Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100392
Jingqi Wang , Rumeng Guo , Jianqiao Xu , Heping Sun
Identifying precursors of large earthquakes is critical for minimizing the losses of life and property. Recently, Bletery and Nocquet (2023) captured a ∼2-h-long exponential acceleration of slip using the high-rate (5-min) Global Navigation Satellite System (GNSS) time series from the 48 hr before the 2011 MW 9.0 Tohoku-oki earthquake, which was obtained by simply concatenating daily kinematic results together. Here, we apply their method to sum the horizontal displacements of 24 high-rate GNSS stations in the direction predicted by fault slip at the hypocenter of the 2023 MW 7.8 Kahramanmaraş earthquake to characterize its precursory phase. Results demonstrate a several-hour accelerating exponential slip before the mainshock. However, considering that single-day processing would lead to discontinuities at the day boundary, we process the multi-day GNSS data in continuous mode, repeat the experiment, and find that the observed acceleration-like signals vanish. Our work shows that inadequate data processing may lead to the detection of false precursory signals, highlighting the need to develop robust processing techniques to identify reliable precursory signals before large earthquakes.
{"title":"Assessing precursory signals with kinematic GNSS: Insights from the 2023 Mw 7.8 Kahramanmaraş earthquake","authors":"Jingqi Wang , Rumeng Guo , Jianqiao Xu , Heping Sun","doi":"10.1016/j.eqrea.2025.100392","DOIUrl":"10.1016/j.eqrea.2025.100392","url":null,"abstract":"<div><div>Identifying precursors of large earthquakes is critical for minimizing the losses of life and property. Recently, Bletery and Nocquet (2023) captured a ∼2-h-long exponential acceleration of slip using the high-rate (5-min) Global Navigation Satellite System (GNSS) time series from the 48 hr before the 2011 <em>M</em><sub>W</sub> 9.0 Tohoku-oki earthquake, which was obtained by simply concatenating daily kinematic results together. Here, we apply their method to sum the horizontal displacements of 24 high-rate GNSS stations in the direction predicted by fault slip at the hypocenter of the 2023 <em>M</em><sub>W</sub> 7.8 Kahramanmaraş earthquake to characterize its precursory phase. Results demonstrate a several-hour accelerating exponential slip before the mainshock. However, considering that single-day processing would lead to discontinuities at the day boundary, we process the multi-day GNSS data in continuous mode, repeat the experiment, and find that the observed acceleration-like signals vanish. Our work shows that inadequate data processing may lead to the detection of false precursory signals, highlighting the need to develop robust processing techniques to identify reliable precursory signals before large earthquakes.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100392"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100393
Xinwei Tang , Gaozhi Xin , Danqing Song
Soft soil is widely distributed and has complex origins. Shield tunnels are inevitably constructed within soft soil interlayers, and under seismic action, tunnels may be subject to severe damages. On the basis of actual engineering, this study utilized dynamic triaxial testing to investigate the dynamic properties of soft soil. Using the PIMY constitutive model, the seismic subsidence characteristics of the soft soil were characterized, and a refined finite element model was established to study the mutual influence mechanism between the soft soil layer and shield tunnels via the open-source software framework OpenSees. The results demonstrate that soil exhibits a softening effect under dynamic loading; soft soil with better structural integrity is less prone to seismic subsidence; and the greater the inertial force acting on the soft soil, the greater the likelihood of settlement. Under seismic action, the presence of the shield tunnel exacerbates the settlement of the soft soil, as the surrounding soil experiences significant inertial forces from the tunnel structure, hindering drainage and accelerating the accumulation of pore water pressure; The soft soil itself has large deformation and displacement under the action of earthquake, which leads to the great stress, deformation, and displacement of the structure. The arch foot position of the tunnel is identified as the most vulnerable to damage.
