journal of Lightwave Technology, 2022, 40(5): p. 1453-1463.\u003c/li\u003e\n\u003cli\u003eFern\u0026aacute;ndez-Ruiz M.R, et al. Distributed acoustic sensing for seismic activity monitoring[J]. APL Photonics, 2020, 5(3): p. 030901.\u003c/li\u003e\n\u003cli\u003eFabelinskii IL. Molecular scattering of light [M]. Springer Science \u0026amp; Business Media, 2012.\u003c/li\u003e\n\u003cli\u003eMarra, et al. Ultrastable laser interferometry for earthquake detection with terrestrial and submarine cables[J], Science, 2018, vol. 361, pp. 486-490, .\u003c/li\u003e\n\u003cli\u003eHuang N E, Shen Z, Long S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear nonstationary time series analysis[C]. Proceedings: Mathematical,Physical and Engineering Sciences. London The Royal Society Press, 1998, 454(1971):903-995.\u003c/li\u003e\n\u003cli\u003eJB Ajo-Franklin, N.J. T.C. Dawe and J.B. Ajo-Franklin. Illuminating seafloor faults and ocean dynamics with dark fiber distributed acoustic sensing[J]. Science (American Association for the Advancement of Science), 2019. 366(6469): p. 1103-1107.\u003c/li\u003e\n\u003cli\u003eJousset P, et al. Dynamic strain determination using fibre-optic cables allows imaging of seismological and structural features[J]. Nature Communications, 2018. 9(1).\u003c/li\u003e\n\u003cli\u003eLindsey N.J, et al. Fiber‐Optic Network Observations of Earthquake Wavefields[J]. Geophysical Research Letters, 2017. 44(23).\u003c/li\u003e\n\u003cli\u003eLindsey N.J, H. Rademacher and J.B. Ajo Franklin. On the Broadband Instrument Response of Fiber‐Optic DAS Arrays[J]. Journal of Geophysical Research: Solid Earth, 2020. 125(2).\u003c/li\u003e\n\u003cli\u003eLior I, et al. Imaging an underwater basin and its resonance modes using optical fiber distributed acoustic sensing[J]. Seismological research letters, 2022, 93(3): p. 1573-1584.\u003c/li\u003e\n\u003cli\u003eLior I, et al. On the Detection Capabilities of Underwater Distributed Acoustic Sensing. Journal of Geophysical Research: Solid Earth, 2021. 126(3).\u003c/li\u003e\n\u003cli\u003eLamare, P. The MEUST deep sea infrastructure in the Toulon site. EPJ Web Conf. 2016, 116, 09001.\u003c/li\u003e\n\u003cli\u003eLu X. Coherent Rayleigh time domain reflectometry:novel applications for optical fibre sensing[D]. EPFL Swiss Federal Institute of Technology, 2016.\u003c/li\u003e\n\u003cli\u003eLi X L, Li Z H, Wang E Y, et al. Analysis of natural mineral earthquake and blast based on Hilbert\u0026ndash;Huang transform (HHT). Science Direct[J]. Journal of Applied Geophysics,2016, 128: 79-86.\u003c/li\u003e\n\u003cli\u003eMartin, E.R. et al. A Seismic Shift in Scalable Acquisition Demands New Processing: Fiber-Optic Seismic Signal Retrieval in Urban Areas with Unsupervised Learning for Coherent Noise Removal[J]. IEEE Signal Processing Magazine, 2018. 35(2): p. 31-40.\u003c/li\u003e\n\u003cli\u003eMatsumoto, H, et al. Detection of hydroacoustic signals on a fiber-optic submarine cable[J]. Scientific Reports, 2021, 11(1).\u003c/li\u003e\n\u003cli\u003eMasoudi A, Newson TP. Contributed Review:Distributed optical fibre dynamic strain sensing[J]. Review of Scientific Instruments, 2016, 87: 011501.\u003c/li\u003e\n\u003cli\u003eMayne W H. Common-reflection-point horizontal data-stacking technique[J]. Geophysics, 1962,32,225-229.\u003c/li\u003e\n\u003cli\u003eNathaniel J. LindseyT. Craig DaweJonathan B. Ajo-Franklin. Illuminating seafloor faults and ocean dynamics with dark fiber distributed acoustic sensing[J]. Science, 2019, 366, 1103-1107.\u003c/li\u003e\n\u003cli\u003eOlshansky R. Propagation in glass optical wave guides[J]. Reviews of Modern Physics. 1979, 51(2): 341.\u003c/li\u003e\n\u003cli\u003eRoy S K, Molla M, Hirose K, et al. Pitch estimation of noisy speech signals using EMD fourier based hybrid algorithm[C]. IEEE International Symposium on Circuits \u0026amp; Systems.IEEE, 2010: 2658-2661.\u003c/li\u003e\n\u003cli\u003eRilling G, Flandrin P, Goncalves P. On empirical mode decomposition and its algorithms[J]. Journal of the Acoustical Society of America, 2003, 265(04): 785- 788.\u003c/li\u003e\n\u003cli\u003eSladen A, et al. Distributed sensing of earthquakes and ocean-solid Earth interactions on seafloor telecom cables[J]. Nature Communications, 2019. 10(1).\u003c/li\u003e\n\u003cli\u003eSatoshi Ide, E.A.H.M. Very broadband strain-rate measurements along a submarine ber-optic cable off Cape Muroto, Nankai subduction zone, Japan[J]. Earth, Planets and Space, 2021.\u003c/li\u003e\n\u003cli\u003eSchroeder J, Mohr R,Macedo PB. Rayleigh and Brillouin scattering in K2OSiO2 glasses [J] . Journal of the American Ceramic Society, 1973, 56(10): 510-514.\u003c/li\u003e\n\u003cli\u003eSchoenberger M. Optimum weighted stack for multiple suppression [J]. Geophysics, 1996, 61: 891-901.\u003c/li\u003e\n\u003cli\u003eTyapkin Yu K. Optimized estimates of a complicated model of the multichannel seismic record with statistic and deterministic regularization[J]. Russian Geology and Geophysics, 1994, 35(1): 109-115.