Founded in 1948, Acta Geophysical Sinica (ISSN: 0001-5733) records the glorious progress of the development of continental geophysical science. Jointly sponsored by the Institute of Geology and Geophysics of the Chinese Academy of Sciences and the Chinese Geophysical Society, this journal is a comprehensive academic journal related to geophysical science. It mainly publishes research papers in solid geophysics, applied geophysics, space geophysics and atmospheric, marine geophysics, and interdisciplinary disciplines closely related to geophysics. Emphasis is placed on reporting on innovative research results.
Chinese Journal of Geophysics is one of the representative journals in the field of geosciences and has been listed as a core journal of natural sciences in China. At present, there are a number of well-known international search systems to search this journal, such as SCI, SA, CA, All-Russian Institute of Scientific Information Abstracts Journal, Japanese Science and Technology Abstracts Flash, World Index of Translations, Ulicy International Journal Guide, Water Science Abstracts, TULSA Petroleum Abstracts, American Geological Abstracts, French Literature Bulletin, etc. This journal is included in relevant domestic retrieval institutions, such as "China Science and Technology Paper Statistical Source Database", "China Science Citation Index Database", etc.
Journal Catalog
What to see
Space Physics and Geodesy
01
Luo Taorong et al.: Three-dimensional data assimilation modeling of electrons in extraterrestrial radiation belts based on Fortran language
Highlights:
The development of a data assimilation model of the Earth's electron radiation belt is of great significance for understanding the dynamic evolution process of radiation belt electrons and the spatial weather prediction of the radiation belt. Combined with the electronic observation data of the Van Allen satellite and the three-dimensional diffusion model of the outer radiation belt, and using the Kalman filter algorithm, a three-dimensional data assimilative model of Outer radiation belt electrons (TDAMORE) based on Fortran language is developed to achieve the effect of L*=3~7, Three-dimensional reconstruction of the spatiotemporal variation process of electrons in the outer radiation band with an energy range of 0.1~5 MeV and a throw angle range of 5°~90°. By reconstructing the evolution process of electron flux in the outer radiation band during August 2018, it is confirmed that the TDAMORE model can well reproduce the evolution characteristics of electron flux with different energies and different throwing angles before and after magnetic storms. By analyzing the correlation coefficient, mean error, mean absolute error and mean square error between the observed and assimilation results of the electron flux, it is found that for electrons with an energy of less than 4 MeV, the correlation coefficient between the observed and assimilation results is basically greater than 0.8 and the error is relatively low. For higher energy electrons, the error between the observation and the assimilation results is relatively high, which may be caused by the assimilation model ignoring the scattering loss of electrons by the electromagnetic ion cyclotron wave.
And download the PDF file for free
02
Liu Xinming et al.: Numerical study of the anisotropy of positrons of galactic cosmic rays induced by solar modulation
Highlights:
When the positrons of galactic cosmic rays with energies lower than 30 GeV enter the heliosphere, they will interact with the solar wind and the heliospheric magnetic field frozen in it, and the combined modulation effects of convection, diffusion and drift of the solar wind will lead to anisotropy in the positron flux of galactic cosmic rays. In this study, the Parker transport equation was used to describe the positron transport process in the heliosphere, the local interstellar energy spectrum at the heliospheric apex (120 AU) was used as the boundary condition, the transmission equation was solved by the alternating direction implicit method (ADI), the flux of the positron of the galactic cosmic rays was calculated, and then the gradient of the positrons of the galactic cosmic rays was calculated, and finally the polar anisotropy and radial anisotropy of the positrons with energies of 0.01 GeV, 0.1 GeV, and 1 GeV were calculated. The results show that: (1) Since the drift of the low-energy positron is almost zero, the polar anisotropy is determined only by diffusion, and its absolute value is equal at the same latitude in the northern and southern hemispheres of the heliosphere and 0 at the ecliptic plane; (2) The radial anisotropy of low-energy positrons is determined only by diffusion and convection, while the radial anisotropy of high-energy positrons is determined by diffusion, convection, and drift. In addition, the polar gradient of the positron at the ecliptic plane is 0, so the radial anisotropy of the positron of any energy here is determined only by diffusion and convection, and the positron has a large radial anisotropy due to the large diffusion and drift in the high latitudes of the heliosphere.
