Dense Seismic Array Research Articles

Three-dimensional velocity structure of the MS 6.0 Luxian earthquake source region and adjacent areas based on a dense seismic array

Three-dimensional velocity structure of the MS 6.0 Luxian earthquake source region and adjacent areas based on a dense seismic array

High-Resolution 3D Shear-Wave velocity structure in xiong’an New Area, Beijing (China), revealed by short-period dense seismic array

High-Resolution 3D Shear-Wave velocity structure in xiong’an New Area, Beijing (China), revealed by short-period dense seismic array

Microseismicity and fault structure in the Daliangshan block within the Southeastern Tibetan Plateau

The Daliangshan block is located between the Tibetan Plateau and the South China block and has accommodated several M > 6.5 damaging earthquakes in the past ∼600 years, as well as intense tectonic deformation and complex fault structures. In this study, we analyze more than two years of continuous seismic data recorded by a recently deployed dense seismic array. We used a recently developed machine learning-based earthquake location workflow (ESPRH) to construct a high-precision earthquake catalog for the region and obtained 3539 earthquakes, which is approximately three times as many as the National Earthquake Data Center (NEDC) catalog contains. The seismicity distribution not only confirms the nature of the faults marked on the map but also delineates the detailed geometry of the unmapped faults, including the en échelon faults at the northern end of the Zemuhe Fault and the “V”-shaped conjugate fault within the Mabian Fault Zone. The Zemuhe Fault and Lianfeng Fault are prone to hosting large earthquakes according to the derived low b-value. The western side of the Daliangshan block is dominated by strike-slip faults. Combining the fault geometry presented in this paper, we observed that the fault properties on the eastern side are complex. This tectonic phenomenon is attributed to the fact that during the lateral extrusion of the southeastern edge of the Chuandian fragments, the northeastern part of the Daliangshan block was squeezed by the South China block more strongly than its southwestern part. We provide the first precise earthquake catalog for Daliangshan block, which can be used as important seismological data for regional hazard assessment and research on the southeastern (SE) margin of the Tibetan Plateau.

Seismic mapping of the central and southern segments of the Tanlu fault zone using P-wave receiver functions

The central and southern segments of the Tanlu fault zone, located in the collision boundary between the Yangtze Plate and the North China Craton, underwent complex tectonic deformation associated with the Pacific Plate subduction. The crustal thickness and Vp/Vs ratio are important parameters for comprehending tectonic evolution and geodynamic processes. By integrating a newly dense seismic array, our results based on P-wave receiver function analyses reveal the crust thickness varies significantly across the Tanlu fault zone. The Yangtze Plate is characterized by a thinner crust, contrasting a thicker crust imaged in the North China Craton, indicating the Tanlu fault zone is a prominent structural block boundary. The Vp/Vs ratio within the Tanlu fault zone is visibly higher than its surroundings, probably correlating with abundant fluids activity along the fault zone, as a result of the subduction of the Pacific Plate. Additionally, we observe a gradually deepening Moho discontinuity beneath the Sulu orogenic belt with a low-angle dip, presenting direct seismic evidence for supporting the ancient Yangtze Plate underthrusting beneath the Sulu orogenic belt.

Shallow crustal structure of the northern Longmen Shan fault zone revealed by a dense seismic array with ambient noise analysis

The Longmen Shan (LMS) fault zone located in the southwest of China is not only the boundary tectonic zone of the Tibetan Plateau and South China block, but also an important part of the north–south seismic belt of China. To investigate the detailed crustal structure of the LMS fault zone, we deployed a dense seismic array of 151 short-period temporary stations in its northern section from March 16 to April 6, 2023. Continuous vertical component ambient noise waveforms recorded by these stations were cross-correlated for all station pairs, enabling the extraction of Rayleigh wave phase velocity dispersion curves across periods ranging from 0.5 s to 6 s. We then inverted these data to derive the crustal shear wave velocity using direct surface wave tomography. Meanwhile, a linear sub-array consisting of 54 stations and crossing the surface rupture of the 2008 Wenchuan earthquake was used to image the fault structure by applying the Transmitted Surface Wave Reverse Time Migration method (TSW-RTM). The ambient noise tomography results show that strong lateral heterogeneity exists in the shallow crust in the study area, with a high shear velocity in the northwest and a low shear velocity in the southeast. The low shear velocity is generally distributed between the Yingxiu-Beichuan fault (YBF) and the Guanxian-Jiangyou fault (GJF). The results of TSW-RTM indicate that the Wenchuan-Maoxian fault (WMF) and the YBF are both characterized with a steep dip to the northwest. Our results could serve as the important basis for the study of segmentation characteristics of the LMS fault and seismic hazard preparation and mitigation.

