Late at night on 17 June 2019,a magnitude 6.0 earthquake struck Shuanghe Town and its surrounding area in Changning County,Sichuan,China,becoming the largest earthquake recorded within the southern Sichuan Basin.A ser...Late at night on 17 June 2019,a magnitude 6.0 earthquake struck Shuanghe Town and its surrounding area in Changning County,Sichuan,China,becoming the largest earthquake recorded within the southern Sichuan Basin.A series of earthquakes with magnitudes up to 5.6 occurred during a short period after the mainshock,and we thus refer to these earthquakes as the Changning M6 earthquake sequence(or swarm).The mainshock was located very close to a salt mine,into which for^3 decades fresh water had been extensively injected through several wells at a depth of 2.7–3 km.It was also near(within^15 km)the epicenter of the 18 December 2018 M5.7 Xingwen earthquake,which is thought to have been induced by shale gas hydraulic fracturing(HF),prompting questions about the possible involvement of industrial activities in the M6 sequence.Following previous studies,this paper focuses on the relationship between injection and seismicity in the Shuanghe salt field and its adjacent Shangluo shale gas block.Except for a period of serious water loss after the start of cross-well injection in 2005–2006,the frequency of earthquakes shows a slightly increasing tendency.Overall,there is a good correlation between the event rate in the Shuanghe area and the loss of injected water.More than 400 M≥3 earthquakes,including 40 M≥4 and 5 M≥5 events,had been observed by the end of August 2019.Meanwhile,in the Shangluo area,seismicity has increased during drilling and HF operations(mostly in vertical wells)since about 2009,and dramatically since the end of 2014,coincident with the start of systematic HF in the area.The event rate shows a progressively increasing background with some fluctuations,paralleling the increase in HF operations.More than 700 M≥3 earthquakes,including 10 M≥4 and 3 M≥5 in spatially and temporally clustered seismic events,are correlated closely with active fracturing platforms.Well-resolved centroid moment tensor results for M≥4 earthquakes were shown to occur at very shallow depths around shale formations with 展开更多
The vertical structure of the crustal block of the Songliao Basin can be divided into upper, middle and low Earth's crust according to density. There is an about 3-km-thick low density interval between the upper c...The vertical structure of the crustal block of the Songliao Basin can be divided into upper, middle and low Earth's crust according to density. There is an about 3-km-thick low density interval between the upper crust and the middle crust. This interval may be a magma chamber accumulated in crust by 'fluid phase' which is precipitated and separated from upper mantle meltmass. The abiogenetic natural gas, other gaseous mass and hydrothermal fluids are provided to the Songliao rifted basin through crustal faults and natural earthquakes. This is a basic condition to form an abiogenetic gas reservoir in the Songliao Basin. On both flanks of the upper crust (or named basin basement) fault there are structural traps in and above the basement and unconformity surface or lateral extended sand, which contains communicated pores, as migration pathway and natural gas reservoir; up to gas reservoirs there is shale as enclosed cap rock, and the suitable arrangement of these conditions is the basic features of abiogenetic gas reservoir.展开更多
The Panxi region is located in the frontal zone of positive squeezing subduction and side squeezing shearing between the Indian plate and the Eurasian plate. The long-period magnetotelluric （LMT） and broadband magne...The Panxi region is located in the frontal zone of positive squeezing subduction and side squeezing shearing between the Indian plate and the Eurasian plate. The long-period magnetotelluric （LMT） and broadband magnetotelluric （MT） techniques are both used to study the deep electrical conductivity structure in this region; magnetic and gravity surveys are also performed along the profile. According to the 2-D resistivity model along the Yanyuan-Yongshan profile, a high- conductivity layer （HCL） exists widely in the crust, and a high-resistivity block （HRB） exists widely in the upper mantle in general, as seen by the fact that a large HCL exists from the western Jinpingshan tectonic zone to the eastern Mabian tectonic zone in the crust, while the HRB found in the Panxi tectonic zone is of abnormally high resistivity in that background compared to both sides of Panxi tectonic zone. In addition, the gravity and magnetic field anomalies are of high value. Combined with geological data, the results indicate that there probably exists basic or ultrabasic rock with a large thickness in the lithosphere in the Panxi axial region, which indicates that fracture activity once occurred in the lithosphere. As a result, we can infer that the high-resistivity zone in the Panxi lithosphere is the eruption channel for Permian Emeishan basalt and the accumulation channel for basic and ultrabasic rock. The seismic sources along the profile are counted according to seismic record data. The results indicate that the most violent earthquake sources are located at the binding site of the HRB and the HCL, where the tectonic activity zone is generally acknowledged to be; however, the earthquakes occurring in the HCL are not so violent, which reflects the fact that the HCL is a plastic layer, and the fracture threshold of a plastic layer is low generally, making high stress difficult to accumulate but easy to release in the layer. As a result, a higher number of smaller earthquakes occurred in the HCL at Daliangshan tectonic展开更多
