Wenwen Zhang, Yongqian Zhang, Qingtian Lü, Yutao Shi, Yao Xu, Jiayong Yan
Abstract Intracontinental deformation is out of the theory of conventional plate tectonics. It is widely recognized with deformation within the continental interior instead of the plate margin, yet its formation mechanism has long been controversial. The eastern Sichuan–Wuling mountains (ESWM) area is located ∼1300 km away from the subduction plate boundary and had developed intracontinental deformations, including crustal shortening and fold-and-thrust (FAT) tectonics, making it an ideal place to understand the mechanism of intracontinental deformation. In this study, we obtain a new seismic image of the 3D crustal structure of the ESWM area using the continuous ambient noise data of 67 broadband seismic stations. We invert the Rayleigh-wave dispersions of 5–30 s derived from cross-correlating the Z-component of all station pairs and obtain the fine crustal VS model. Our new seismic image reveals distinct velocity characteristics between the thin-skinned chevron anticline FAT tectonics in the eastern Sichuan basin and the thick-skinned chevron syncline FAT tectonics in the Wuling mountains area. Specifically, a low-VS layer observed beneath the Wuling mountains area, together with the crystalline basement beneath the eastern Sichuan basin, marks the ductile décollements confining the folding and thrusting deformation. Based on our new VS model and some previous studies, we propose a geodynamic model, which is associated with the far-field effect of the westward paleo-Pacific subduction during the late Mesozoic. Our model meets all the structural investigations at surface and geophysical observations at depth, and is reliable and valuable for further studies on similar intracontinental deformation in other regions.
{"title":"New Seismic Imaging of the Crustal Structure beneath the Eastern Sichuan and Wuling Mountains, South China: Insights into the Formation of Fold-and-Thrust Belts","authors":"Wenwen Zhang, Yongqian Zhang, Qingtian Lü, Yutao Shi, Yao Xu, Jiayong Yan","doi":"10.1785/0220230105","DOIUrl":"https://doi.org/10.1785/0220230105","url":null,"abstract":"Abstract Intracontinental deformation is out of the theory of conventional plate tectonics. It is widely recognized with deformation within the continental interior instead of the plate margin, yet its formation mechanism has long been controversial. The eastern Sichuan–Wuling mountains (ESWM) area is located ∼1300 km away from the subduction plate boundary and had developed intracontinental deformations, including crustal shortening and fold-and-thrust (FAT) tectonics, making it an ideal place to understand the mechanism of intracontinental deformation. In this study, we obtain a new seismic image of the 3D crustal structure of the ESWM area using the continuous ambient noise data of 67 broadband seismic stations. We invert the Rayleigh-wave dispersions of 5–30 s derived from cross-correlating the Z-component of all station pairs and obtain the fine crustal VS model. Our new seismic image reveals distinct velocity characteristics between the thin-skinned chevron anticline FAT tectonics in the eastern Sichuan basin and the thick-skinned chevron syncline FAT tectonics in the Wuling mountains area. Specifically, a low-VS layer observed beneath the Wuling mountains area, together with the crystalline basement beneath the eastern Sichuan basin, marks the ductile décollements confining the folding and thrusting deformation. Based on our new VS model and some previous studies, we propose a geodynamic model, which is associated with the far-field effect of the westward paleo-Pacific subduction during the late Mesozoic. Our model meets all the structural investigations at surface and geophysical observations at depth, and is reliable and valuable for further studies on similar intracontinental deformation in other regions.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135859127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Computing theoretical seismograms from a point source in a given Earth model is essential for modeling and inversion of observed seismic waveforms for Earth’s structure and earthquake source parameters. Here, we derived the propagator matrices and source terms for a spherical multilayered Earth model using the exact earth flattening transformation. We found that their differences from their counterparts in horizontal layered media are inversely proportional to the nondimensional horizontal wavenumber and its higher order. In addition, all the source terms in a spherical layered model have a source-depth dependent scaling factor that differs from in a horizontal layered model by up to 6% for deep earthquakes. The surface displacement produced by a point source can be obtained in a similar form as in horizontal layered media. Computation of theoretical seismograms was implemented using the generalized reflection and transmission coefficients method. Numerical tests show that our formulae and implementation are correct and efficient for computing full-wave seismograms, including the permanent displacements, at teleseismic distances up to 100°. Individual seismic phases can be isolated and analyzed semianalytically because the generalized reflection and transmission method is used. Furthermore, our analytic expression of displacement in terms of the propagator matrices and source terms can be used to derive analytic derivatives of seismograms for full-wave waveform inversion.