Imaging microearthquake rupture processes using a dense array in Oklahoma

Both large and small earthquakes rupture in complex ways. However, microearthquakes are often simplified as point sources and their rupture properties are challenging to resolve. We leverage seismic wavefields recorded by a dense array in Oklahoma to image microearthquake rupture processes. We construct machine-learning enabled catalogs and identify four spatially disconnected seismic clusters. These clusters likely delineate near-vertical strike-slip faults. We develop a new approach to use the maximum absolute SH-wave amplitude distributions (S-wave wavefields) to compare microearthquake rupture processes. We focus on one cluster with earthquakes located beneath the dense array and have a local magnitude range of -1.3 to 2.3. The S-wave wavefields of single earthquakes are generally coherent but differ slightly between the low-frequency (<12 Hz) and high-frequency (>12 Hz) bands. The S-wave wavefields are coherent between different earthquakes at low frequencies with average correlation coefficients greater than 0.95. However, the wavefield coherence decreases with increasing frequency for different earthquakes. This reduced coherence is likely due to the rupture differences among individual earthquakes. Our results suggest that earthquake slip of the microearthquakes dominates the radiated S-wave wavefields at higher frequencies. Our method suggests a new direction in resolving small earthquake source attributes using dense seismic arrays without assuming a rupture model.

Fluid‐Driven Seismicity in the Baihetan Reservoir Area Revealed by 3D Seismic Tomography Based on Dense Seismic Arrays

AbstractAfter impounding the Baihetan Water Reservoir in a historically seismically active area, numerous earthquakes occurred, but it is uncertain if they are connected to the impoundment. Based on the dense seismic network, this study used local earthquake tomography to construct high‐resolution VP (P‐wave velocity) and VP/VS models (P‐ to S‐wave velocity ratio) and update earthquake locations after the impoundment of Baihetan reservoir. Our study revealed that the reservoir water spreads from the dam site to the Qiaojia Basin and the Heishuihe branch, and pore pressure diffusion along faults induces many earthquakes. Reservoir water migrates through hidden faults and fractures beneath Hulukou, spreading to both sides and saturating some rocks below 7 km beneath the dam site‐Lianhuatang section, causing multiple magnitude 3.5+ earthquakes. This study reveals that reservoir water migration drives earthquakes in the reservoir area, offering new insights into seismogenesis following the Baihetan reservoir's impoundment, potentially applicable to understanding reservoir‐induced earthquakes in other reservoirs.

High‐Resolution Seismicity and Ground Motion Variability Across the Highly Locked Southern Anninghe Fault With Dense Seismic Arrays and Machine Learning Techniques

AbstractFault activity and structure are important factors for the assessment of seismic hazards. The Anninghe fault is one of the most active strike‐slip faults in southwestern China but has been experiencing seismic quiescence for M > 4 earthquakes since the 1970s. To understand better the characteristics of its highly locked southern segment, we investigate seismicity and ground motion variability using recently deployed multi‐scale dense arrays. Assisted by machine learning (ML) seismic phase picking and event discrimination models, we first compile a high‐resolution catalog of local seismic events. We find limited earthquakes that occurred on the Anninghe fault, consistent with its generally acknowledged high locking degree. Whereas, most newly detected events appear within off‐fault clusters, among which four are closely related to anthropogenic activities (e.g., mining blasts), and two neighboring faults host the remaining ones. We further apply an ML‐based first‐motion polarity (FMP) classifier and successfully obtain a reliable small earthquake focal mechanism, which agrees well with the geologically inferred north‐south trending and eastward dipping of the Anninghe fault. Analyses of ground motion variations along two across‐fault linear arrays show abrupt changes in FMPs and obvious frequency‐dependent site amplifications near the mapped fault traces. It further suggests that, at finer scales, the damaged Anninghe fault zone may have split into two smaller damaged zones at shallower depths, resulting in a typical “flower‐type” fault structure. The efficient workflow developed in this study can be well applied for the longer‐term monitoring and better characterization of the southern Anninghe fault, or other similar regions.