And download the PDF file for free
03
Hu Anxun et al.: Spatial and temporal distribution of glacial albedo in the Greater Caucasus Mountains and its relationship with mass balance
Highlights:
Based on the MOD10A1 and MYD10A1 ice and snow albedo data, the temporal and spatial distribution characteristics of the albedo of the Galabasi Glacier and the Deyankut Glacier in the Greater Caucasus Mountains were analyzed, and the correlation between the albedo and mass balance of the two glaciers in the summer of 2002-2019 was studied by using five albedo aggregation methods. The results show that the albedo of Galabrasi Glacier and Deyancourt Glacier decreases significantly with the increase of year, the lowest albedo of both glaciers in summer, the highest albedo of Galabasi Glacier in winter, and the highest albedo of Deyankut Glacier in spring, the difference of albedo between high and low altitude areas of Galabasi Glacier in summer is more obvious than that of Deyankut Glacier, and the summer albedo of both glaciers shows a certain positive correlation with the summer mass balance, and the correlation coefficient of Galabasi Glacier is more significant, and the correlation coefficient between the two glaciers can reach 0.874 (P<0.05) when the average minimum albedo method is used , the maximum correlation coefficient between the two is 0.765 (P<0.05) when the weighted average albedo method is used in Deyangkut Glacier. The reason for the low correlation between albedo and mass balance in Deyankut Glacier may be that the slope and low altitude make it more susceptible to warm and humid air currents, and the short-wave radiant energy in glacier ablation is relatively low, and the albedo has little influence on its ablation.
And download the PDF file for free
04
Wu Yilin et al.: Using Machine Learning Methods to Extract Hydrocarbon Microleakage Anomalies from Sentinel-2 Images: Taking Marsel Exploration Area as an Example
Highlights:
Hydrocarbon micro-leakage in oil and gas reservoirs is a common phenomenon, and remote sensing technology provides an efficient and fast method for surface detection of hydrocarbon micro-leakage in oil and gas reservoirs. The traditional method based on hydrocarbon micro-leakage mechanism is simple but multi-solution by detecting the response of surface anomalies (vegetation, alteration minerals, etc.) caused by hydrocarbon micro-leakage. In this paper, a machine learning-based extraction method for anomalous hydrocarbon microleakage from remote sensing images was proposed. Firstly, the training samples were made based on the surface microbial detection results in the study area, and in order to compare the learning results of different samples, the patch sample dataset and pixel sample dataset were made respectively, and on this basis, the learning models of two types of datasets were constructed by logistic regression, support vector machine, random forest, LeNet, AlexNet, GoogLeNet, and ResNet algorithms. The results show that for the classical machine learning algorithm, the highest accuracy of the plaque sample is 0.833 and the highest pixel sample is 0.771, and for the deep learning algorithm, the highest accuracy of the plaque sample is 0.782 and the highest pixel sample is 0.914. Finally, the four algorithm models with the highest accuracy are applied to the Marsel exploration area of Kazakhstan, and compared with the geological and seismic data, it is found that the prediction results of the ResNet-18-1D pixel sample have the best correspondence with the seismic geological analysis data, and the accuracy is 0.914, and the Kappa coefficient is 0.892.
And download the PDF file for free
05
Hu Wenjie et al.: The influence of tidal effects on the monitoring of surface deformation of multi-strip time series InSAR in coastal strips is Lg wave attenuation imaging in Chinese mainland
Highlights:
The influence of tidal effects on the large-scale and high-precision interferometric synthetic Aperture Radar (SAR) technology on surface deformation monitoring along the coast cannot be ignored. The displacement of solid Earth Tide (SET) and ocean tide loading (OTL) in the tidal effect will produce decimeter-level errors in the monitoring of large-scale coastal InSAR surface deformation in the temporal and spatial domains, and the temporal and spatial differences of surface tidal deformation in different bands will produce large errors in the stitching of InSAR images in multiple bands. In this paper, we focus on the analysis of the influence of tidal effects on the monitoring of InSAR surface subsidence in different bands in coastal areas, and use a variety of tidal displacement estimation methods to correct the surface tidal deformation of multiple bands. The results show that the surface tidal deformation in the time-series InSAR of different bands in the study area has a large temporal and spatial difference, resulting in a difference of 1~2 cm·a-1 in the temporal series InSAR of adjacent bands in the coastal area. The correction of tidal displacement in the ocean tide load model or GPS reference station network can eliminate the spatial high-order nonlinear errors caused by the temporal and spatial differences of surface tidal deformation in different bands on the splicing of temporal series InSAR deformation results, and make up for the shortcomings of traditional splicing methods in fitting tidal displacement deviations. On the basis of the correction of atmospheric delay error, the standard deviation (STD) of the time series InSAR deformation residuals is reduced from 2.3 cm before the tidal correction to 0.75 cm.