{"title":"Mutual influence mechanism between soft soil interlayers and shield tunnels under seismic action: Laboratory tests and numerical simulations","authors":"Xinwei Tang , Gaozhi Xin , Danqing Song","doi":"10.1016/j.eqrea.2025.100393","DOIUrl":"10.1016/j.eqrea.2025.100393","url":null,"abstract":"<div><div>Soft soil is widely distributed and has complex origins. Shield tunnels are inevitably constructed within soft soil interlayers, and under seismic action, tunnels may be subject to severe damages. On the basis of actual engineering, this study utilized dynamic triaxial testing to investigate the dynamic properties of soft soil. Using the PIMY constitutive model, the seismic subsidence characteristics of the soft soil were characterized, and a refined finite element model was established to study the mutual influence mechanism between the soft soil layer and shield tunnels via the open-source software framework OpenSees. The results demonstrate that soil exhibits a softening effect under dynamic loading; soft soil with better structural integrity is less prone to seismic subsidence; and the greater the inertial force acting on the soft soil, the greater the likelihood of settlement. Under seismic action, the presence of the shield tunnel exacerbates the settlement of the soft soil, as the surrounding soil experiences significant inertial forces from the tunnel structure, hindering drainage and accelerating the accumulation of pore water pressure; The soft soil itself has large deformation and displacement under the action of earthquake, which leads to the great stress, deformation, and displacement of the structure. The arch foot position of the tunnel is identified as the most vulnerable to damage.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100393"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154214","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100376
Xingao Li , Zhongtai He , Long Guo , Linlin Li
The northern section of the Xiaojiang fault is the most active section in the Xiaojiang Fault Zone, and a detailed interpretation of this fault is highly important. In this work, KeyHole-4B images and Landsat 8 images of the northern section of the Xiaojiang fault were collected, and remote sensing interpretation and tectonic geomorphological analysis of the northern section of the Xiaojiang fault were carried out to obtain a more detailed fault distribution. The results reveal that the northern section of the Xiaojiang fault is a group of faults that are subparallel to each other with a space of 2–4 km. The fault is located along the Jinshajiang Valley and the Xiaojiang Valley. At the same time, we counted the large-scale left-lateral dislocations of the gullies and ridges. Combined with the results of previous studies, the long-term average slip rate of the northern section of the Xiaojiang fault is 6.2 ± 1.1 mm/a since the late Middle Pleistocene, 11.4 ± 2.8 mm/a since the middle of the late Pleistocene, and 8.0 ± 2.0 mm/a since the middle and late Pleistocene. The high slip rate in the northern section of the Xiaojiang fault represents the response of the local strain of the central Yunnan subblock, which rotates clockwise along the boundary fault. This finding is consistent with the pattern of northwards and north-east wards thrusting of the Indian plate, leading to eastwards extrusion and the escape of material from the Qinghai-Xizang Plateau.
{"title":"Using historical remote sensing images for detailed tectonic geomorphological interpretation in the study of active faults: Application to the Xiaojiang fault case study","authors":"Xingao Li , Zhongtai He , Long Guo , Linlin Li","doi":"10.1016/j.eqrea.2025.100376","DOIUrl":"10.1016/j.eqrea.2025.100376","url":null,"abstract":"<div><div>The northern section of the Xiaojiang fault is the most active section in the Xiaojiang Fault Zone, and a detailed interpretation of this fault is highly important. In this work, KeyHole-4B images and Landsat 8 images of the northern section of the Xiaojiang fault were collected, and remote sensing interpretation and tectonic geomorphological analysis of the northern section of the Xiaojiang fault were carried out to obtain a more detailed fault distribution. The results reveal that the northern section of the Xiaojiang fault is a group of faults that are subparallel to each other with a space of 2–4 km. The fault is located along the Jinshajiang Valley and the Xiaojiang Valley. At the same time, we counted the large-scale left-lateral dislocations of the gullies and ridges. Combined with the results of previous studies, the long-term average slip rate of the northern section of the Xiaojiang fault is 6.2 ± 1.1 mm/a since the late Middle Pleistocene, 11.4 ± 2.8 mm/a since the middle of the late Pleistocene, and 8.0 ± 2.0 mm/a since the middle and late Pleistocene. The high slip rate in the northern section of the Xiaojiang fault represents the response of the local strain of the central Yunnan subblock, which rotates clockwise along the boundary fault. This finding is consistent with the pattern of northwards and north-east wards thrusting of the Indian plate, leading to eastwards extrusion and the escape of material from the Qinghai-Xizang Plateau.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100376"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154208","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100399
Joseph Omeiza Alao
Seismic hazard assessment (SHA) is crucial for mitigating earthquake hazards, particularly in tectonically active regions. This study critically examines the emerging roles of 3D and 4D geophysical and geological modelling in assessing SHA, focusing on advancements, applications, and limitations. 3D geophysical modelling provides high-resolution spatial representations of fault networks, stress distributions, and seismic-prone zones. In contrast, 4D geophysical modelling integrates temporal dynamics to analyze subsurface variations or fault systems over time. Based on previous studies, the quantitative data highlight the effectiveness of real-time seismic monitoring, with stress accumulation rates ranging from 0.01% to 50% during seismic events. Time-lapse seismic data improves forecasting precision, with early warning detection reducing seismic uncertainties by over 30%. Additionally, studies show that enhanced fluid migration tracking using 4D seismic modelling, leading to a 25% increase in hydrocarbon recovery efficiency. These advancements aim in urban planning, infrastructure resilience, and hazard mitigation strategies. However, challenges remain in data acquisition, computational demands, and model interpretation. The integration of artificial intelligence and high-performance computing is expected to improve predictive modelling accuracy, ensuring more effective SHA. The findings emphasize the importance of geophysical modelling in disaster preparedness, reinforcing the need for technological advancements to enhance seismic hazard mitigation strategies and infrastructure safety.