{"title":"Computing Theoretical Seismograms from a Point Source in a Spherical Multilayered Medium","authors":"Shaoqian Hu, Lupei Zhu","doi":"10.1785/0220230173","DOIUrl":"https://doi.org/10.1785/0220230173","url":null,"abstract":"Computing theoretical seismograms from a point source in a given Earth model is essential for modeling and inversion of observed seismic waveforms for Earth’s structure and earthquake source parameters. Here, we derived the propagator matrices and source terms for a spherical multilayered Earth model using the exact earth flattening transformation. We found that their differences from their counterparts in horizontal layered media are inversely proportional to the nondimensional horizontal wavenumber and its higher order. In addition, all the source terms in a spherical layered model have a source-depth dependent scaling factor that differs from in a horizontal layered model by up to 6% for deep earthquakes. The surface displacement produced by a point source can be obtained in a similar form as in horizontal layered media. Computation of theoretical seismograms was implemented using the generalized reflection and transmission coefficients method. Numerical tests show that our formulae and implementation are correct and efficient for computing full-wave seismograms, including the permanent displacements, at teleseismic distances up to 100°. Individual seismic phases can be isolated and analyzed semianalytically because the generalized reflection and transmission method is used. Furthermore, our analytic expression of displacement in terms of the propagator matrices and source terms can be used to derive analytic derivatives of seismograms for full-wave waveform inversion.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135858230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Elaine K. Young, Michael E. Oskin, Alba M. Rodriguez Padilla
Abstract We use multiple, independently produced surface-rupture maps of the 2019 Ridgecrest earthquake sequence to test the reproducibility of surface-rupture map interpretation and completeness. The 4 July Mw 6.4 and 5 July Mw 7.1 earthquakes produced surface-rupture zones approximately 20 and 50 km in length, respectively. Three independent mappers with various backgrounds in active tectonics mapped the surface rupture from the postearthquake lidar data without knowledge from postearthquake field or geodetic observations. Visual comparisons of the three remote rupture maps show good agreement for scarps >50 cm in height. For features with less topographic expression, interpretations of the data vary more widely between mappers. Quantitative map comparisons range from 18% to 54% consistency between mapped lines with 1 m buffers. The percent overlap increases with buffer width, reflecting variance in line placement as well as differences in fault-zone interpretation. Overall, map similarity is higher in areas where the surface rupture was simpler and had more vertical offset than in areas with complex rupture patterns or little vertical offset. Fault-zone interpretation accounts for the most difference between maps, while line placement accounts for differences at the meter scale. In comparison to field observations, our remotely produced maps capture the principal rupture well but miss small features and geometric complexity. In general, lidar excels for the detection and measurement of vertical offsets in the landscape, and it is deficient for detecting lateral offset with little or no vertical motion.