EQGraphNet: Advancing single-station earthquake magnitude estimation via deep graph networks with residual connections

Magnitude estimation is a critical task in seismology, and conventional methods usually require dense seismic station arrays to provide data with sufficient spatiotemporal distribution. In this context, we propose the Earthquake Graph Network (EQGraphNet) to enhance the performance of single-station magnitude estimation. The backbone of the proposed model consists of eleven convolutional neural network layers and ten RCGL modules, where a RCGL combines a Residual Connection and a Graph convolutional Layer capable of mitigating the over-smoothing problem and simultaneously extracting temporal features of seismic signals. Our work uses the STanford EArthquake Dataset for model training and performance testing. Compared with three existing deep learning models, EQGraphNet demonstrates improved accuracy for both local magnitude and duration magnitude scales. To evaluate the robustness, we add natural background noise to the model input and find that EQGraphNet achieves the best results, particularly for signals with lower signal-to-noise ratios. Additionally, by replacing various network components and comparing their estimation performances, we illustrate the contribution of each part of EQGraphNet, validating the rationality of our approach. We also demonstrate the generalization capability of our model across different earthquakes occurring environments, achieving mean errors of ±0.1 units. Furthermore, by demonstrating the effectiveness of deeper architectures, this work encourages further exploration of deeper GNN models for both multi-station and single-station magnitude estimation.

Deep structure of the Wulong goldfield, Liaodong Peninsula, China, revealed by receiver functions: implications for the tectonic and mineralization dynamics

During the Mesozoic, the North China Craton experienced intense tectonic movements that resulted in the formation of numerous gold deposits on the Liaodong and Jiaodong Peninsulas in northeastern China. To investigate the relationship between deep crustal structure and gold mineralization in the Liaodong Peninsula, we deployed 334 dense seismic stations in the Wulong goldfield (WLGF) with the idea of analysing numerous receiver functions at different array stations. The purpose focused on knowing the potential for gold mineralization in the area. The study revealed the following: (1) The WLGF is characterized by a crustal thickness of approximately 32 km and an average Vp/Vs ratio of 1.76. The high value of the Vp/Vs ratio near the Wulong gold deposit suggests that mantle materials have penetrated into the crust and contributed to the mineralization process. (2) A low-velocity layer located at a depth of 10–18 km below the WLGF seems to support the existence of a potentially brittle-ductile transition zone. Also, hydrothermal magma upwelling channels are observed in the upper crust beneath the Wulong gold deposit. (3) The presence of a discontinuous low-velocity layer in the middle crust beneath the Liaodong Peninsula suggests promising prospects for gold ore exploration. The receiver functions method based on a dense seismic array employed in this study can offer valuable references and guidance for the fine exploration and research of ore deposits in other regions globally.

Evidence for faulting and fluid-driven earthquake processes from seismic attenuation variations beneath metropolitan Los Angeles

Seismicity in the Los Angeles metropolitan area has been primarily attributed to the regional stress loading. Below the urban areas, earthquake sequences have occurred over time showing migration off the faults and providing evidence that secondary processes may be involved in their evolution. Combining high-frequency seismic attenuation with other geophysical observations is a powerful tool for understanding which Earth properties distinguish regions with ongoing seismicity. We develop the first high-resolution 3D seismic attenuation models across the region east of downtown Los Angeles using 5,600 three-component seismograms from local earthquakes recorded by a dense seismic array. We present frequency-dependent peak delay and coda-attenuation tomography as proxies for seismic scattering and absorption, respectively. The scattering models show high sensitivity to the seismicity along some of the major faults, such as the Cucamonga fault and the San Jacinto fault zone, while a channel of low scattering in the basement extends from near the San Andreas fault westward. In the vicinity of the Fontana seismic sequence, high absorption, low scattering, and seismicity migration across a fault network suggest fluid-driven processes. Our attenuation and fault network imaging characterize near-fault zones and rock-fluid properties beneath the study area for future improvements in seismic hazard evaluation.