{"title":"The emerging roles of 3D and 4D geophysical and geological modelling in evaluating seismic risks: A critical review","authors":"Joseph Omeiza Alao","doi":"10.1016/j.eqrea.2025.100399","DOIUrl":"10.1016/j.eqrea.2025.100399","url":null,"abstract":"<div><div>Seismic hazard assessment (SHA) is crucial for mitigating earthquake hazards, particularly in tectonically active regions. This study critically examines the emerging roles of 3D and 4D geophysical and geological modelling in assessing SHA, focusing on advancements, applications, and limitations. 3D geophysical modelling provides high-resolution spatial representations of fault networks, stress distributions, and seismic-prone zones. In contrast, 4D geophysical modelling integrates temporal dynamics to analyze subsurface variations or fault systems over time. Based on previous studies, the quantitative data highlight the effectiveness of real-time seismic monitoring, with stress accumulation rates ranging from 0.01% to 50% during seismic events. Time-lapse seismic data improves forecasting precision, with early warning detection reducing seismic uncertainties by over 30%. Additionally, studies show that enhanced fluid migration tracking using 4D seismic modelling, leading to a 25% increase in hydrocarbon recovery efficiency. These advancements aim in urban planning, infrastructure resilience, and hazard mitigation strategies. However, challenges remain in data acquisition, computational demands, and model interpretation. The integration of artificial intelligence and high-performance computing is expected to improve predictive modelling accuracy, ensuring more effective SHA. The findings emphasize the importance of geophysical modelling in disaster preparedness, reinforcing the need for technological advancements to enhance seismic hazard mitigation strategies and infrastructure safety.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100399"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100402
Rhommel Grutas, Jhon Philip Camayang, Justine Anne Duka, Miguel Antonio Magandi, John Edward Nachor, Jedrek Angelo G. Tupas, Guia Angela C. Agoncillo
The Philippines' current seismic design framework, grounded in outdated uniform hazard approaches, fails to ensure consistent structural safety due to regional variations in seismic risk and structural fragility. This study aims to develop the first risk-targeted seismic hazard maps for the Philippines, adopting the ASCE 7–16 framework and integrating updated probabilistic seismic hazard data from the SHADE Project. Through the application of risk-integral formulations, the annual probability of structural collapse was computed by convolving seismic hazard curves and lognormal fragility functions. A parametric analysis was conducted using varying fragility dispersions (β = 0.6, 0.7, 0.8) and target collapse probabilities () to evaluate their effects on risk coefficients (), conditional collapse probabilities, and hazard curve slopes (η) for spectral accelerations at 0.2 s and 1.0 s. Results reveal that higher fragility dispersions and lower collapse targets significantly increase required design motions, particularly in short-period structures. The selected baseline parameters, β = 0.6 and = 2 × 10−4 (1 % collapse risk in 50 years), yielded consistent collapse probabilities and aligned with international standards. Spatial analyses showed elevated in high-hazard zones such as Western Luzon, Eastern Visayas and Mindanao, while a strong correlation between and η underscores the importance of hazard curve shape in seismic design. All computations assumed rock site conditions, with future work recommended to address site-specific effects.