{"title":"Reproducibility of Remote Mapping of the 2019 Ridgecrest Earthquake Surface Ruptures","authors":"Elaine K. Young, Michael E. Oskin, Alba M. Rodriguez Padilla","doi":"10.1785/0220230095","DOIUrl":"https://doi.org/10.1785/0220230095","url":null,"abstract":"Abstract We use multiple, independently produced surface-rupture maps of the 2019 Ridgecrest earthquake sequence to test the reproducibility of surface-rupture map interpretation and completeness. The 4 July Mw 6.4 and 5 July Mw 7.1 earthquakes produced surface-rupture zones approximately 20 and 50 km in length, respectively. Three independent mappers with various backgrounds in active tectonics mapped the surface rupture from the postearthquake lidar data without knowledge from postearthquake field or geodetic observations. Visual comparisons of the three remote rupture maps show good agreement for scarps >50 cm in height. For features with less topographic expression, interpretations of the data vary more widely between mappers. Quantitative map comparisons range from 18% to 54% consistency between mapped lines with 1 m buffers. The percent overlap increases with buffer width, reflecting variance in line placement as well as differences in fault-zone interpretation. Overall, map similarity is higher in areas where the surface rupture was simpler and had more vertical offset than in areas with complex rupture patterns or little vertical offset. Fault-zone interpretation accounts for the most difference between maps, while line placement accounts for differences at the meter scale. In comparison to field observations, our remotely produced maps capture the principal rupture well but miss small features and geometric complexity. In general, lidar excels for the detection and measurement of vertical offsets in the landscape, and it is deficient for detecting lateral offset with little or no vertical motion.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135858462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Genevieve L. Coffey, Chris Rollins, Russ J. Van Dissen, David A. Rhoades, Matthew C. Gerstenberger, Nicola J. Litchfield, Kiran K. S. Thingbaijam
Abstract Recurrence intervals of ground-surface rupturing earthquakes are one of numerous datasets used to constrain the rates of fault ruptures in the 2022 revision of the New Zealand National Seismic Hazard Model (NZ NSHM 2022). Paleoearthquake timing and single-event displacement (SED) data in the New Zealand Paleoseismic Site Database version 1.0 alongside geologic and geodetic slip rates from the New Zealand Community Fault Model version 1.0 and NZ NSHM 2022 Geodetic Deformation Model were used to estimate recurrence intervals on faults across New Zealand for inclusion in the NZ NSHM 2022. Past earthquake timings were fit with lognormal, exponential, and Brownian Passage Time recurrence models to derive probability density functions (PDFs) of mean recurrence interval (MRI) in a Bayesian framework. At some sites, SED and slip-rate (SR) data were used to estimate PDFs of MRI; and at sites where timings, slip rate, and displacement data are available, the timings-based and slip-based PDFs were combined to develop tighter constraints on MRI. Using these approaches, we produce a database of maximum-likelihood MRIs and their uncertainties for 80 sites across New Zealand. The resulting recurrence interval dataset is publicly available and is the largest such dataset in New Zealand to date. It provides a valuable resource for future seismic hazard modeling and highlights areas that would benefit from future study.
{"title":"Paleoseismic Earthquake Recurrence Interval Derivation for the 2022 Revision of the New Zealand National Seismic Hazard Model","authors":"Genevieve L. Coffey, Chris Rollins, Russ J. Van Dissen, David A. Rhoades, Matthew C. Gerstenberger, Nicola J. Litchfield, Kiran K. S. Thingbaijam","doi":"10.1785/0220230197","DOIUrl":"https://doi.org/10.1785/0220230197","url":null,"abstract":"Abstract Recurrence intervals of ground-surface rupturing earthquakes are one of numerous datasets used to constrain the rates of fault ruptures in the 2022 revision of the New Zealand National Seismic Hazard Model (NZ NSHM 2022). Paleoearthquake timing and single-event displacement (SED) data in the New Zealand Paleoseismic Site Database version 1.0 alongside geologic and geodetic slip rates from the New Zealand Community Fault Model version 1.0 and NZ NSHM 2022 Geodetic Deformation Model were used to estimate recurrence intervals on faults across New Zealand for inclusion in the NZ NSHM 2022. Past earthquake timings were fit with lognormal, exponential, and Brownian Passage Time recurrence models to derive probability density functions (PDFs) of mean recurrence interval (MRI) in a Bayesian framework. At some sites, SED and slip-rate (SR) data were used to estimate PDFs of MRI; and at sites where timings, slip rate, and displacement data are available, the timings-based and slip-based PDFs were combined to develop tighter constraints on MRI. Using these approaches, we produce a database of maximum-likelihood MRIs and their uncertainties for 80 sites across New Zealand. The resulting recurrence interval dataset is publicly available and is the largest such dataset in New Zealand to date. It provides a valuable resource for future seismic hazard modeling and highlights areas that would benefit from future study.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135923259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract On 14 August 2021, an Mw 7.2 earthquake struck Nippes, Haiti, 11 yr after the devastating 2010 Mw 7.0 Port-au-Prince earthquake. This earthquake occurred in a remote region where the structure at the depth of the main boundary Enriquillo Plantain Garden fault (EPGF) is less known. Using Synthetic Aperture Radar imagery, we retrieve the coseismic and early postseismic deformation of the 2021 Haiti earthquake to constrain its fault geometry and slip distribution. Our modeling results show that the 2021 earthquake ruptured the high-angle Ravine du Sud fault and a bend fault ∼64° dipping to the north at depth. Although not only conclusive, the combination of coseismic and postseismic deformation, along with geomorphic features, and relocated aftershocks, suggest a nonplanar fault structure with significant variations in dip angles along both the depth and track of the EPGF in this region. East of the epicenter, we document a 25 km section along the EPGF that crept for ∼15 days. This distribution of aseismic slip utilizing stacked deformation indicates that only a small fraction of the accumulated strain near the surface was released during the earthquake, suggesting a high potential for seismic hazard in the region along the EPGF from the ruptured segment to the east, before reaching the 2010 rupture.
