The detection of hydrothermal plumes is crucial for discovering active seafloor hydrothermal fields and exploring seafloor resources. To improve the efficiency of autonomous underwater vehicles (AUVs) in detecting hydrothermal plumes, this study proposes a comprehensive online identification algorithm that employs deep learning models based on the historical dataset collected by the Qianlong-II AUV at the Southwest Indian Ridge (SWIR). The algorithm integrates sliding window technique, the gated recurrent unit network, and the multi-layer perceptron network to achieve real-time detection of hydrothermal anomaly data, including temperature, turbidity, methane concentration, and oxidation-reduction potential (ORP), and further classifies the fluid types of hydrothermal plumes. To train the deep learning models, point anomaly data (temperature and turbidity) are labeled via the K-medoids clustering algorithm, while subsequence anomaly data (methane and ORP) are labeled using the adaptive iterative algorithm. Experimental results demonstrate that the algorithm achieves accuracies exceeding 98.1% for anomaly detection and 98.2% for fluid types identification. Meanwhile, the ablation experiment confirmed that the anomaly detection module significantly improved the overall identification performance. The proposed algorithm provides an effective and reliable solution for autonomous online identification of hydrothermal plumes, with broad application potential in deep-sea exploration.

Microfibers are recognized as the most prevalent form of microplastics, with a widespread distribution across various ecosystems. The distinctive morphology of microfibers provides additional insights into their detection. In contrast to most studies targeting microplastic identification, this study proposed a rapid detection method for microfibers utilizing Raman spectroscopy and machine learning. A Raman spectral data set comprising 15 types of plastic fibers was created, and an autoencoder (AE) model was employed to process the data, effectively reducing spectral interferences such as noise and fluorescence through spectral reconstruction. Four machine learning models, including support vector machine (SVM), random forest (RF), and two convolutional neural networks (CNN, one-dimensional CNN1D, and two-dimensional CNN2D), were developed and evaluated by using the reconstructed spectra. CNN models demonstrated superior performance, with CNN1D achieving the highest accuracy of 99.03%. When applied to microfiber samples from real-world environments, the CNN1D-based method achieved an overall classification accuracy of 85.71%, achieving perfect identification of polyethylene (PE) and polyethylene terephthalate (PET). Finally, AE and CNN1D were applied to surface water samples from the East China Sea to validate the proposed microfiber identification approach in real-world scenarios. A total of 775 environmental microfibers were efficiently classified, achieving high-throughput classification with an average abundance of 2.92 ± 2.30 items/L, and the primary material types included cotton and polyester. Through the method in this study, the time needed to complete spectral processing and matching for all microfiber samples was shortened to less than 5 min. This study established a rapid detection framework for microfibers, addressing a critical gap in environmental monitoring. Importantly, the integration of morphology in our analytical pipeline shows potential for the future development of source tracking systems, which could significantly contribute to microfiber pollution control strategies.

Although anthropogenic disturbances on the landscape and ecology of individual oceanic islands over the past millennia have been documented from archaeological and paleo-ecological archives, the quantitative evaluation on a global scale allows for the exploration of more dimensions of anthropogenic impacts on islands. Here, we analyze sedimentation rates and charcoal concentrations integrated from 45 islands worldwide to evaluate soil erosion and fire regimes associated with human pre-contact, initial-colonization, and post-settlement stages. Compared to the relatively low and constant pre-contact background values, both parameters experienced unprecedented changes during the initial-colonization stage. The increasing rates are primarily driven by human activities, including large-scale fire-induced deforestation and intensive agricultural practices. In the post-settlement stage, sedimentation rates and fire occurrences either decreased or increased, but were still higher than their background levels. A comparable level and pattern of human disturbance during the initial-colonization stage is evident across the Atlantic, Indian, and Pacific Oceans. The Polynesian Triangle, the last main migration frontier in the Pacific Ocean after 1000 CE, however, experienced the most intensive anthropogenic influence. The study underscores that, apart from the Industrial Revolution, the initial-colonization stage of similar to 300 years had the most profound impact. Following the post-colonization stage, these ecosystems transitioned to a new, more dynamic state, rather than reverting to pre-human conditions.