{"title":"Risk-targeted seismic hazard model for the Philippines","authors":"Rhommel Grutas, Jhon Philip Camayang, Justine Anne Duka, Miguel Antonio Magandi, John Edward Nachor, Jedrek Angelo G. Tupas, Guia Angela C. Agoncillo","doi":"10.1016/j.eqrea.2025.100402","DOIUrl":"10.1016/j.eqrea.2025.100402","url":null,"abstract":"<div><div>The Philippines' current seismic design framework, grounded in outdated uniform hazard approaches, fails to ensure consistent structural safety due to regional variations in seismic risk and structural fragility. This study aims to develop the first risk-targeted seismic hazard maps for the Philippines, adopting the ASCE 7–16 framework and integrating updated probabilistic seismic hazard data from the SHADE Project. Through the application of risk-integral formulations, the annual probability of structural collapse was computed by convolving seismic hazard curves and lognormal fragility functions. A parametric analysis was conducted using varying fragility dispersions (β = 0.6, 0.7, 0.8) and target collapse probabilities (<span><math><mrow><msub><mi>P</mi><mtext>fail</mtext></msub></mrow></math></span>) to evaluate their effects on risk coefficients (<span><math><mrow><msub><mi>C</mi><mi>R</mi></msub></mrow></math></span>), conditional collapse probabilities, and hazard curve slopes (η) for spectral accelerations at 0.2 s and 1.0 s. Results reveal that higher fragility dispersions and lower collapse targets significantly increase required design motions, particularly in short-period structures. The selected baseline parameters, β = 0.6 and <span><math><mrow><msub><mi>P</mi><mtext>fail</mtext></msub></mrow></math></span> = 2 × 10<sup>−4</sup> (1 % collapse risk in 50 years), yielded consistent collapse probabilities and aligned with international standards. Spatial analyses showed elevated <span><math><mrow><msub><mi>C</mi><mi>R</mi></msub></mrow></math></span> in high-hazard zones such as Western Luzon, Eastern Visayas and Mindanao, while a strong correlation between <span><math><mrow><msub><mi>C</mi><mi>R</mi></msub></mrow></math></span> and η underscores the importance of hazard curve shape in seismic design. All computations assumed rock site conditions, with future work recommended to address site-specific effects.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100402"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100398
Huifang Chen , Binhua Lin , Tairan Xu , Yanming Zhang , Yuanhong Yang
This study proposes a method for determining earthquake focal depths by combining the sPn phase with the waveform cross-correlation technique, based on waveform data recorded by the Fujian Seismic Network from the 2024 M 7.3 Hualien offshore earthquake and the 2025 M 6.2 Tainan earthquake. The Pn phase onset was precisely aligned using waveform cross-correlation, and the arrival time difference () between the sPn and Pn phases was extracted via a sliding time-window correlation method. The focal depths were derived using a layered velocity model for the Taiwan region. Results show that the calculated focal depth for the Hualien earthquake is 23.1 km ( = 6.9 s), with a relative error of 2.7% compared to the official result (22.5 km) from the Central Weather Administration of Taiwan. For the Tainan earthquake, the depth is 17.9 km ( = 6.1 s), with a relative error of 13.3%. In this study, we show that a cross-correlation threshold of 0.8 and a bandpass filtering of 0.1–0.3 Hz are efficient to suppress noise and significantly improve depth accuracy for shallow earthquakes with depth <30 km. Compared to traditional travel-time location methods, this approach exhibits superior noise resistance and computational efficiency. Future work will focus on optimizing 3D velocity structures, integrating multiple phases, and applying deep learning techniques such as convolutional neural networks, aiming to improve the results in a more reliable and automatic way, and to provide efficient support on earthquake emergency response.