And download the PDF file for free
06
Wei Na et al.: A pole-shift prediction method that takes into account the effective angular momentum and IGS ultrafast solution data
Highlights:
Pole-shift high-precision forecasting is crucial for real-time satellite orbit determination and deep space probe navigation. In this paper, we propose a method to combine Effective Angular Momentum (EAM) with ultrafast solution data provided by the International GNSS Service (IGS) for pole shift forecasting. Based on the prediction value of pole shift on the first day obtained from the IGS ultrafast solution data, the pole shift prediction results obtained by introducing EAM are corrected to obtain the joint forecast value. Firstly, based on the LS (Least Squares) + AR (Auto-Regressive) model, the pole shift prediction with EAM is realized, and compared with the Bulletin A data provided by the International Earth Rotation and Reference Systems Service (IERS), it is more effective in the ultra-short term (No. 1~10 days) forecast, in which the angular momentum of the atmosphere and the ocean play a major role. Secondly, in view of the high accuracy and fast update of IGS ultrafast solution data, based on the IGS ultrafast solution data, the prediction value of the first day of extreme shift can be obtained based on the LS+AR model, which is significantly better than the prediction value of the first day of extreme shift in IERS Bulletin A. Finally, the joint forecast values are obtained by using the forecast values on the first day to correct the forecast results considering the effective angular momentum, which further improves the accuracy of ultra-short-term pole shift prediction (especially on the 1st~5th day). The results of the joint forecast from July 24, 2020 to January 30, 2022 show that the forecast value of the 1~20 days is generally better than the forecast value of IERS Bulletin A. On the first day of the forecast, the accuracy of the forecast values in the X and Y directions can be increased by 39.5%~62.3% and 24.5%~51.9% respectively compared with the forecast values of Communiqué A, and on the 10th day of forecasting, the accuracy of the forecast values relative to the forecast values of Communiqué A can be increased by 28.0%~28.9% and 21.9%~23.4%, respectively.
And download the PDF file for free
07
Zhou Shuai et al.: Multilayer sensor inversion of the seabed topography of the South China Sea fused with multi-source marine geodetic data
Highlights:
In this paper, the perpendicular deviation, gravity anomaly and vertical gravity gradient data released by the Scripps Institution of Oceanography (SIO) and the shipborne bathymetric data released by the National Centers for Environmental Information (NCEI) are fused, and the multi-layer perceptron neural network (Multi-Layer) is used Perceptron (MLP) established a seabed topography model (MLP_Depth) with a resolution of 1'×1' in the South China Sea (108°E-121°E, 6°N-23°N). Firstly, the position information of 642716 shipborne bathymetric control points and the earth gravity information (perpendicular deviation, gravity anomaly, vertical gravity gradient) at the surrounding 4'×4' grid points were used as input data, and the measured bathymetric values at the shipboard bathymetric control points were used as the output data to train the MLP neural network model, and the determination coefficient R2 was 99% and the average absolute error MAE was 39.33 m at the end of training. Then, the input data at the positive center point of the 1'×1' grid in the study area were input into the MLP model to obtain the predicted sea depth value at the positive center point of the grid. Finally, a MLP_Depth model with a resolution of 1'×1' in the study area was established according to the predicted sea depth value. Compared MLP_Depth with the measured water depth at 160679 checkpoints, the standard deviation STD (75.38 m), the mean absolute percentage error (MAPE) and the mean absolute error MAE (42.91 m) were better than those of the GEBCO_2021 model, the topo_23.1 model, the ETOPO1 model and the STD (108.88 m, 113.41 m, 229.67) of the difference between the model, the .1 model and the ETOPO1 model and the measured water depth at the checkpoints m), MAPE (6.11%, 6.94%, 18.37%) and MAE (47.33 m, 52.24 m, 130.08 m). At the same time, in order to study the accuracy of the seabed topography model established by this method in different areas, the seabed topography models of regions A and B are established in the study area. It has been verified that the MLP_Depth_A and MLP_Depth_B models have higher accuracy than the MLP_Depth models, and can better reflect the change trend of seabed topography.
And download the PDF file for free
Geodynamics, Seismology, Geomagnetism
08
Zhang Geng et al.: Using borehole strain data to detect the low-frequency free oscillation of the Earth ring and its spectral peak splitting triggered by the Tohoku earthquake in Japan in 2011
Highlights:
In earthquake-excited free oscillations of the Earth, the accurate detection of annular free oscillations (especially low-frequency ring free oscillations) is more difficult than that of spherical free oscillations. The annular free oscillation is more sensitive to the SH wave velocity structure and can better monitor the strike-slip fault movement, so the accurate estimation of the spectral peak parameters of the annular free oscillation can provide stronger constraints for the large-scale SH wave velocity structure and the source mechanism of large earthquakes in the earth's interior. In this study, based on the observation data of the four-component borehole strain gauge after the Tohoku earthquake in Japan in 2011, the low-frequency ring and spherical free oscillation signals excited by the earthquake below 1.3 mHz were extracted, and compared with the results of the STS-1 broadband seismograph, it was found that the four-component borehole strain gauge had stronger detection ability for the annular free oscillation below 1 mHz. In this study, the frequency domain autoregressive method is used to accurately estimate the degenerate frequency of 0T2-0T6, and the difference between the results and the theoretical value of the PREM model is very small, and the average relative error is only 2.52×10-4. In addition, the 0T2 peak splitting phenomenon was detected by borehole strain gauge observations. The results show that the borehole strain gauge can effectively detect the low-frequency free oscillation signal of the earth ring excited by large earthquakes, which can provide high-quality basic data for the study of large-scale anisotropic structure of the mantle.