SS precursors are crucial for probing mantle discontinuities and thermal structure but constructing high‐quality SS datasets—essential for precursor analysis—remains labor‐intensive due to limited automated solutions. While deep learning has advanced P‐ and S‐wave analysis, its application to systematic SS phase selection remains underexplored. We present PhaseSelectNet (PSNet), a deep neural network integrating Time‐Net and Polarity‐Net to facilitate data set construction. Polarity‐Net—the first automated tool for determining SS polarities—ensures reliable stacked SS precursors by preventing polarity misjudgments, which can distort waveforms. Trained on a global data set composed of high‐quality and low‐quality SS waveforms, Time‐Net and Polarity‐Net both demonstrate the capability to accurately identify high‐quality SS seismograms, while respectively provide precise polarities and arrival times. Our application of PSNet to the Iceland region—where a mantle plume interacts with the Mid‐Atlantic Ridge (MAR)—has revealed mantle discontinuity topography extending beyond previous observational limits constrained by seismic station coverage. The results identify the thinnest portion of the mantle transition zone (MTZ) southeast of Iceland, suggesting this region as the probable upwelling center at MTZ depths. This finding provides new insights into Icelandic key tectonic characteristic: the systematic southeastward migration of its rift system over the past 24 Myr, which has produced a distinct curvature in the otherwise linear trajectory of the North Atlantic Ridge. The continuous flow of high‐temperature magma, originating from a southeastern mantle upwelling along the base of the lithosphere, triggered thermomechanical weakening preferentially southeast of existing rift structures, systematically shifting the rift axis over time.

Determining the mechanisms responsible for intraplate volcanism - such as slab devolatilization melting versus active mantle plumes - remains a challenge. The greater South China Sea (SCS) region has experienced extensive intraplate Cenozoic volcanism across areas including Hainan, Southeast Indochina, northern Borneo, the northern SCS, and the post-spreading SCS basin. The prevalence of volcanism distributed widely across this region prompts fundamental questions about the key geodynamic processes driving such diverse magmatic activities. In this study, we elucidate the mantle transition zone (MTZ) discontinuities in this region using SS precursors, which helps to overcome the sparse seismic coverage due to its predominantly oceanic setting. We collected over 16,000 high-quality seismograms that sample the upper mantle and MTZ beneath this region from global earthquakes and stations. After correcting for the effects of shallow crustal variations and upper mantle heterogeneity on traveltimes of SS phases and their precursors, we unveil lateral variations in the MTZ boundaries (d410 and d660) and intricate features of the mid-MTZ reflectors (S520S). Significant MTZ thinning and normal S520S waveforms beneath Hainan provide compelling evidence for mantle upwelling through the MTZ. Conversely, the evident splitting of S520S beneath the northern SCS, Southeast Indochina, and northern Borneo, all characterized by stagnant subducted slabs, indicates that the volcanism in these regions likely originated from a mechanism distinct from the active upwelling beneath Hainan. Dehydration melting attributed to devolatilizing stagnant slabs in the MTZ is a potential cause for Cenozoic volcanism in these regions. Determining the causes of volcanic activity far from tectonic plate boundaries remains a challenge. The greater South China Sea region experienced widespread volcanism during the Cenozoic era, prompting questions about the driving geological processes behind this phenomenon. This study investigated the boundaries of the mantle transition zone (MTZ) beneath the region using seismic wave analysis of over 16,000 seismic recordings. By analyzing the travel times of signals reflected from the MTZ boundaries and internal reflectors, lateral variations in the structure of the MTZ boundaries and internal features were revealed. Beneath Hainan, thinning of the MTZ and normal waveforms of the internal MTZ reflector suggest upwelling of hot mantle material through the MTZ, likely causing the surface volcanism. In contrast, beneath the northern South China Sea, Southeast Indochina, and northern Borneo, the distorted waveforms of signals from the internal MTZ reflector indicate a different mechanism. These latter regions contain ancient, subducted slabs present in their MTZs. Dehydration melting, caused by the release of water from these stagnant slabs within the MTZ, is a potential cause of the Cenozoic volcanism in these areas, distinct from the mantle upwelling observed beneath Hainan. SS precursors from the mantle transition zone (MTZ), notably the S520S, provide insights into mantle dynamics beneath the greater South China Sea (SCS) region Evidence of MTZ thinning and normal S520S waveform beneath Hainan suggests active mantle upwelling through the MTZ The S520S splitting beneath areas with stagnant subducted slabs indicates a possible magmatic origin through flux melting