本文以福建省台南地震台网记录的2024年花莲7.3级地震和2025年台南6.2级地震波形资料为基础，提出了sPn相位与波形互相关技术相结合的震源深度确定方法。利用波形互相关精确对准Pn相位起始点，利用滑动时间窗相关方法提取sPn与Pn相位的到达时间差（Δt）。采用层状速度模型对台湾地区进行震源深度反演。结果表明，花莲地震震源深度为23.1 km （Δt = 6.9 s），与台湾中央气象局官方结果（22.5 km）的相对误差为2.7%。台南地震震源深度为17.9 km （Δt = 6.1 s），相对误差为13.3%。在本研究中，我们发现0.8的互相关阈值和0.1-0.3 Hz的带通滤波可以有效地抑制噪声，并显著提高深度为30 km的浅层地震的深度精度。与传统的行时定位方法相比，该方法具有更好的抗噪声性能和计算效率。未来的工作将侧重于优化三维速度结构，整合多阶段，并应用卷积神经网络等深度学习技术，以更加可靠和自动化的方式提高结果，为地震应急响应提供有效的支持。
{"title":"Determination of focal depths for the 2024 M 7.3 Hualien offshore and 2025 M 6.2 Tainan earthquakes in Taiwan, China: An enhanced method based on sPn phase and waveform cross-correlation techniques","authors":"Huifang Chen , Binhua Lin , Tairan Xu , Yanming Zhang , Yuanhong Yang","doi":"10.1016/j.eqrea.2025.100398","DOIUrl":"10.1016/j.eqrea.2025.100398","url":null,"abstract":"<div><div>This study proposes a method for determining earthquake focal depths by combining the sPn phase with the waveform cross-correlation technique, based on waveform data recorded by the Fujian Seismic Network from the 2024 <em>M</em> 7.3 Hualien offshore earthquake and the 2025 <em>M</em> 6.2 Tainan earthquake. The Pn phase onset was precisely aligned using waveform cross-correlation, and the arrival time difference (<span><math><mo>Δ</mo><mi>t</mi></math></span>) between the sPn and Pn phases was extracted via a sliding time-window correlation method. The focal depths were derived using a layered velocity model for the Taiwan region. Results show that the calculated focal depth for the Hualien earthquake is 23.1 km (<span><math><mo>Δ</mo><mi>t</mi></math></span> = 6.9 s), with a relative error of 2.7% compared to the official result (22.5 km) from the Central Weather Administration of Taiwan. For the Tainan earthquake, the depth is 17.9 km (<span><math><mo>Δ</mo><mi>t</mi></math></span> = 6.1 s), with a relative error of 13.3%. In this study, we show that a cross-correlation threshold of 0.8 and a bandpass filtering of 0.1–0.3 Hz are efficient to suppress noise and significantly improve depth accuracy for shallow earthquakes with depth <30 km. Compared to traditional travel-time location methods, this approach exhibits superior noise resistance and computational efficiency. Future work will focus on optimizing 3D velocity structures, integrating multiple phases, and applying deep learning techniques such as convolutional neural networks, aiming to improve the results in a more reliable and automatic way, and to provide efficient support on earthquake emergency response.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100398"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154206","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100390
Lue Yang , Lun Li , Xiuwei Ye , Jinming Zhang , Jialong He , Yuan Gao , Pengfei Li , Yu He , Ziwei Li
The Shantou region, one of the most seismically active zones of Guangdong Province (South China), has experienced multiple strong earthquakes, including two significant events with magnitudes greater than 7.0 that occurred in 1600 and 1918, respectively. To investigate the seismogenic structures responsible for these major earthquakes and their potential triggering mechanisms, we construct a high-resolution shear-wave velocity model from the surface to ∼15 km depth based on a dense nodal seismic array using ambient noise tomography. The model reveals a pronounced low-velocity zone (LVZ) at depths of 2–15 km, with a perturbation of −2 to −8%, coinciding with the northwestern extension of the Huanggangshui Fault previously identified in the offshore region. Integrating our results with previous field geological surveys and shallow reflection seismic exploration, we interpret the LVZ as a wide fault zone, potentially comprising multiple fault branches that possibly include two NW-trending faults (i.e., the Rongjiang Fault and the Hanjiang Fault). Notably, the interaction between the Huanggangshui Fault and NE-trending Littoral Fault is suggested to have triggered the 1918 M 7.3 Nan'ao earthquake. Additionally, the 1895 M 6.2 earthquake seems to have occurred at the edge of the LVZ near the Rongjiang Fault, a possible branch of the Huanggangshui Fault, further supporting the association between this structure and seismic activity. These findings imply that the LVZ may represent a region of concentrated tectonic stress, making it a potential site for future strong earthquakes. Consequently, this area should be prioritized in seismic hazard assessments. This study provides valuable insights into the seismogenic characteristics of the Shantou region and contributes to improving seismic hazard evaluations in South China.