And download the PDF file for free
09
Zhu Chenglin et al.: Research on the interference characteristics and disturbance mechanism of engineering precipitation on borehole strain
Highlights:
Pumping disturbance is common and has a significant impact on borehole strain observation, and its disturbance mechanism should be studied in order to truly capture the seismic precursor information. With the process of urbanization, the precipitation of foundation pit engineering and its accompanying engineering gradually increases, and the interference of engineering precipitation on borehole strain is more serious than that of ordinary single-well pumping interference. It is of great significance for the detection of seismic anomalies that engineering precipitation has the interference characteristics and interference mechanism of borehole strain observation, and whether the interference data can be repaired. Based on the observation case of Laiyang body strain being disturbed by the nearby engineering precipitation of the Yulongwan project, the interference characteristics of the engineering precipitation are studied, the interference distortion is repaired, and the interference mechanism is determined by numerical simulation. The results show that: (1) The engineering precipitation superimposes the solid tide distortion with clear direction and stable magnitude on the basis of the change of the strain trend of the interferor, which can be repaired by the first-order differential inverse operation. After repairing the solid tidal distortion and deducting the background rate, the interference magnitude of engineering precipitation reaches -15.65×10-9/d;(2) By establishing a grid algorithm based on the concentrated load model, the concentrated force load of engineering precipitation in the early and late stages of the bulk strain anomaly is simulated to be 6.61×109 N and 1.73×1010 N, respectively, and the interference magnitude is significantly greater than that of the common single-well pumping interference, and the pressure interference is concentrated in the vertical direction of the connection between the bulk strain gauge and the pumping well;(3) Through the analysis of the characteristics and mechanical mechanism of solid tidal distortion, it is determined that the disturbance mechanism of solid tidal distortion caused by engineering precipitation is the periodic behavior of "decreasing pressure trend caused by pumping, rising tensile step caused by pumping pause, and decreasing pressure trend caused by pumping", and there is a significant synchronization between the time interval (frequency) of solid tide distortion and the borehole water level, and the linear coefficient is 0.737.
And download the PDF file for free
10
Kou Huadong et al.: Seasonal changes and possible physical mechanisms of seismic wave travel in Binchuan Basin and surrounding areas of Yunnan Province
Highlights:
Based on the background noise autocorrelation method, this paper calculates the relative time variation (dt/t) of seismic waves in the Binchuan Basin and surrounding areas of Yunnan Province in four frequency bands, and finds that there are significant annual and seminal variation characteristics, and the maximum variation amplitude gradually decreases from 0.8% of ±0.1~0.2 Hz (the most sensitive depth of about 7 km) to ±0.05% of 1.0~2.0 Hz (corresponding to shallow medium below 1 km). Seasonal variations in noise sources may be a potential factor in the periodic fluctuations of dt/t, but rain and temperature are more directly correlated with it. The pore pressure change caused by rainwater infiltration is an important contribution factor for annual and semi-annual variation, while the thermodynamic deformation of medium caused by temperature fluctuation mainly has an obvious annual variation effect on the medium with a certain depth range (about 3 km is the most sensitive). There are amplitude differences and asynchronous changes in different frequency bands, which may be related to the depth range of sensitive cores in each frequency band and the infiltration/propagation process of water and thermodynamic deformation from the surface to the depths. The results show that environmental factors have a significant impact on the seismic wave travel time of underground media, and only by distinguishing their contributions can we more reliably evaluate the seismic tectonic stress or the changes in the human ecological environment.
And download the PDF file for free
11
Pan Yang et al.: Teleseismic time-tomography in a global heterogeneous model
Highlights:
Teleseismic travel time tomography is an important tool for studying the internal structure of the Earth. Typically, teleseismic travel time tomography uses a one-dimensional Earth reference model to calculate seismic rays outside the study area. Due to the extensive heterogeneous structure of the earth, the traditional teleseismic travel time tomography ignores the influence of the heterogeneous structure outside the study area on the radiation when calculating the rays, which ultimately leads to the impact on the imaging results in the study area. In order to reduce the influence of the heterogeneous structure of the earth on the teleseismic travel time tomography of the model, this paper uses the windward differential format numerically to solve the range function equation to obtain the teleseismic rays in the global heterogeneous model, and develops a teleseismic travel time imaging method based on the global heterogeneous model. In order to test the effectiveness of the imaging method in this paper to improve the imaging accuracy, the northeastern edge of the Qinghai-Tibet Plateau was selected as the research object. Synthetic tests and real-world imaging results show that the imaging results are more accurate and show more detail after considering the inhomogeneous structure of the Earth.