{"title":"Coseismic and Early Postseismic Slip of the 2021 Mw 7.2 Nippes, Haiti, Earthquake: Transpressional Rupture of a Nonplanar Dipping Fault System","authors":"Zhen Li, Teng Wang","doi":"10.1785/0220230160","DOIUrl":"https://doi.org/10.1785/0220230160","url":null,"abstract":"Abstract On 14 August 2021, an Mw 7.2 earthquake struck Nippes, Haiti, 11 yr after the devastating 2010 Mw 7.0 Port-au-Prince earthquake. This earthquake occurred in a remote region where the structure at the depth of the main boundary Enriquillo Plantain Garden fault (EPGF) is less known. Using Synthetic Aperture Radar imagery, we retrieve the coseismic and early postseismic deformation of the 2021 Haiti earthquake to constrain its fault geometry and slip distribution. Our modeling results show that the 2021 earthquake ruptured the high-angle Ravine du Sud fault and a bend fault ∼64° dipping to the north at depth. Although not only conclusive, the combination of coseismic and postseismic deformation, along with geomorphic features, and relocated aftershocks, suggest a nonplanar fault structure with significant variations in dip angles along both the depth and track of the EPGF in this region. East of the epicenter, we document a 25 km section along the EPGF that crept for ∼15 days. This distribution of aseismic slip utilizing stacked deformation indicates that only a small fraction of the accumulated strain near the surface was released during the earthquake, suggesting a high potential for seismic hazard in the region along the EPGF from the ruptured segment to the east, before reaching the 2010 rupture.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136013334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Giovanna Calderoni, Luigi Improta, Rita Di Giovambattista
Abstract We investigate the variability of Brune stress drop (Δσ), apparent stress (τa), and Savage–Wood radiation efficiency (ηsw=τa/Δσ), in the 2013–2014 Mw 5.0 earthquake sequence that struck the Matese area in the southern Apennines range of Italy. The sequence is clustered in a relatively small crustal volume in the 13–22 km depth range, which is greater than that of background seismicity and normal-faulting sequences that occurred under the range axis, usually located in the first 15 km of the crust. We find high Savage–Wood radiation efficiency values for most of the analyzed earthquakes located in a narrow crustal volume, with values ranging from well above the self-similarity value to very high values as high as 0.55. In addition, a large variability in radiation efficiency (up to 90%) is observed for two similar magnitude events at different depths. Previous studies reported seismic evidence of fluid involvement in the nucleation process of the Matese earthquakes. By integrating our results with crustal geophysical data published recently, we propose that most of the earthquakes characterized by high values of ηsw are nucleated within high pore pressure zones located in the crystalline midcrust of Adria. We reckon that high pore pressure fluids of deep origin played a role in the rupture process and were responsible for the mixed shear-tensile sources inferred from the analysis of the S-wave/P-wave spectral amplitude ratio for most of 2013–2014 earthquakes.