{"title":"The seismogenic structure of strong intraplate earthquakes in Shantou region, South China: Insights from upper crustal shear-wave velocity structure","authors":"Lue Yang , Lun Li , Xiuwei Ye , Jinming Zhang , Jialong He , Yuan Gao , Pengfei Li , Yu He , Ziwei Li","doi":"10.1016/j.eqrea.2025.100390","DOIUrl":"10.1016/j.eqrea.2025.100390","url":null,"abstract":"<div><div>The Shantou region, one of the most seismically active zones of Guangdong Province (South China), has experienced multiple strong earthquakes, including two significant events with magnitudes greater than 7.0 that occurred in 1600 and 1918, respectively. To investigate the seismogenic structures responsible for these major earthquakes and their potential triggering mechanisms, we construct a high-resolution shear-wave velocity model from the surface to ∼15 km depth based on a dense nodal seismic array using ambient noise tomography. The model reveals a pronounced low-velocity zone (LVZ) at depths of 2–15 km, with a perturbation of −2 to −8%, coinciding with the northwestern extension of the Huanggangshui Fault previously identified in the offshore region. Integrating our results with previous field geological surveys and shallow reflection seismic exploration, we interpret the LVZ as a wide fault zone, potentially comprising multiple fault branches that possibly include two NW-trending faults (i.e., the Rongjiang Fault and the Hanjiang Fault). Notably, the interaction between the Huanggangshui Fault and NE-trending Littoral Fault is suggested to have triggered the 1918 <em>M</em> 7.3 Nan'ao earthquake. Additionally, the 1895 <em>M</em> 6.2 earthquake seems to have occurred at the edge of the LVZ near the Rongjiang Fault, a possible branch of the Huanggangshui Fault, further supporting the association between this structure and seismic activity. These findings imply that the LVZ may represent a region of concentrated tectonic stress, making it a potential site for future strong earthquakes. Consequently, this area should be prioritized in seismic hazard assessments. This study provides valuable insights into the seismogenic characteristics of the Shantou region and contributes to improving seismic hazard evaluations in South China.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100390"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154207","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100397
Yujiang Li, Cheng Yang, Xingping Hu, Jie Yuan, Rui Yao, Hong Li
On 28 March 2025, a strong MW 7.7 earthquake struck the seismic gap in the central section of the Sagaing Fault in Myanmar, causing significant damages and casualties in Myanmar and neighboring countries. Major earthquakes like this are expected to transfer stresses to nearby active regions and change their seismic hazards in the near future. In this study, based on a stratified viscoelastic model and a coseismic slip model, we calculated the co- and post-seismic Coulomb stress change (△CFS) imparted by the MW 7.7 Myanmar earthquake to the main active faults in the adjacent southwestern Yunnan region in China. Our results show that five fault segments experience up to 3 kPa of coseismic stress increase, including the Longling-Lancang Fault, the Nantinghe Fault, the Menglian Fault, the Heihe Fault, and the Red River Fault, respectively. The pattern of postseismic △CFS is similar to that of coseismic △CFS, suggesting that with the increasing elapsed time, the stress level continues to increase in these fault zones. The coseismic auxiliary stress fields show that the orientation of the principal tensile stress is predominantly NE-SW in the northern part of the southwestern Yunnan region, and shows clockwise rotation to NW-SE in the south. This stress regime controls the additional slip motion, consistent with that reflected by the coseismic shear stress change. Combined with other geophysical and geodetic data, we propose that more attention should be paid to the Longling-Lancang Fault, the Nantinghe Fault, the Menglian Fault, and the Heihe Fault, potential candidates for the next strong earthquakes in this region.