利用国家数字地震台网的 137 个固定台站以及 332 个 ChinArray 流动台站的数据,基于背景噪声和远震面波成像方法,共同约束瑞利面波相速度,并通过非线性方法(马尔科夫链蒙特卡洛方法)反演青藏高原东南缘的地壳三维剪切波速度结构.反演结果表明,青藏高原东南缘存在大范围连通的下地壳流,表现为下地壳存在连通的波速小于 3.55 km/s 的近水平的剪切波低速区,且与地表地形有很好的对应关系.推断青藏高原东南缘存在 3 支连通的地壳流,第一支位于攀枝花一带以西,第二支位于攀枝花一带以东,第三支位于攀枝花一带的下地壳.来自青藏高原的下地壳流受到四川块体坚硬下地壳的阻挡而转向,通过峨眉山一带向南流动,与第三支通过峨眉山大火成岩省(ELIP)内带(攀枝花一带)向南流动的下地壳流一起,改造了峨眉山大火成岩省中带南部的下地壳,并使地壳增厚.推断青藏高原东南缘连通的下地壳流的南端目前大约在北纬24°附近,并将随着时间的推移向南迁移,即通过下地壳流导致该地区地壳增厚的范围也随时间的推移向南扩展.

We develop a novel approach for multi-frequency, elliptical-anisotropic eikonal tomography based on physics-informed neural networks (pinnEAET). This approach simultaneously estimates the medium properties controlling anisotropic Rayleigh waves and reconstructs the traveltimes. The physics constraints built into pinnEAET's neural network enable high-resolution results with limited inputs by inferring physically plausible models between data points. Even with a single source, pinnEAET can achieve stable convergence on key features where traditional methods lack resolution. We apply pinnEAET to ambient noise data from a dense seismic array (ChinArray-Himalaya II) in the northeastern Tibetan Plateau with only 20 quasi-randomly distributed stations as sources. Anisotropic phase velocity maps for Rayleigh waves in the period range from 10-40 s are obtained by training on observed traveltimes. Despite using only about 3% of the total stations as sources, our results show low uncertainties, good resolution and are consistent with results from conventional tomography. Anisotropy refers to the directional dependence of seismic wave velocities, which can arise from a variety of factors such as crystal alignment, stress fields, or fluid-filled cracks. Elliptical-anisotropic eikonal tomography is a variant of eikonal tomography that can be used to estimate medium properties and reconstructed traveltimes from ambient noise data. In this study, we propose a new algorithm to implement multi-frequency, elliptical-anisotropic eikonal tomography based on physics-informed neural networks (pinnEAET), which combine data-driven models with theory-based models that include physics constraints on the system. We apply this architecture to data from a dense seismic array deployed on the northeastern Tibetan Plateau. Our results can achieve at least the same resolution as traditional methods while requiring less traveltime data. This strategy can provide new insights into the seismic imaging in case of limited or noisy data. We present a physics-informed deep learning eikonal tomography method for anisotropic velocity modelingThe algorithm incorporates wave physics to simultaneously process multi-frequency data, ensuring reliable tomographic modelsWe successfully recover the anisotropic velocity structure of the northeastern Tibet using less data than in traditional models

<p>Machine learning is rapidly becoming ubiquitous in the Earth Sciences promising to provide scalable algorithms for data-mining, interpretation, and model building. Initially heralded for its ability to exclude complicated physics from data analysis, recent innovations seek to merge machine learning  solutions with conventional physics-based methods in order to enhance their capability</p> <p>We present a novel eikonal tomography approach for Rayleigh wave phase velocity and azimuthal anisotropy based on the elliptical-anisotropic eikonal equation, by formulating the tomography problem as the training of a physics informed neural network (PINN). The PINN eikonal tomography (pinnET) neural network utilizes deep neural networks as universal function approximators and extracts traveltimes and medium properties during the optimization process. Whereas classical eikonal tomography uses a generic non-physics-based interpolation and regularization step to reconstruct traveltime surfaces, optimizing the network parameters in pinnET means solving a physics constrained traveltime surface reconstruction inversion, tackling measurement noise and resolving the underlying velocities that govern the physics. The fast and slow velocity and the anisotropic direction information can be directly evaluated from the trained medium property networks. Checkerboard tests indicate that the input velocity model can be well recovered by using this approach and synthetic data.</p> <p>We demonstrate this approach by applying it to multi-frequency surface wave data from ChinArray phase II sampling the north-eastern Tibetan plateau. We are able to use much less data to achieve similar subsurface images because of the benefit of including the physics constraint while reconstructing the traveltime surfaces. We are able to obtain excellent results using only 10 sources.  Comparing results from pinnET with conventional eikonal tomography, we find good agreement with distinct low velocity structures beneath the Songpang-Ganzi block, Qilian and Western Qinling Orogen. Large phase velocity uncertainties occur in a small part of the southeastern Ordos Block, the western Songpan-Ganzi Block and the eastern Sichuan basin, which correspond to the reduced data coverage dependent on the selection of the 10 sources. We also verify the accuracy and reliability of the pinnET by choosing only one station as virtual source, the retrieved velocities show relatively good resolution which is much better than in conventional eikonal tomography using similar sized datasets. The method is memory efficient because compressing the traveltimes as outputs to a NN is a concept akin to compressed sensing and offers advantages over traditional anisotropic eikonal tomography or neural network approaches.</p>