Insight into the Evolution of the Eastern Margin of the Wyoming Craton from Complex, Laterally Variable Shear Wave Splitting

Abstract Dense seismic arrays such as EarthScope’s Transportable Array (TA) have enabled high-resolution seismic observations that show the structure of cratonic lithosphere is more heterogeneous and complex than previously assumed. In this study, we pair TA data with data from the Bighorn Arch Seismic Experiment and the Crust and lithosphere Investigation of the Easternmost expression of the Laramide Orogeny (CIELO) to provide unprecedented detail on the seismic anisotropic structure of the eastern margin of the Wyoming Craton, where several orogens emerged from nominally strong cratonic lithosphere during the Laramide Orogeny. In this study, we use the splitting of teleseismic shear waves to characterize fabrics associated with deformation in the Earth’s crust and mantle. We constrain distinct anisotropic domains in the study area, and forward modeling shows that each of these domains can be explained by a single layer of anisotropy. Most significantly, we find a fast direction in the southern part of the Powder River Basin, which we refer to as the Thunder Basin Block (TBB), that deviates from absolute plate motion (APM). This change in splitting behavior coincides with changes in other modeled geophysical observations, such as active source P-wave velocity models, potential field modeling, and seismic attenuation analysis, which all show a significant change moving from the Bighorn Mountains to the TBB. We argue that these results correspond to structure predating the Laramide Orogeny, and most likely indicate a Neoarchean boundary preserved within the lithosphere.

Integration of Machine learning and equal differential time method for enhanced hypocenter localization in earthquake early warning systems: application to dense seismic arrays in Taiwan

The Earthquake Early Warning System (EEWS) acts as a vital instrument for reducing seismic risks in regions with high seismic vulnerability. A rapid and accurate hypocenter estimation is pivotal for the EEWS, providing the groundwork for more reliable magnitude and intensity assessments necessary for effective earthquake warnings. This study presents an algorithm that integrates machine-learning-based (near) real-time phase picking with an Equal Differential Time (EDT) rapid hypocenter location algorithm, applying it to a 3D velocity model. The phase-picking model, refined through data augmentation, enhances the precision of phase detection in continuous recordings and simultaneous multiple events while ensuring the swift detection of the P-phase, which is critical for early earthquake warnings. Our rapid earthquake location method calculates theoretical P arrivals from potential hypocenters, which are grid points in a 3D velocity model, to stations that are close to their grid points, with the arrivals being stored by the station. As P arrivals are detected, the differences in arrival times across stations are utilized in EDT for estimating hypocenters. Furthermore, our earthquake location algorithm is adept at localizing multiple seismic events, a capability that can diminish the risk of unreported cases in scenarios where events occur in close temporal and spatial succession in high seismicity regions. We applied the algorithm to real waveform recordings of recent earthquakes in Taiwan that satisfied the early warning criteria. The results suggest that our algorithm consistently yields more reliable hypocenter estimates compared to those from the currently operational EEWS in Taiwan. Moreover, our algorithm succeeded in locating an earthquake that the current EEWS overlooked due to its failure to recognize P arrivals. These results showcase the potential of our algorithm to provide more accurate hypocenter estimates and to locate earthquake events with complex seismic recordings.Graphical

Double-Difference Tomography with a Deep Learning–Based Phase Arrival Weighting Scheme and Its Application to the Anninghe–Xiaojiang Fault Zone