And download the PDF file for free
12
Li Xinyue et al.: Near-seismic body wave travel time and gravity jointly invert the P-wave velocity structure at the top of the mantle in the crust of North China
Highlights:
North China is a typical case of craton failure, and it is also one of the areas with the most intense and frequent intracontinental strong earthquake activity, and the study of the velocity structure of the crust and mantle in this area can provide important deep constraints for deepening the understanding of the formation and evolution of craton, intraplate strong earthquakes and volcanic mechanisms. Based on the P-wave data of natural seismic events recorded by 202 fixed stations and 360 mobile stations and the measured Bug gravity anomaly, the P-wave velocity structure of the crust and upper mantle was obtained by the joint inversion method of near-seismic body wave travel time and gravity. The introduction of gravity data not only provides more reasonable constraints on the crust-mantle structure, but also effectively improves the imaging ability of shallow and marginal ray distribution sparse regions. The joint imaging results show that the Ordos massif and the shallow layer of the North China Plain show obvious low-velocity anomalies, which may be related to the existence of thick soft sedimentary layers, and there is an obvious P-wave high-velocity anomaly in the upper crust of eastern Ordos, which is speculated to be related to the formation of the Paleoproterozoic Kongzi belt. The 15~25 km and 25~40 km in the Datong volcanic area show obvious low-velocity anomalies, and the two low-velocity layers are connected, which is speculated to be related to the multi-stage magmatic inversion caused by the upwelling of mantle thermal materials. The epicenters of the Sanhe-Pinggu magnitude 8.0 earthquake and the Xingtai magnitude 7.2 earthquake were both located in the brittle upper crust, and both were anomalous at low velocity, so it is speculated that the creep in the ductile crust leads to the local accumulation of elastic strain energy in the brittle upper crustal rocks under the loading of the regional dynamic environment, which leads to the occurrence of earthquakes.
And download the PDF file for free
13
Gao Zhanyong et al.: Study on the velocity structure of shallow three-dimensional S-wave velocity structure of the earth's crust in Guangzhou-Foshan region based on short-period dense arrays
Highlights:
The Guangzhou-Foshan region is located in the core area of the Pearl River Delta Economic Zone, through which three regional faults are staggered. The fine three-dimensional S-wave velocity structure of shallow crust in the Guangzhou-Foshan area is of great significance for urban planning and construction, seismic hazard assessment, and strong ground motion simulation in the study area. In this paper, based on the continuous waveform data recorded by 1104 stations in the short-period dense seismic array in the Guangzhou-Foshan area, the cross-correlation function between station pairs is calculated by using the background noise cross-correlation method, and 39488 high-quality surface wave group velocity dispersion curves in the period range of 0.1~5 s are further extracted, and then the high-resolution three-dimensional S-wave velocity structure at a depth of 0.3~3 km in the study area is calculated by the direct surface wave inversion method based on ray tracing. The results show that there is a significant lateral difference in the S-wave velocity in the shallow crust in the study area, and the S-wave velocity structure in the Pearl River Delta fault basin in the south is relatively complex, while the S-wave velocity structure in the Guanghua Basin in the north is relatively simple. The low-velocity anomalies in the study area are mainly located near the fault, and most of them are located in the descending disk of the fault. The eastern margin of the Sanshui Basin in the Pearl River Delta is characterized by a saddle-shaped low-velocity anomaly, extending to a maximum of 2 km downward, and the boundaries of the north and south sides of the low-velocity body are basically consistent with the location of the basin-control fault. There are two parallel high-velocity anomalies in the Guanghua Basin, and a low-velocity anomaly with a parallel structural strike in the middle, and the overall velocity structure characteristics are consistent with the geological structure background of the Guanghua Basin as a compound dip. The low-velocity anomaly extends down-dip to a depth of 1.4 km, which may indicate the maximum buried depth of the stratum involved in the inclined stratum. The development of banded low-velocity anomalies along the Guangzhou-Conghua fault is related to the sedimentary strata formed along the fault. The banded anomaly is divided into two segments in the superficial part, indicating that the Guangzhou-Conghua fault may have been segmented in history. Yudai Park is surrounded by a high-speed anomaly, which extends upwards to about 1.3 km from the surface, and it is speculated that the high-speed anomaly may be a reflection of early magmatic intrusion in the shallow part of the earth's crust.