PhaseNet and EQTransformer are two state-of-the-art earthquake detection methods that have been increasingly applied worldwide.To evaluate the generaliz-ation ability of the two models and provide insights for the dev...PhaseNet and EQTransformer are two state-of-the-art earthquake detection methods that have been increasingly applied worldwide.To evaluate the generaliz-ation ability of the two models and provide insights for the development of new models,this study took the sequences of the Yunnan Yangbi M6.4 earthquake and Qinghai Maduo M7.4 earthquake as examples to compare the earthquake detection effects of the two abovementioned models as well as their abilities to process dense seismic sequences.It has been demonstrated from the corresponding research that due to the differences in seismic waveforms found in different geographical regions,the picking performance is reduced when the two models are applied directly to the detection of the Yangbi and Maduo earthquakes.PhaseNet has a higher recall than EQTransformer,but the recall of both models is reduced by 13%-56%when compared with the results rep-orted in the original papers.The analysis results indicate that neural networks with deeper layers and complex structures may not necessarily enhance earthquake detection perfor-mance.In designing earthquake detection models,attention should be paid to not only the balance of depth,width,and architecture but also to the quality and quantity of the training datasets.In addition,noise datasets should be incorporated during training.According to the continuous waveforms detected 21 days before the Yangbi and Maduo earthquakes,the Yangbi earthquake exhibited foreshock,while the Maduo earthquake showed no foreshock activity,indicating that the two earthquakes’nucleation processes were different.展开更多
Seismic phase pickers based on deep neural networks have been extensively used recently,demonstrating their advantages on both performance and efficiency.However,these pickers are trained with and applied to different...Seismic phase pickers based on deep neural networks have been extensively used recently,demonstrating their advantages on both performance and efficiency.However,these pickers are trained with and applied to different data.A comprehensive benchmark based on a single dataset is therefore lacking.Here,using the recently released DiTing dataset,we analyzed performances of seven phase pickers with different network structures,the efficiencies are also evaluated using both CPU and GPU devices.Evaluations based on F1-scores reveal that the recurrent neural network(RNN)and EQTransformer exhibit the best performance,likely owing to their large receptive fields.Similar performances are observed among PhaseNet(UNet),UNet++,and the lightweight phase picking network(LPPN).However,the LPPN models are the most efficient.The RNN and EQTransformer have similar speeds,which are slower than those of the LPPN and PhaseNet.UNet++requires the most computational effort among the pickers.As all of the pickers perform well after being trained with a large-scale dataset,users may choose the one suitable for their applications.For beginners,we provide a tutorial on training and validating the pickers using the DiTing dataset.We also provide two sets of models trained using datasets with both 50 Hz and 100 Hz sampling rates for direct application by end-users.All of our models are open-source and publicly accessible.展开更多
Fast and accurate P-wave arrival picking significantly affects the performance of earthquake early warning(EEW)systems.Automated P-wave picking algorithms used in EEW have encountered problems of falsely picking up no...Fast and accurate P-wave arrival picking significantly affects the performance of earthquake early warning(EEW)systems.Automated P-wave picking algorithms used in EEW have encountered problems of falsely picking up noise,missing P-waves and inaccurate P-wave arrival estimation.To address these issues,an automatic algorithm based on the convolution neural network(DPick)was developed,and trained with a moderate number of data sets of 17,717 accelerograms.Compared to the widely used approach of the short-term average/long-term average of signal characteristic function(STA/LTA),DPick is 1.6 times less likely to detect noise as a P-wave,and 76 times less likely to miss P-waves.In terms of estimating P-wave arrival time,when the detection task is completed within 1 s,DPick′s detection occurrence is 7.4 times that of STA/LTA in the 0.05 s error band,and 1.6 times when the error band is 0.10 s.This verified that the proposed method has the potential for wide applications in EEW.展开更多
基金the State Scholarship Fund of China (No. 201804190004)