Abstract High-precision seismic phase arrivals are a prerequisite for building reliable velocity models with travel-time tomography. There has recently been a growing use of seismic phase arrival data obtained through deep learning techniques in travel-time tomography research. Nevertheless, a significant challenge that has emerged pertains to the assessment of the quality of these automatic arrivals. In this article, we used PhaseNet, a deep learning method, to automatically detect the arrival times of the P wave and S wave of 3086 seismic events recorded by dense seismic arrays, obtaining 87,553 high-quality arrivals. To evaluate the quality of the arrival times subsequently used for travel-time tomography inversion, we applied a weighting scheme that includes both detection probability value and signal-to-noise ratio. This new weighting scheme can effectively reduce the overall travel-time residual by 7%. The weighted data were then used in the double-difference tomography method to invert for the crustal velocity structure of the Anninghe–Xiaojiang fault zone. The resulting new model exhibits a lateral resolution of up to 0.25° and reveals velocity anomalies that exhibit a strong correlation with major geological features and block boundaries. Notably, the presence of low-VP and low-VS in the middle crust of the Ludian–Qiaojia seismic zone suggests the existence of hot and weak felsic rocks, as well as possible fluid presence beneath the seismogenic layer of this area. This study not only validates the practicality of using deep learning-based phase picking arrivals in travel-time tomography but also proposes a new weighting scheme to refine the tomographic velocity models.

Imaging the Garlock Fault Zone With a Fiber: A Limited Damage Zone and Hidden Bimaterial Contrast

AbstractThe structure of fault zones and the ruptures they host are inextricably linked. Fault zones are narrow, which has made imaging their structure at seismogenic depths a persistent problem. Fiber‐optic seismology allows for low‐maintenance, long‐term deployments of dense seismic arrays, which present new opportunities to address this problem. We use a fiber array that crosses the Garlock Fault to explore its structure. With a multifaceted imaging approach, we peel back the shallow structure around the fault to see how the fault changes with depth in the crust. We first generate a shallow velocity model across the fault with a joint inversion of active source and ambient noise data. Subsequently, we investigate the fault at deeper depths using travel‐time observations from local earthquakes. By comparing the shallow velocity model and the earthquake travel‐time observations, we find that the fault's low‐velocity zone below the top few hundred meters is at most unexpectedly narrow, potentially indicating fault zone healing. Using differential travel‐time measurements from earthquake pairs, we resolve a sharp bimaterial contrast at depth that suggests preferred westward rupture directivity.

Detection of the low-velocity layer using a convolutional neural network on passive surface-wave data: An application in Hangzhou, China

Passive surface-wave methods using dense seismic arrays have gained growing attention in near-surface high-resolution imaging in urban environments. Deep learning (DL) in the extraction of dispersion curves and inversion can release a tremendous workload brought by dense seismic arrays. We presented a case study of imaging shear-wave velocity (Vs) structure and detecting low-velocity layer (LVL) in the Hangzhou urban area (eastern China). We used traffic-induced passive surface-wave data recorded by dense linear arrays. We extracted phase-velocity dispersion curves from noise recordings using seismic interferometry and multichannel analysis of surface waves. We adopted a convolutional neural network to estimate near-surface Vs models by inverting Rayleigh-wave fundamental-mode phase velocities. To improve the accuracy of the inversion, we utilized the sensitivities to weight the loss function. The average root mean square error from the weighted inversion is 46% lower than that from the unweighted DL inversion. The estimated pseudo-2D Vs profiles correspond to the velocities obtained from downhole seismic measurements. Compared with an investigation on the same survey area, our inversion results are more consistent with the Vs provided by downhole seismic measurements within 50–60 m where the LVL exists. The trained neural network successfully identified that the LVL is located at 50–60 m deep. To check the applicability of the trained neural network, we applied it to a nearby passive surface-wave survey and the inversion results agree with the existing investigation results. The two applications demonstrate the accuracy and efficiency of delineating near-surface Vs structures with the LVL from traffic-induced noise using the DL technique. The DL inversion has great potential for monitoring subsurface medium changes in urban areas.