And download the PDF file for free
14
Zhang Jianyong et al.: Seismic structure and genesis of the MS6.8, MS5.0 and MS5.6 earthquake sequences in Luding, Sichuan Province from 2022 to 2023
Highlights:
The strong MS6.8 earthquake occurred in Luding, Sichuan Province on September 5, 2022, followed by the strong aftershocks of MS5.0 and MS5.6 on October 22 and January 26, 2023, respectively. In this study, the continuous waveform data and seismic phase data of the three earthquake sequences recorded by the National Seismic Network were used to carry out double-difference relocation, and the CAP waveform inversion method and the P wave initial dynamic polarity inversion method were used to obtain the solution of 79 aftershock source mechanisms of ML2.5+ in the study area. The results show that the main rupture is along the western section of the Xianshui River fault, and the rupture is complete, and the aftershock activity is not high. The aftershocks distributed along the main fault have obvious segmented characteristics, and the fault is nearly upright, shallow in the northwest and deep in the southeast. Most of the aftershocks in the main shock and Moxi section are strike-slip mechanisms, which are typical strong seismic activities triggered by stress instability in the Xianshuihe strike-slip fault zone caused by the extrusion and rotation of the Indo-Eurasian plate. The aftershocks that occurred in the Gongga Mountain area were the earthquake swarm activities triggered by the MS6.8 mainshock, and the magnitude level was not high and the distribution was diffuse, which did not trigger the secondary faults of Yanzigou, Hailuogou and Mozigou conjugated with the main fault, but triggered the secondary hidden faults nearly parallel to the main fault, the Gonggashan fault. The two strong aftershocks of MS5.0 and MS5.6 occurred on the secondary hidden fault, with a fault dip angle of 40°~50° and a shallow source depth. The tension-type seismic activity in the Gongga Mountains area is obviously different from that of the strike-slip Xianshuihe fault zone, which may be caused by the rapid uplift of the Gongga Mountains and the rapid surface stripping under the compression and rotation of the Indo-Eurasian plate, which leads to the local collapse due to gravity unloading, and the MS6.8 mainshock has an obvious triggering effect. The occurrence of the three strong earthquakes in Luding released part of the stress in the western section of Moxi and the Gongga Mountain area on the west side, but did not change the possibility of strong earthquakes in the "Y" shaped intersection area.
And download the PDF file for free
15
Mo Lihua et al.: Study on the focal parameters of the MS6.0 earthquake in Luxian County, Sichuan Province in 2021
Highlights:
On September 16, 2021, the largest earthquake (MS6.0) on record struck Luxian County, Sichuan. In order to understand the seismic mechanism of the earthquake, we used the near-seismic and far-seismic waveform data to accurately determine the mechanism solution, depth, horizontal position, rupture direction and other source parameters of the 2021 Luxian earthquake and MS≥3.0 aftershocks. The depth of the center of mass of the MS6.0 earthquake is 4 km, the moment magnitude is MW5.25, and the strike, inclination and slip angles of the two nodes are 95°/47°/84° and 283°/43°/96°, respectively, and the waveform fitting results of the two phases of sPL and sP show that the depth of the center of mass is about 5 km. We also used the InSAR deformation data to obtain the centroid position of the MS6.0 earthquake, and then determined the horizontal positions of other earthquakes by using the relative localization method of surface wave to time difference. Based on the empirical Green's function method, the rupture direction of the MS6.0 earthquake is roughly north-west-west. In 2021, the depth of the centroid of the MS≥3.0 earthquake in Luxian County was 3~5 km, and the epicenter of the centroid was located within 3 km of the shale gas exploitation water injection well.
And download the PDF file for free
16
Wang Tongli et al.: Microseismic detection and activity study of Guye MS5.1 earthquake sequence in 2020
Highlights:
On July 12, 2020, the MS5.1 earthquake occurred in Guye (39.78°N, 118.44°E), Hebei Province, with an epicenter depth of 10 km. In this paper, the seismic detection and seismic activity analysis of the continuous data of 84 fixed stations within 500 km of the earthquake were continuously observed by using the GPU-M&L (GPU-M&L) for three months of continuous data before and after the earthquake (June 1, 2020 to August 31, 2020). Firstly, we selected 118 earthquakes with high signal-to-noise ratio from the 362 Guye earthquake sequence events in the Taiwan network directory as templates, and obtained a total of 1017 earthquakes through template scanning and detection, which is about 2.8 times that of the Taiwan network directory, and then used the double-difference localization method (hypoDD) to relocate 669 seismic events. Finally, we analyze the spatiotemporal distribution, source mechanism solution and spatial distribution of b-value of the event sequence of the Guye earthquake. The results show that before the Guye MS5.1 earthquake, the seismic activity around the epicenter area was significant, and the dominant direction of the seismic sequence distribution before and after the earthquake was NE direction, and the depth of the epicenter was nearly upright, and the seismic fault was the Tangshan-Guye fault or its extension, which was the same as the Tangshan MS7.8 earthquake in 1976, and the seismic fault trend was basically the same. Based on this, we speculate that the Guye MS5.1 earthquake was triggered by the stress adjustment in the aftershock region after the Tangshan MS7.8 earthquake in 1976.