文摘Late at night on 17 June 2019,a magnitude 6.0 earthquake struck Shuanghe Town and its surrounding area in Changning County,Sichuan,China,becoming the largest earthquake recorded within the southern Sichuan Basin.A series of earthquakes with magnitudes up to 5.6 occurred during a short period after the mainshock,and we thus refer to these earthquakes as the Changning M6 earthquake sequence(or swarm).The mainshock was located very close to a salt mine,into which for^3 decades fresh water had been extensively injected through several wells at a depth of 2.7–3 km.It was also near(within^15 km)the epicenter of the 18 December 2018 M5.7 Xingwen earthquake,which is thought to have been induced by shale gas hydraulic fracturing(HF),prompting questions about the possible involvement of industrial activities in the M6 sequence.Following previous studies,this paper focuses on the relationship between injection and seismicity in the Shuanghe salt field and its adjacent Shangluo shale gas block.Except for a period of serious water loss after the start of cross-well injection in 2005–2006,the frequency of earthquakes shows a slightly increasing tendency.Overall,there is a good correlation between the event rate in the Shuanghe area and the loss of injected water.More than 400 M≥3 earthquakes,including 40 M≥4 and 5 M≥5 events,had been observed by the end of August 2019.Meanwhile,in the Shangluo area,seismicity has increased during drilling and HF operations(mostly in vertical wells)since about 2009,and dramatically since the end of 2014,coincident with the start of systematic HF in the area.The event rate shows a progressively increasing background with some fluctuations,paralleling the increase in HF operations.More than 700 M≥3 earthquakes,including 10 M≥4 and 3 M≥5 in spatially and temporally clustered seismic events,are correlated closely with active fracturing platforms.Well-resolved centroid moment tensor results for M≥4 earthquakes were shown to occur at very shallow depths around shale formations with
基金Project supported by the National Natural Science Foundation of China
文摘The vertical structure of the crustal block of the Songliao Basin can be divided into upper, middle and low Earth's crust according to density. There is an about 3-km-thick low density interval between the upper crust and the middle crust. This interval may be a magma chamber accumulated in crust by 'fluid phase' which is precipitated and separated from upper mantle meltmass. The abiogenetic natural gas, other gaseous mass and hydrothermal fluids are provided to the Songliao rifted basin through crustal faults and natural earthquakes. This is a basic condition to form an abiogenetic gas reservoir in the Songliao Basin. On both flanks of the upper crust (or named basin basement) fault there are structural traps in and above the basement and unconformity surface or lateral extended sand, which contains communicated pores, as migration pathway and natural gas reservoir; up to gas reservoirs there is shale as enclosed cap rock, and the suitable arrangement of these conditions is the basic features of abiogenetic gas reservoir.
基金supported by National High-Tech R&D Program of China (Grant 2014AA06A612)the project of the China Geological Survey (Grants 1212011220263,1212010914049 and 1212011121273)
文摘The Panxi region is located in the frontal zone of positive squeezing subduction and side squeezing shearing between the Indian plate and the Eurasian plate. The long-period magnetotelluric （LMT） and broadband magnetotelluric （MT） techniques are both used to study the deep electrical conductivity structure in this region; magnetic and gravity surveys are also performed along the profile. According to the 2-D resistivity model along the Yanyuan-Yongshan profile, a high- conductivity layer （HCL） exists widely in the crust, and a high-resistivity block （HRB） exists widely in the upper mantle in general, as seen by the fact that a large HCL exists from the western Jinpingshan tectonic zone to the eastern Mabian tectonic zone in the crust, while the HRB found in the Panxi tectonic zone is of abnormally high resistivity in that background compared to both sides of Panxi tectonic zone. In addition, the gravity and magnetic field anomalies are of high value. Combined with geological data, the results indicate that there probably exists basic or ultrabasic rock with a large thickness in the lithosphere in the Panxi axial region, which indicates that fracture activity once occurred in the lithosphere. As a result, we can infer that the high-resistivity zone in the Panxi lithosphere is the eruption channel for Permian Emeishan basalt and the accumulation channel for basic and ultrabasic rock. The seismic sources along the profile are counted according to seismic record data. The results indicate that the most violent earthquake sources are located at the binding site of the HRB and the HCL, where the tectonic activity zone is generally acknowledged to be; however, the earthquakes occurring in the HCL are not so violent, which reflects the fact that the HCL is a plastic layer, and the fracture threshold of a plastic layer is low generally, making high stress difficult to accumulate but easy to release in the layer. As a result, a higher number of smaller earthquakes occurred in the HCL at Daliangshan tectonic
基金funded by the National Key R&D Program of China(No.2021YFC3000702)the National Natural Science Foundation of China(No.41774067)the Fundamental Research Funds for the Institute of Geophysics,China Earthquake Administration(Nos.DQ JB21Z05,DQJB20X07).