Seismic and geological evidence of hidden faults in the Yinpan Reservoir area based on a dense seismic array

Seismic and geological evidence of hidden faults in the Yinpan Reservoir area based on a dense seismic array

Fine Shear-Wave Velocity Structures of Subsurface beneath the Guangdong–Hong Kong–Macao Greater Bay Area with Dense Seismic Array and SPAC Method

Abstract We apply the spatial autocorrelation (SPAC) method to construct the 3D subsurface shear-wave velocity structure model using the short-period dense seismic array (containing 725 nodal geophones) located at the Guangdong–Hong Kong–Macao Greater Bay area (GBA). We first divided the dense array into numerous subarrays, with each subarray consisting of nine nodal geophones, and obtained 562 subarrays that can provide 1D VS profiles of the same quantity. Then, the SPAC method and genetic algorithm are utilized to extract the dispersion curve of the Rayleigh wave from the raw microtremor data and invert VS structure, respectively. Finally, a 3D VS structure model from the surface to 3.3 km depth is derived by combining all 1D VS structures. Relatively low-velocity anomalies above 700 m are considered unconsolidated shallow sediments as well as relatively high-velocity anomalies beneath 1100 m are attributed to consolidated granite bedrock. Meanwhile, low-velocity anomalies that are identified through the vertical VS profile at a depth of about 900–3000 m can be contributed to the fractured zone, and striped low-velocity anomalies in the horizontal VS maps reveal the location of the deeply buried faults in the study area. The results also mean that the SPAC method combined with the records of short-period dense seismic array can be effectively applied to image subsurface structures in high-populated urban area. The development of this noise-resistance and environment-friendly geophysical technique provides a reliable and effective way to explore the complicated subsurface geological structures, which is of great significance to urban engineering construction and earthquake disaster reduction work in densely populated urban agglomerations.

Development of improved short-period geophone: Implementation of low-frequency compensation

Short-period geophones are widely used instruments in dense seismic array investigations. However, inherent limitations in capturing low-frequency signals restrict their versatility. This study implements a low-frequency compensation algorithm via a state-variable filtering circuit to enhance the frequency response of a traditional short-period geophone. Through simulation experiments and vibration testing, we validated the performance of the proposed bandwidth-extension (BE) system. Additionally, time–frequency earthquake signal analysis compared outcomes between our BE seismometers, benchmark short-period seismometers, and high-fidelity broadband seismometers across three seismic events with epicentral distances up to 5,186 km. Results demonstrate the BE system successfully extends the natural frequency cutoff from 4.5 Hz to below 0.2 Hz while maintaining strong consistency with broadband counterparts. Moreover, negligible power demands (10 mW) facilitate embedding the BE circuit directly within acquisition hardware. Given these advantages, the BE method shows promise for upgrading future sensors of seismic networks to record low-frequency content with minimal costs.

Long Period Rayleigh Wave Focal Spot Imaging Applied to USArray Data

AbstractWe demonstrate the effectiveness of seismic dense array surface wave focal spot imaging using USArray data from the western‐central United States. We study dispersion in the 60–310 s period range and assess the image quality of fundamental mode Rayleigh wave phase velocity maps. We apply isotropic spatial autocorrelation models to the time domain zero lag noise correlation wavefield data at distances of about one wavelength. Local estimates of the phase velocity, its uncertainty, and the regression quality imply overall better ZZ relative to ZR or RZ results. The extension of the depth resolution compared to passive surface wave tomography is demonstrated by the inversion of three clustered dispersion curves from different tectonic units. We observe anisotropic surface wave energy flux and the influence of body wave energy, but sensitivity tests at 60 s targeting the data range, correlation component, and processing choices show that the ZZ focal spots yield consistent high‐quality images compared to regional tomography results in the 60–150 s period range. In contrast, at 200–300 s the comparatively small scales of the imaged structures and the imperfect agreement with low‐resolution global tomography results highlight the persistent challenge to reconcile imaging results based on different data sources, theories, and techniques. Our study shows that surface wave focal spot imaging is an accurate, robust, local imaging approach. Better control over clean autocorrelation fields can further improve applications of this seismic imaging tool for increased resolution of the elastic structure below dense seismic arrays.