Double beamforming tomography (abbreviated as DBF) is a new method to image underground velocity structure using seismic ambient noise records. DBF can improve the signalto-noise ratio by superposing the coherent energy of ambient noise field, which is helpful to identify and extract different types of seismic waves and their incident azimuth (weak body wave, reflected surface wave, etc.), so as to conduct imaging research on underground velocity structure. Particularly for ambient noise surface wave imaging ,DBF does not require tomographic inversion, as it ensures directly finding the appropriate phase velocities from local slowness (azimuthal) searching. With a dense seismic nodal array, we image the shallow structure of Hualong fault in Panyu, Guangzhou based on ambient noise records. Fundamental Rayleigh wave phase velocities are extracted for periods of 0. 6 similar to 1. 2 s using DBF method. Then, a 2-D V, profile model from the surface to 1200 m depth is derived by shear wave inversion. V, results show that the shallow low-velocity anomalies in the northeast side of the array (near the Pearl River Delta) extend to the depth of 600 m, reflecting differences in sediment thickness, compaction rate and lithologic composition in different blocks on both sides of Hualong fault. In addition, the relative high-velocity anomalies below depth of 900 m correspond to the bedrock formed by Tertiary magmatic activities. The near-surface structural characteristics of Hualong fault effectively improve our cognition of the fault system in the Pearl River Delta, which also provide valuable references for further studies on regional structural characteristics,focal mechanisms and seismic hazard assessment.

The broadband ocean bottom seismograph (OBS) is the core instrument for seismological studies such as imaging the oceanic lithosphere. This paper presents the design of the domestic seismometer-detached broadband OBS (Pankun) and its tests in the South China Sea (SCS), as well as a detailed analysis of its performance and data quality. In addition to the seismometer-detached type, Pankun has the following features: a unique anti-current structure, an automatic device separating the seismometer, a broadband seismometer (100 Hz similar to 120 s), and low-power consumption and long operational time. Compared with the OBSs in the previous experiment in 2012, Pankun has much better performances in recovery rate, horizontal orientation determination, leveling device, and clock accuracy. More importantly, the quality of earthquake waveforms has also been greatly improved benefits from its unique anti-current structure, especially for the horizontal components. The quality of horizontal components of Pankun is comparable to nearby land stations, while low-frequency components (>10 s) have lower noise levels than international counterparts. These analyses show that Pankun's data can be used for most seismological applications, including receiver function and anisotropy analysis that require three-component seismic data. Finally, the broadband characteristics of Pankun' s data, especially high-quality, low-frequency signals, are more conducive to studying the structure and processes of the oceanic lithosphere and upper mantle using the long-period surface waves and the finite frequency body waves.

Methodology, supplemental Figures S1–S11, and Tables S1 and S2.