{"title":"Investigating the Role of Fluids in the Source Parameters of the 2013–2014 Mw 5 Matese Seismic Sequence, Southern Italy","authors":"Giovanna Calderoni, Luigi Improta, Rita Di Giovambattista","doi":"10.1785/0220230046","DOIUrl":"https://doi.org/10.1785/0220230046","url":null,"abstract":"Abstract We investigate the variability of Brune stress drop (Δσ), apparent stress (τa), and Savage–Wood radiation efficiency (ηsw=τa/Δσ), in the 2013–2014 Mw 5.0 earthquake sequence that struck the Matese area in the southern Apennines range of Italy. The sequence is clustered in a relatively small crustal volume in the 13–22 km depth range, which is greater than that of background seismicity and normal-faulting sequences that occurred under the range axis, usually located in the first 15 km of the crust. We find high Savage–Wood radiation efficiency values for most of the analyzed earthquakes located in a narrow crustal volume, with values ranging from well above the self-similarity value to very high values as high as 0.55. In addition, a large variability in radiation efficiency (up to 90%) is observed for two similar magnitude events at different depths. Previous studies reported seismic evidence of fluid involvement in the nucleation process of the Matese earthquakes. By integrating our results with crustal geophysical data published recently, we propose that most of the earthquakes characterized by high values of ηsw are nucleated within high pore pressure zones located in the crystalline midcrust of Adria. We reckon that high pore pressure fluids of deep origin played a role in the rupture process and were responsible for the mixed shear-tensile sources inferred from the analysis of the S-wave/P-wave spectral amplitude ratio for most of 2013–2014 earthquakes.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136352913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract The Anninghe fault (ANHF), located in southwest China, was a major block boundary that hosted M 7.5 earthquakes historically. For seismic hazard assessment, it is critical to investigate fault properties before future earthquakes. To investigate the fault structure, we deployed three linear dense arrays with an aperture of ∼8–9 km across different segments of the ANHF from October 2019 to March 2020. More importantly, we detonated a new methane source to generate seismic waves, which is environmentally friendly and can be used in different regions such as mountainous and urban areas. After data acquisition, we first removed the noise to accurately pick up the first arrivals of seismic waves. Then, we conducted the first-arrival seismic tomography, a method commonly used in the petroleum industry, to obtain the high-resolution P-wave velocity structure. The tomographic results showed distinct low-velocity zones (LVZs) of ∼1000–1500 m in width and ∼300–400 m in depth along the fault, well consistent with the lateral distribution of site amplification that was derived from regional earthquake waveforms. These LVZs may have formed as a combined result of the fault damage zone and ANHF-controlled sediments. As the Anning River Valley is densely populated, our newly identified LVZs shed lights on earthquake hazard in the region. In addition, we demonstrate that using a combination of methane detonation sources, linear dense arrays, and active source tomography can effectively determine the shallow P-wave velocity model in complex environments (i.e., mountains and urban areas).
{"title":"High-Resolution Shallow Structure along the Anninghe Fault Zone, Sichuan, China, Constrained by Active Source Tomography","authors":"Xinru Mu, Junhao Song, Hongfeng Yang, Jianping Huang, Huajian Yao, Baofeng Tian","doi":"10.1785/0220230137","DOIUrl":"https://doi.org/10.1785/0220230137","url":null,"abstract":"Abstract The Anninghe fault (ANHF), located in southwest China, was a major block boundary that hosted M 7.5 earthquakes historically. For seismic hazard assessment, it is critical to investigate fault properties before future earthquakes. To investigate the fault structure, we deployed three linear dense arrays with an aperture of ∼8–9 km across different segments of the ANHF from October 2019 to March 2020. More importantly, we detonated a new methane source to generate seismic waves, which is environmentally friendly and can be used in different regions such as mountainous and urban areas. After data acquisition, we first removed the noise to accurately pick up the first arrivals of seismic waves. Then, we conducted the first-arrival seismic tomography, a method commonly used in the petroleum industry, to obtain the high-resolution P-wave velocity structure. The tomographic results showed distinct low-velocity zones (LVZs) of ∼1000–1500 m in width and ∼300–400 m in depth along the fault, well consistent with the lateral distribution of site amplification that was derived from regional earthquake waveforms. These LVZs may have formed as a combined result of the fault damage zone and ANHF-controlled sediments. As the Anning River Valley is densely populated, our newly identified LVZs shed lights on earthquake hazard in the region. In addition, we demonstrate that using a combination of methane detonation sources, linear dense arrays, and active source tomography can effectively determine the shallow P-wave velocity model in complex environments (i.e., mountains and urban areas).","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136294647","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Predicting surface-wave travel-time shifts