{"title":"Coulomb stress transfer from the 2025 MW 7.7 Myanmar earthquake to active faults in southwestern Yunnan, China: Implications for seismic hazard","authors":"Yujiang Li, Cheng Yang, Xingping Hu, Jie Yuan, Rui Yao, Hong Li","doi":"10.1016/j.eqrea.2025.100397","DOIUrl":"10.1016/j.eqrea.2025.100397","url":null,"abstract":"<div><div>On 28 March 2025, a strong <em>M</em><sub>W</sub> 7.7 earthquake struck the seismic gap in the central section of the Sagaing Fault in Myanmar, causing significant damages and casualties in Myanmar and neighboring countries. Major earthquakes like this are expected to transfer stresses to nearby active regions and change their seismic hazards in the near future. In this study, based on a stratified viscoelastic model and a coseismic slip model, we calculated the co- and post-seismic Coulomb stress change (△<em>CFS</em>) imparted by the <em>M</em><sub>W</sub> 7.7 Myanmar earthquake to the main active faults in the adjacent southwestern Yunnan region in China. Our results show that five fault segments experience up to 3 kPa of coseismic stress increase, including the Longling-Lancang Fault, the Nantinghe Fault, the Menglian Fault, the Heihe Fault, and the Red River Fault, respectively. The pattern of postseismic △<em>CFS</em> is similar to that of coseismic △<em>CFS</em>, suggesting that with the increasing elapsed time, the stress level continues to increase in these fault zones. The coseismic auxiliary stress fields show that the orientation of the principal tensile stress is predominantly NE-SW in the northern part of the southwestern Yunnan region, and shows clockwise rotation to NW-SE in the south. This stress regime controls the additional slip motion, consistent with that reflected by the coseismic shear stress change. Combined with other geophysical and geodetic data, we propose that more attention should be paid to the Longling-Lancang Fault, the Nantinghe Fault, the Menglian Fault, and the Heihe Fault, potential candidates for the next strong earthquakes in this region.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100397"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-01DOI: 10.1016/j.eqrea.2025.100391
Dongdong Cao , Jun Zhang , Ming Li , Baoqiang Chen , Jia Li , Xiaolong Wu
To address the insufficient understanding of the dynamic coupling between surface subsidence and multi-depth rock mass deformation induced by underground mining, this study focuses on the 520109 working face of the Shagoucha Coal Mine in Shaanxi Province. Most existing subsidence prediction models rely heavily on surface deformation data and often overlook the temporal evolution of deep rock mass responses, limiting their predictive accuracy under complex geological conditions. In this context, we implement a fully integrated GNSS–borehole monitoring system to obtain high-frequency continuous GNSS observations and internal deformation time series at three key depths (14 m, 92 m, and 132 m). To reveal the dynamic correlation between strata deformations and surface subsidence across multiple time scales, cross-wavelet transform (XWT) analysis is applied to quantify both amplitude and phase relationships in the time–frequency domain. The results demonstrate that surface subsidence consistently lags behind deep rock mass deformation, with the deepest monitored stratum (132 m) showing the earliest and largest deformation. The 92 m layer (primary subsidence deformation zone) also displays a leading response, particularly in high-frequency bands, indicating its role in stress redistribution and transmission. In contrast, the shallow 14 m loess layer exhibits a lagging and hydrologically sensitive behavior, responding passively to overlying subsidence. These results highlight the stratified and frequency-dependent nature of deformation evolution, emphasizing the significance of deep rock mass signals as early indicators of subsidence progression. By integrating multi-depth deformation monitoring with time–frequency correlation analysis, this study provides novel insights into the temporal hierarchy of mining-induced subsidence. It provides theoretical support for refining subsidence prediction models and early warning systems. Compared with previous studies that focus primarily on surface or single-depth data, our approach provides a more comprehensive framework for interpreting the spatiotemporal dynamics of stratified deformation processes in mining areas.