SUMMARY The ocean is the primary source of seismic ambient noise. Therefore, seismic recordings at seafloor stations should reveal noise characteristics more directly than land stations. However, due to a lack of broad-band seismic instrumentation, seafloor noise studies using seafloor stations have been inadequate compared to land-based instrumentation. In this study, we use seismic data collected at the South China Sea (SCS) seafloor by newly developed ocean bottom seismographs (OBSs) to analyze the ambient noise features in this marginal sea. The broad-band OBS, dubbed ‘Pankun’, has unique shielding to isolate its sensor from the influences of bottom currents. A side-by-side land test between the OBS sensor unit and a standalone seismometer showed that the self-noise caused by the gimbal and the pressure case is insignificant. The recordings on the SCS seafloor have distinct noise spectra. The double frequency microseisms (DFMs) have a single instead of double peak like that seen for Pacific stations. The peak appears in a lower period range (1–5 s) than in the global noise model, indicating that the primary source region for the DFM is the SCS itself. The high-frequency content of the DFM is attenuated more as it propagates from its source region (seafloor) to land stations. The single frequency microseism (SFM) peak on the spectrum is weak, reflecting that SFMs, generated in shallow water along the coast, have difficulties propagating back into the deep ocean due to the substantial increase in seafloor depth. A long-period Earth's hum signal is also identifiable on the vertical component at periods greater than 50 s, probably due to the anti-current design of the OBS. Although the seasonal sea state mainly affects the noise level, extreme events such as typhoons can produce short-term abnormally high DFMs in the basin. However, the DFM highs caused by such events exhibit complex patterns, depending on the wind speed, duration, and area covered by the events.

We present a novel eikonal tomography approach using physics‐informed neural networks (PINNs) for Rayleigh wave phase velocities based on the eikonal equation. The PINN eikonal tomography (pinnET) neural network utilizes deep neural networks as universal function approximators and extracts traveltimes and velocities of the medium during the optimization process. Whereas classical eikonal tomography uses a generic non‐physics based interpolation and regularization step to reconstruct traveltime surfaces, optimizing the network parameters in pinnET means solving a physics constrained traveltime surface reconstruction inversion tackling measurement noise and satisfying physics. We demonstrate this approach by applying it to 25 s surface wave data from ChinArray II sampling the northeastern Tibetan plateau. We validate our results by comparing them to results from conventional eikonal tomography in the same area and find good agreement.

During the last 50 Ma, the eastern Eurasian continent has experienced widespread crustal and lithospheric deformation. However, mechanisms in creating such widespread intra‐continental deformation are still not well understood. Here, we present a 3‐D S‐wave velocity model of the northeastern Tibetan Plateau and adjacent regions by jointly inverting teleseismic body and surface waves. The resulting model clearly suggests that lateral extrusion of asthenosphere from northern Tibet is being blocked by the thick cratonic keels of Ordos and Sichuan, and therefore the asthenosphere is channeled into a strong eastward flow beneath the northern Qinglin between these two cratonic keels. Our results also demonstrate that continental deformation in eastern Eurasia is likely influenced by the small‐scale convection when the lateral asthenospheric flow from NE Tibet encounters preexisting continental lithospheric steps to the east of Ordos.

Deforestation and intensive land use have accelerated soil erosion, reshaped topography, and altered carbon reservoirs for thousands of years. The timing, scope, and magnitude of long‐term anthropogenic soil erosion across China are especially important to understand the global scale of this process. Here, sediment accumulation rates (SARs) from 191 sediment archives are found to be temporally correlated with monsoon intensity during 6–40 ka BP, indicating that hydroclimate was the main driver of soil erosion in this time interval. The rapid increase in SARs after ca. 5 ka BP is decoupled from persistently weakened hydroclimate but instead follows the trend of increasing population and related agricultural activities in China, implying a change in the primary controlling factor since then. Early human activities in China therefore appear to have had profound implications on Earth's surface at a continental scale.

We have presented the 410 km and 660 km mantle discontinuity structure and mantle transition zone (MTZ) thickness beneath the middle Southern China Block using seismic data recorded by 84 portable seismic stations in the region by receiver function method. We found that transition zone thickness beneath Cathaysia Block is thinned 10 similar to 25 km obviously, the MTZ thinned anomalously within an area approximately 200 km in diameter centered (23. 7 degrees N, 114. 5 degrees E), where the Hainan mantle plume may upwell into upper mantle, the continuous upwelling of hot mantle materials resulted in a wide range of low velocity anomalies in the upper mantle beneath the Cathaysia Block (above 410 km), also a large range of Cenozoic basalts in Leizhou Peninsula and coastal areas.