文摘PhaseNet and EQTransformer are two state-of-the-art earthquake detection methods that have been increasingly applied worldwide.To evaluate the generaliz-ation ability of the two models and provide insights for the development of new models,this study took the sequences of the Yunnan Yangbi M6.4 earthquake and Qinghai Maduo M7.4 earthquake as examples to compare the earthquake detection effects of the two abovementioned models as well as their abilities to process dense seismic sequences.It has been demonstrated from the corresponding research that due to the differences in seismic waveforms found in different geographical regions,the picking performance is reduced when the two models are applied directly to the detection of the Yangbi and Maduo earthquakes.PhaseNet has a higher recall than EQTransformer,but the recall of both models is reduced by 13%-56%when compared with the results rep-orted in the original papers.The analysis results indicate that neural networks with deeper layers and complex structures may not necessarily enhance earthquake detection perfor-mance.In designing earthquake detection models,attention should be paid to not only the balance of depth,width,and architecture but also to the quality and quantity of the training datasets.In addition,noise datasets should be incorporated during training.According to the continuous waveforms detected 21 days before the Yangbi and Maduo earthquakes,the Yangbi earthquake exhibited foreshock,while the Maduo earthquake showed no foreshock activity,indicating that the two earthquakes’nucleation processes were different.
基金jointly supported by the National Natural Science Foundation of China (No. 42074060)the Special Fund, Institute of Geophysics, China Earthquake Administration (CEA-IGP) (Nos. DQJB19B29, DQJB20B15, and DQJB22Z01)supported by XingHuo Project, CEA (No. XH211103)
文摘Seismic phase pickers based on deep neural networks have been extensively used recently,demonstrating their advantages on both performance and efficiency.However,these pickers are trained with and applied to different data.A comprehensive benchmark based on a single dataset is therefore lacking.Here,using the recently released DiTing dataset,we analyzed performances of seven phase pickers with different network structures,the efficiencies are also evaluated using both CPU and GPU devices.Evaluations based on F1-scores reveal that the recurrent neural network(RNN)and EQTransformer exhibit the best performance,likely owing to their large receptive fields.Similar performances are observed among PhaseNet(UNet),UNet++,and the lightweight phase picking network(LPPN).However,the LPPN models are the most efficient.The RNN and EQTransformer have similar speeds,which are slower than those of the LPPN and PhaseNet.UNet++requires the most computational effort among the pickers.As all of the pickers perform well after being trained with a large-scale dataset,users may choose the one suitable for their applications.For beginners,we provide a tutorial on training and validating the pickers using the DiTing dataset.We also provide two sets of models trained using datasets with both 50 Hz and 100 Hz sampling rates for direct application by end-users.All of our models are open-source and publicly accessible.
基金National Natural Science Foundation of China under Grant Nos.51968016 and 5197083806the Guangxi Innovation Driven Development Project(Science and Technology Major Project,Grant No.Guike AA18118008).
文摘Fast and accurate P-wave arrival picking significantly affects the performance of earthquake early warning(EEW)systems.Automated P-wave picking algorithms used in EEW have encountered problems of falsely picking up noise,missing P-waves and inaccurate P-wave arrival estimation.To address these issues,an automatic algorithm based on the convolution neural network(DPick)was developed,and trained with a moderate number of data sets of 17,717 accelerograms.Compared to the widely used approach of the short-term average/long-term average of signal characteristic function(STA/LTA),DPick is 1.6 times less likely to detect noise as a P-wave,and 76 times less likely to miss P-waves.In terms of estimating P-wave arrival time,when the detection task is completed within 1 s,DPick′s detection occurrence is 7.4 times that of STA/LTA in the 0.05 s error band,and 1.6 times when the error band is 0.10 s.This verified that the proposed method has the potential for wide applications in EEW.