is valuable for analyzing potential effects caused by changes in medium properties, station clock errors, instrument response errors, and other factors. Many current neural networks used in seismology are single-station models trained using single-station (pair) data. However, most seismic methods require knowledge of the spatial positions between multiple stations. Multiple stations contain rich interrelationships and spatial information that cannot be exploited by single-station models. We proposed a multistation neural network structure Transformer Graph Convolutional Network (TGCN) that utilizes temporal attention and spatial attention to capture spatiotemporal information for predicting relative travel-time shifts. Before that, we introduced a method that treats station pairs as nodes and constructs a graph with multiple station pairs. We collected original ambient noise waveforms from 2017 to 2019 in the Alaska region and 2010 to 2014 in the southern California region to obtain relative travel-time shift sequences of station pairs for model training and testing. To showcase the improvement of spatial information to the model, we compared TGCN with two other baseline single-station models—temporal convolutional network and long short-term memory. Our proposed method predicted travel-time values more accurately than the two baseline models, and it also exhibited slower decay in performance when predicting over larger intervals. We also found that the number of station pairs has an impact on the model. When there are a sufficient number of station pairs, the model can effectively utilize the rich spatial information and achieve higher accuracy. Our approach, which incorporates spatiotemporal information, provides outputs that are more efficient and accurate compared with the traditional single-station (pair) method that only considers temporal information, suggesting that spatial information does enhance the performance of the model.
{"title":"Transformer Graph Convolutional Network for Relative Travel-Time Shift Prediction","authors":"Chunwei Jin, Fang Ye, Jinhui Cai, Yan Yao","doi":"10.1785/0220230158","DOIUrl":"https://doi.org/10.1785/0220230158","url":null,"abstract":"Abstract Predicting surface-wave travel-time shifts is valuable for analyzing potential effects caused by changes in medium properties, station clock errors, instrument response errors, and other factors. Many current neural networks used in seismology are single-station models trained using single-station (pair) data. However, most seismic methods require knowledge of the spatial positions between multiple stations. Multiple stations contain rich interrelationships and spatial information that cannot be exploited by single-station models. We proposed a multistation neural network structure Transformer Graph Convolutional Network (TGCN) that utilizes temporal attention and spatial attention to capture spatiotemporal information for predicting relative travel-time shifts. Before that, we introduced a method that treats station pairs as nodes and constructs a graph with multiple station pairs. We collected original ambient noise waveforms from 2017 to 2019 in the Alaska region and 2010 to 2014 in the southern California region to obtain relative travel-time shift sequences of station pairs for model training and testing. To showcase the improvement of spatial information to the model, we compared TGCN with two other baseline single-station models—temporal convolutional network and long short-term memory. Our proposed method predicted travel-time values more accurately than the two baseline models, and it also exhibited slower decay in performance when predicting over larger intervals. We also found that the number of station pairs has an impact on the model. When there are a sufficient number of station pairs, the model can effectively utilize the rich spatial information and achieve higher accuracy. Our approach, which incorporates spatiotemporal information, provides outputs that are more efficient and accurate compared with the traditional single-station (pair) method that only considers temporal information, suggesting that spatial information does enhance the performance of the model.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135480714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Wenhuan Kuang, Congcong Yuan, Zhihui Zou, Jie Zhang, Wei Zhang
Abstract Recent advances in artificial intelligence allow seismologists to upgrade the workflow for locating earthquakes. The standard workflow concatenates a sequence of data processing modules, including event detection, phase picking, association, and event location, with elaborately fine-tuned parameters, lacking automation and convenience. Here, we leverage deep reinforcement learning and develop a state-of-the-art earthquake robot (EQBot) to help advance automated earthquake location. The EQBot learns from tremendous trial-and-error explorations, which aims to best align the observed P and S waves, complying with the geophysical principle of gather alignments in source imaging. After training on earthquakes (M ≥ 2.0) for a decade in the Los Angeles region, it can locate earthquakes directly from waveforms with mean absolute errors of 1.32 km, 1.35 km, and 1.96 km in latitude, longitude, and depth, respectively, closely comparable to the cataloged locations. Moreover, it can automatically implement quality control by examining the alignments of P and S waves. Our study provides a new solution to advance the earthquake location process toward full automation.