And download the PDF file for free
17
Sun Xiaoqian et al.: The HVSR method was used to detect the velocity structure of landslide in Xishan Village, Lixian County, Sichuan
Highlights:
Horizontal-to-Vertical Spectral Ratio (HVSR) is the spectral ratio of the horizontal component to the vertical component of seismic background noise, and is often used to detect shallow surface velocity structures with small undulating terrain. However, the applicability of this method in highly undulating terrain needs to be carefully studied. In this paper, the HVSR method was used to detect the velocity structure of the landslide in Xishan Village, Lixian County, Sichuan. Firstly, we rotate the three-component seismic data in the ENZ coordinate system to the LQT coordinate system perpendicular to the free surface of the terrain, and calculate the HVSR curves of the 2-hour seismic background noise data in the two coordinate systems. Then, the HVSR curves of the LQT coordinate system of 29 stations are inverted, and the one-dimensional S-wave velocity structure of the landslide mass and bedrock is obtained. The S-wave velocity range of the landslide mass in Xishan Village is 50~1000 m·s-1, and the thickness of the landslide is 23~130 m. There is a low-velocity layer of 10~20 m at 40 ~120 m underground on the west edge of the landslide, indicating that the shear strength of this position is relatively weak and is a potential sliding danger area. After the coordinate system conversion of the data, the HVSR method can obtain a reliable landslide velocity structure by using the short observation data, which provides important information for landslide volume estimation, vulnerable location identification, and disaster assessment.
And download the PDF file for free
18
Wang Qiao et al.: Distribution and propagation characteristics of extremely low-frequency electromagnetic fields of underground electric dipole sources
Highlights:
The forward algorithm of dipole source response in layered earth is used to calculate the extremely low-frequency seismic electromagnetic field excited by the electric dipole source in the earth and analyze its propagation characteristics in the earth environment. A number of horizontal layered earth models were designed to simulate and demonstrate the response and spatial distribution characteristics of electric dipole sources in deep burial. The amplitude and frequency characteristics of the excitation response of the inclined electric dipole source with the offset distance and vertical position of the observation point, the depth of the excitation source, the structure and parameters of the crust waveguide in the high-impedance earth with crust waveguide + LAI waveguide model are analyzed. The results show that it is feasible to simulate and analyze the response and propagation characteristics of seismic electromagnetic radiation in high-impedance earth by using an inclined electric dipole source, and the simulated electromagnetic radiation in high-impedance earth exhibits slow attenuation or amplitude enhancement characteristics in both crustal waveguide and LAI waveguide, but the two waveguide effects have different frequency selectivity, and the electromagnetic radiation in high-impedance earth has the characteristics of strong amplitude and slow attenuation in the high-impedance medium of the waveguide. It is recommended to install stations on the surface, in wells or in high-resistance rock formations on the seabed, and in the atmosphere.
And download the PDF file for free
Applied Geophysics
19
Wang Xiao et al.: Research on qP wave transmission stereochromatography method for three-dimensional TTI medium
Highlights:
He is committed to the study of qP wave stereochromatography inversion algorithm in three-dimensional TTI medium. The ray perturbation theory is applied to the equation of mimetic acoustic wave function in a three-dimensional transverse isotropic (TTI) medium with an oblique axis of symmetry, and all Frechét derivatives between the three-dimensional TTI medium qP wave data space and the three-dimensional TTI media model space required for stereo tomography are derived. AFTER VERIFYING THE CORRECTNESS OF THE CORRESPONDING FRECCHÉT DERIVATIVES FOR THE THREE ANISOTROPIC THOMSEN PARAMETERS BY SENSITIVITY TESTS, THE TWO-STEP INVERSION STRATEGY WAS USED. In the case of three-dimensional qP wave TTI media, ideal inversion results can be obtained for the three anisotropic THOMSEN parameters, which is theoretically prepared for the application of transmission stereochromatography in three-dimensional qP wave TTI media. A typical theoretical data example is used to confirm the above view.
And download the PDF file for free
20
Zhang Zhaoqian et al.: Research on shale acoustic wave anisotropy correction method based on staggered grid finite difference algorithm
Highlights:
Shale has the characteristics of bedding development, which will cause strong acoustic anisotropy, resulting in obvious differences between the acoustic logging data of vertical wells and horizontal wells, so the petrophysical interpretation model based on vertical wells cannot be directly applied in the calculation of reservoir parameters of horizontal wells. In order to solve this problem, based on the experimental data of shale wave velocity anisotropy, this paper introduces the finite difference algorithm of staggered grid, firstly simulates the transmission and reception of sound waves in the vertical wellbore model (VTI medium), and then simulates the sound field propagation in the wellbore when the well inclination angle is not 0 (TTI medium) through the Bond transform of the elastic coefficient matrix, and the relative value of the longitudinal wave slowness difference when the arbitrary well inclination angle and the well inclination angle is 0 as the ordinate, and the corresponding well inclination angle is used as the abscissa, and the longitudinal wave anisotropy correction formula is established。 The simulation and application results show that the waveform curve in the borehole is consistent with the waveform curve obtained by the real axis integration method (RAI), and the formation longitudinal wave slowness obtained by the slowness similarity correlation algorithm (STC) is in good agreement with the given experimental measured value, and the average relative error is 2.3% under the condition of 28 formation elastic parametersWhen the inclination angle of the well is 90°, that is, in the horizontal well mode, the relative value of the longitudinal wave slowness difference reaches the maximum. According to the longitudinal wave anisotropy correction formula, the longitudinal wave curve of the horizontal well is slowly corrected. The analysis of the relative error of the effective porosity of the horizontal well and the effective porosity of the core analysis of the pilot hole calculated before and after the longitudinal wave correction shows that the calculation accuracy of the effective porosity calculated after the longitudinal wave correction has been significantly improved, which proves that the method has a very good application effect and can be used for the longitudinal wave slowness correction of shale horizontal wells.