\u003c/li\u003e\n\u003cli\u003eVerschuur DJ, Berkhout AJ, Wapenaarc PA. Adaptive surface related multiple elimination [J]. Geophysics, 1992, 57:1167-1177.\u003c/li\u003e\n\u003cli\u003eWalter F, et al. Distributed acoustic sensing of microseismic sources and wave propagation in glaciated terrain[J]. Nature Communications, 2020. 11(1).\u003c/li\u003e\n\u003cli\u003eWidess MB, et al. Quanfifying resolving power of seismic systems[J]. Geophysics, 1982, 47(8): 1160-1173.\u003c/li\u003e\n\u003cli\u003eWhite R E. The performance of optimum stacking filters in suppressing uncorrelated noise [J].Geophysical Prospecting,1977,25:165-178.\u003c/li\u003e\n\u003cli\u003eYu, C, Zhan, Z. Lindsey, N. J. Ajo‐ Franklin, J. B. \u0026amp; Robertson, M. The potential of DAS in teleseismic studies: Insights from the Goldstone experiment[J]. Geophysical Research Letters, 2019,46, 1320-1328.\u003c/li\u003e\n\u003cli\u003eZhao A B, Shen G N, Chen Y, et al. The application of HHT and neural network in feature extraction of ship targets[J]. Technical Acoustics, 2012, 31(03): 272-276.\u003c/li\u003e\n\u003cli\u003eZhan, Z. Distributed acoustic sensing turns fiber-optic cables into sensitive seismic antennas[J]. Seismological research letters, 2019. 91(1): p. 1-15.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-4811181/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4811181/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"As a new seismological monitoring technology, distributed acoustic sensing (DAS) not only provides a high-precision seismic observation method, but also uses undersea communication fiber or dark fiber to reduce the layout cost, and supports long-term observation for several years, and the sensing resolution reaches the scale of meters. This technique provides a broad prospect for seismic observation in oceanic area which is difficult to be covered by traditional seismic observation instruments.\nIn this paper, we analyzed the DAS waveforms of 10 local-earthquake events waveforms recorded by offshore submarine communication cable in the northern South China Sea, and discussed the response ability of DAS system based on shallow sea communication cable to near-earthquake. It is found that the signals are affected by surface gravitational waves and micro seismic noise (Scholte Wave) generated by solid-liquid coupling motion. Then, waveform stacking, discrete wavelet transform, and empirical mode decomposition are used to denoise the signal in shallow sea area for extracting earthquake waveform. Submarine optical cable is the most widely distributed submarine signal monitoring system in the world. Long-term and cheap deep-sea seismic wave observation based on such the large-scale global ocean observation network can provide important data for the development of seismology.","manuscriptTitle":"Local Earthquake Response on the Submarine Communication Cable in The Northern South China Sea","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-06 06:32:11","doi":"10.21203/rs.3.rs-4811181/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-09-20T04:26:18+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-17T06:33:00+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-15T08:14:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"81689979763289796829741609556255816785","date":"2024-09-11T19:18:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"2094951326532687442141002411919834422","date":"2024-08-29T10:50:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"223808217416825446542936936246712000869","date":"2024-08-26T06:36:49+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-08-21T13:48:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-08-20T01:02:45+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-08-13T14:12:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-08-10T05:37:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-07-27T05:32:46+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7624bbfb-e89b-4e81-9c0b-da5be8bd2f26","owner":[],"postedDate":"September 6th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":37131496,"name":"Earth and environmental sciences/Solid earth sciences/Seismology"},{"id":37131497,"name":"Physical sciences/Optics and photonics/Applied optics/Optical sensors"}],"tags":[],"updatedAt":"2025-04-21T16:06:23+00:00","versionOfRecord":{"articleIdentity":"rs-4811181","link":"https://doi.org/10.1038/s41598-025-93682-2","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-04-15 15:56:57","publishedOnDateReadable":"April 15th, 2025"},"versionCreatedAt":"2024-09-06 06:32:11","video":"","vorDoi":"10.1038/s41598-025-93682-2","vorDoiUrl":"https://doi.org/10.1038/s41598-025-93682-2","workflowStages":[]},"version":"v1","identity":"rs-4811181","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4811181","identity":"rs-4811181","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}