{"title":"Autonomous Earthquake Location via Deep Reinforcement Learning","authors":"Wenhuan Kuang, Congcong Yuan, Zhihui Zou, Jie Zhang, Wei Zhang","doi":"10.1785/0220230118","DOIUrl":"https://doi.org/10.1785/0220230118","url":null,"abstract":"Abstract Recent advances in artificial intelligence allow seismologists to upgrade the workflow for locating earthquakes. The standard workflow concatenates a sequence of data processing modules, including event detection, phase picking, association, and event location, with elaborately fine-tuned parameters, lacking automation and convenience. Here, we leverage deep reinforcement learning and develop a state-of-the-art earthquake robot (EQBot) to help advance automated earthquake location. The EQBot learns from tremendous trial-and-error explorations, which aims to best align the observed P and S waves, complying with the geophysical principle of gather alignments in source imaging. After training on earthquakes (M ≥ 2.0) for a decade in the Los Angeles region, it can locate earthquakes directly from waveforms with mean absolute errors of 1.32 km, 1.35 km, and 1.96 km in latitude, longitude, and depth, respectively, closely comparable to the cataloged locations. Moreover, it can automatically implement quality control by examining the alignments of P and S waves. Our study provides a new solution to advance the earthquake location process toward full automation.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"33 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135689216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Monitoring the temporal variation in seismic velocity plays a critical role in understanding the dynamic processes of the subsurface at different scales. Many seismic velocity changes related to earthquakes and volcanic activities have been obtained using ambient noise correlation in recent years; however, their temporal resolution is limited, typically from a few to dozens of days, which makes it challenging to explore the valuable but short-duration changes in subsurface media. In this article, we develop a method based on the correlation of the coda of the ambient noise correlation (C3) with a multiple-component combination and introduced singular value decomposition-based Wiener filter denoising technique. Using permanent network data, we achieved subdaily ambient noise monitoring at Parkfield, California, using 4-hr cross-correlation stacking with 2-hr step. We identified that the maximum seismic velocity drop delayed the mainshock of the 2004 Mw 6.0 Parkfield earthquake by ∼41 hr, during which the temporal velocity process may have been affected by strong aftershocks, including an Mw 5.0 aftershock that occurred one day after the mainshock; however, no significant precursory change was detected. Our method provides an opportunity for monitoring the short-term change of underground structures based on the widely distributed seismic networks. In addition, the idea of obtaining reliable subsurface information within a short time through high-order noise correlation in this work has important enlightenment for ambient noise imaging and monitoring in broader fields.
{"title":"Subdaily Ambient Noise Monitoring at Parkfield, California, by Combining C1 and C3","authors":"Yi Meng, Zhikun Liu, Tiancheng Li, Rui Zhang","doi":"10.1785/0220230119","DOIUrl":"https://doi.org/10.1785/0220230119","url":null,"abstract":"Abstract Monitoring the temporal variation in seismic velocity plays a critical role in understanding the dynamic processes of the subsurface at different scales. Many seismic velocity changes related to earthquakes and volcanic activities have been obtained using ambient noise correlation in recent years; however, their temporal resolution is limited, typically from a few to dozens of days, which makes it challenging to explore the valuable but short-duration changes in subsurface media. In this article, we develop a method based on the correlation of the coda of the ambient noise correlation (C3) with a multiple-component combination and introduced singular value decomposition-based Wiener filter denoising technique. Using permanent network data, we achieved subdaily ambient noise monitoring at Parkfield, California, using 4-hr cross-correlation stacking with 2-hr step. We identified that the maximum seismic velocity drop delayed the mainshock of the 2004 Mw 6.0 Parkfield earthquake by ∼41 hr, during which the temporal velocity process may have been affected by strong aftershocks, including an Mw 5.0 aftershock that occurred one day after the mainshock; however, no significant precursory change was detected. Our method provides an opportunity for monitoring the short-term change of underground structures based on the widely distributed seismic networks. In addition, the idea of obtaining reliable subsurface information within a short time through high-order noise correlation in this work has important enlightenment for ambient noise imaging and monitoring in broader fields.","PeriodicalId":21687,"journal":{"name":"Seismological Research Letters","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135689215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：