Employing the seismic data observed by 41 temporary seismic stations deployed by South University of Science and Technology in Shenzhen City from December, 2017 to February, 2018, we first obtain Rayleigh wave phase velocity of 0.5 similar to 5 s period by Ambient Noise tomography method in this region, then invert S wave velocity structure from Earth's surface to 8 km depth. Results show that the velocity structures near the Earth's surface are well correspond to local tectonics. A salient feature is that high and low velocity staggered distribution exist in all depths, and the demarcations of high and low velocity anomalies are consistent with faults exposed. High velocity zones horizontally located at different depths are well connected via fault zones and obviously contact with the exposed areas of volcanic-intrusive mass on the Earth' s surface, interbedded with low velocity anomaly sheets. Deep rising high anomaly zones are correspond with areas lacking of low velocity anomalies at shallow depths, where Mesozoic granites exposed, usually hills. This phenomenon implies that deep hot material intruded in Mesozoic via faults and flooded between sedimentary rock layers, deforming the crust structure. The results show that velocity structure and exposed geological tectonic features are consistent in this region; also show potential for well studying the Earth's deep structure in urban areas by employing densely deployed short-period seismic array data.

利用横波分裂方法,对北京大学、南方科技大学及桂林理工大学在广东和广西布设的宽频带流动台阵以及国家地震台网在广东、广西和海南的固定台站记录的XKS震相进行计算,得到海南及邻近地区的上地幔各向异性参数,分裂延迟时间在0.35～1.90s之间.综合研究区域内前人发表的和本文的横波分裂结果,认为自新生代以来,伴随东部太平洋板块俯冲带后撤,华南地区西部软流圈物质向东流动进行补充,导致广东及广西的中部和北部大部分地区上地幔橄榄石晶格大体上沿E-W方向定向排列.受控于海岸线处大陆岩石圈迅速减薄,广东沿海地区横波分裂多数沿NE-SW方向,与海岸线近似平行.特别是,海南岛附近沿海地区分布若干显著不具有横波分裂的台站;海南岛东北角的分裂方向很乱,且横波分裂快慢波到时差(δt)很小.这些观测结果都支持海南地幔柱可能存在于雷州半岛的东南方向,大致位于20.5°N,111.5°E.

The occurrence of large earthquakes is not only related to the fault geometry and state of stress accumulation, but also closely related to the structure of surrounding rocks. With the continuous enrichment of observation methods and development of techniques in recent decades, various studies have been conducted to explore the fine structure of large earthquake source regions and link the observations to the mechanism of large earthquakes. Seismologists generally use the asperity model to explain the generation mechanism of large earthquakes, that is, the process from locking to sudden slip of the seismogenic faults. The detailed geometrical shape of the fault planes and accurate state of stress accumulation are usually tough to depict during the process, while the structure features of the surrounding rocks are available through seismic tomography and magnetotelluric studies. In this paper, we aim to systematically summarize the results of structure studies in the focal areas of large earthquakes, and discuss the relationship between structure features and asperities. We therefore made statistics on the studies of asperity in the source regions of worldwide large earthquakes with magnitude larger than 6.0. There are totally 123 events with different focal mechanism, among which 54 events are intraplate earthquakes, while the other 69 events are interplate earthquakes. These large earthquakes are well studied by structure and/or source rupture process researches. Structure studies carried on 17 intraplate and 14 interplate earthquakes suggest the presence of high strength (tomographic high-velocity and/or magnetotelluric high-resistance) bodies surrounding the seismogenic faults. These anomalous bodies are considered to be the asperities, which play an important role in the generation of large earthquakes. Among the 31 large earthquakes, there are 4 events that took place on subducted seamounts, which are characterized as patches of anomalous structure features as well as elevated effective normal stress and are asperities with high strength in the evolution of these large earthquakes. The structure studies provide direct evidence for the important role of asperity in the generation of large earthquakes. The indirect evidences come from the source rupture studies of 92 large earthquakes, which show large coseismic slip zones near the seismogenic faults, manifesting the existence of asperities with high strength. The predominance of unilateral rupture in earthquakes has also been revealed, which has a close relationship with the structural heterogeneity. Through statistical analysis of asperity mechanism studies on large earthquakes, we find the asperity characterized by high velocity and resistivity plays an important role in the generation of large earthquakes. The asperities are strong in mechanical strength and could accumulate tectonic stress more easily during the long frictional locking periods, large earthquakes are therefore prone to generate in these areas. If the close relationship between asperity and high-velocity and high-resistivity bodies is valid for most of the large earthquakes, it can be used to predict potential large earthquakes and estimate the seismogenic capability of faults in light of structure studies, which will provide important clues for earthquake prevention and disaster reduction. Furthermore, the asperity model is irrelevant to focal mechanism and plays a same important role in thrusting, normal and strike-slip faults.