{"title":"Spatiotemporal correlation of multi-depth rock mass deformation and mining-induced subsidence: A case study of the Shagoucha Coal Mine","authors":"Dongdong Cao , Jun Zhang , Ming Li , Baoqiang Chen , Jia Li , Xiaolong Wu","doi":"10.1016/j.eqrea.2025.100391","DOIUrl":"10.1016/j.eqrea.2025.100391","url":null,"abstract":"<div><div>To address the insufficient understanding of the dynamic coupling between surface subsidence and multi-depth rock mass deformation induced by underground mining, this study focuses on the 520109 working face of the Shagoucha Coal Mine in Shaanxi Province. Most existing subsidence prediction models rely heavily on surface deformation data and often overlook the temporal evolution of deep rock mass responses, limiting their predictive accuracy under complex geological conditions. In this context, we implement a fully integrated GNSS–borehole monitoring system to obtain high-frequency continuous GNSS observations and internal deformation time series at three key depths (14 m, 92 m, and 132 m). To reveal the dynamic correlation between strata deformations and surface subsidence across multiple time scales, cross-wavelet transform (XWT) analysis is applied to quantify both amplitude and phase relationships in the time–frequency domain. The results demonstrate that surface subsidence consistently lags behind deep rock mass deformation, with the deepest monitored stratum (132 m) showing the earliest and largest deformation. The 92 m layer (primary subsidence deformation zone) also displays a leading response, particularly in high-frequency bands, indicating its role in stress redistribution and transmission. In contrast, the shallow 14 m loess layer exhibits a lagging and hydrologically sensitive behavior, responding passively to overlying subsidence. These results highlight the stratified and frequency-dependent nature of deformation evolution, emphasizing the significance of deep rock mass signals as early indicators of subsidence progression. By integrating multi-depth deformation monitoring with time–frequency correlation analysis, this study provides novel insights into the temporal hierarchy of mining-induced subsidence. It provides theoretical support for refining subsidence prediction models and early warning systems. Compared with previous studies that focus primarily on surface or single-depth data, our approach provides a more comprehensive framework for interpreting the spatiotemporal dynamics of stratified deformation processes in mining areas.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"6 1","pages":"Article 100391"},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146154209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-01DOI: 10.1016/j.eqrea.2025.100374
Bang Zhang , Yu-Qi Cai , Zi-Ye Yu , Kai Li
Ground-Penetrating Radar (GPR) technology, with its characteristics of being fast, non-destructive, and high-resolution, has become an important tool for detecting underground structures. However, GPR data inevitably suffer from environmental noise and electromagnetic interference during the acquisition process, leading to decreased data quality and increased complexity in data processing. Traditional filtering algorithms have limitations such as low discrimination between noise and signal, poor adaptability, and inability to process data in real time. This paper proposes a filtering model based on deep neural networks, called FilterNet. FilterNet combines Convolution Neural Networks (CNN) and recurrent neural networks (RNN) for processing multi-channel data. It can perform end-to-end filtering directly on the raw tunnel lining GPR data, achieving functions such as removing air reflection waves, denoising, and automatic gain. Using PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index) as statistical indicators, it is shown that the FilterNet model improves filtering precision. The SSIM of all three models is 0.997, and the PSNR of FilterNet1D and FilterNet are 19.06 and 19.41, respectively. Furthermore, tests on the model's processing efficiency indicate that FilterNet requires less memory and is more efficient than the UNet model. FilterNet's parameters are only 48 % of those of UNet. Its GFLOPS (Giga Floating Point Operations Per Second) is only one-third of UNet's, and it can process data in real time. Additionally, FilterNet performs exceptionally well in suppressing random noise.
{"title":"FilterNet: A CNN-RNN based filter model used for raw tunnel lining GPR data","authors":"Bang Zhang , Yu-Qi Cai , Zi-Ye Yu , Kai Li","doi":"10.1016/j.eqrea.2025.100374","DOIUrl":"10.1016/j.eqrea.2025.100374","url":null,"abstract":"<div><div>Ground-Penetrating Radar (GPR) technology, with its characteristics of being fast, non-destructive, and high-resolution, has become an important tool for detecting underground structures. However, GPR data inevitably suffer from environmental noise and electromagnetic interference during the acquisition process, leading to decreased data quality and increased complexity in data processing. Traditional filtering algorithms have limitations such as low discrimination between noise and signal, poor adaptability, and inability to process data in real time. This paper proposes a filtering model based on deep neural networks, called FilterNet. FilterNet combines Convolution Neural Networks (CNN) and recurrent neural networks (RNN) for processing multi-channel data. It can perform end-to-end filtering directly on the raw tunnel lining GPR data, achieving functions such as removing air reflection waves, denoising, and automatic gain. Using PSNR (Peak Signal-to-Noise Ratio) and SSIM (Structural Similarity Index) as statistical indicators, it is shown that the FilterNet model improves filtering precision. The SSIM of all three models is 0.997, and the PSNR of FilterNet1D and FilterNet are 19.06 and 19.41, respectively. Furthermore, tests on the model's processing efficiency indicate that FilterNet requires less memory and is more efficient than the UNet model. FilterNet's parameters are only 48 % of those of UNet. Its GFLOPS (Giga Floating Point Operations Per Second) is only one-third of UNet's, and it can process data in real time. Additionally, FilterNet performs exceptionally well in suppressing random noise.</div></div>","PeriodicalId":100384,"journal":{"name":"Earthquake Research Advances","volume":"5 4","pages":"Article 100374"},"PeriodicalIF":0.0,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145420106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号