And download the PDF file for free
21
Dai Yongshou et al.: A pre-stack time-varying wavelet extraction method based on an improved gated recurrent unit
Highlights:
In view of the limitations of various assumptions of the traditional time-varying wavelet extraction method, and the need to extract the wavelet amplitude spectrum and phase spectrum respectively, this paper proposes a pre-stack time-varying seismic wavelet extraction method based on the improved Gated Recurrent Unit (GRU) network. In order to expand the extracted time series features and improve the processing ability of the traditional GRU network for long time series, an improved gated recurrent unit network model with multi-layer GRU modules and fully connected neural networks is built, the network model is trained by using the established training dataset to make the network model have the ability to extract time-varying wavelets, and in order to improve the training efficiency and extraction accuracy, the method applies a custom WaveLoss loss function to measure the error in the backpropagation process of training, and finally realize the estimation of pre-stack time-varying wavelets. The pre-stack time-varying wavelet extraction method proposed in this paper has higher accuracy through the simulation experiments of synthetic data and different methods, and the proposed method can effectively improve the resolution of the pre-stack seismic profile in the target area through the actual pre-stack seismic data processing and deconvolution verification analysis in different areas of western China.
And download the PDF file for free
22
Yang Yang et al.: Research on reservoir thickness prediction method of lacustrine carbonate rock based on two-dimensional scattering transform
Highlights:
The lacustrine carbonate rocks of the fourth member of the Sha 4th Member of the Jiyang depression are controlled by the sedimentary environment and tectonic movement in the lake basin, and the reservoir thickness varies greatly, the heterogeneity is strong, and the oil and gas development is difficult. In order to accurately predict the thickness of lacustrine carbonate reservoirs, this paper proposes a reservoir thickness prediction method based on two-dimensional scattering transform and random forest. Firstly, the two-dimensional scattering transform is introduced to extract the seismic time-frequency attributes, which is realized on the basis of the two-dimensional wavelet transform through iterative wavelet decomposition and nonlinear operation. Compared with the traditional two-dimensional wavelet transform, the time-frequency attribute extracted by the scattering transform has the advantages of local deformation stability and robustness to noise, which is helpful to improve the accuracy of reservoir thickness prediction. On this basis, under the condition of limited logging data, the random forest algorithm is used to establish a nonlinear relationship between multi-scale time-frequency attributes and logging interpretation thickness, so as to realize the prediction of lacustrine carbonate reservoirs. The prediction results of the model data show that the thickness prediction method proposed in this paper has the best performance compared with the thickness prediction based on the traditional seismic amplitude attribute and the reservoir thickness prediction based on the two-dimensional wavelet transform. The prediction results of post-stack 3D seismic data show that compared with the thickness prediction based on the traditional seismic amplitude attribute and the reservoir thickness prediction based on the two-dimensional wavelet transform, the thickness prediction results of the proposed method in this paper have less error with the actual drilling data, which improves the accuracy of reservoir thickness prediction and clearly depicts the spatial distribution of three sedimentary subfacies, such as ash reef, ash beach and plaster, which is conducive to the subsequent well location deployment and optimization.
And download the PDF file for free
23
Xu Binsen et al.: Research on reservoir parameter evaluation based on serial and parallel multi-task learning networks
Highlights:
The choice of reservoir parameter evaluation network modeling based on serial or parallel multi-task learning is a recent problem. Based on the experimental data of reservoir parameters in an old area of an oilfield in western China, this paper compares the evaluation results of porosity, saturation and permeability of a multi-task learning network composed of 20 different basic neuron modules, and proposes a model selection strategy: the determination coefficient is the evaluation index, and the condition for selecting the serial rather than parallel structure network model is the reference value of the model parameter quantity less than 1000, and the serial multi-task network is not as good as the parallel multi-task network when the model parameter quantity is greater than 1000. When the average absolute error is the evaluation index, the premise of serial multi-task is that the number of model parameters is less than 10000 reference values. When the number of model parameters is greater than 10,000, the results of serial and parallel multi-task networks are similar. If the average absolute error and the amount of model parameters are within the allowable range, both architecture networks are feasible. The purpose of this paper is to provide support for the design and application of different types of multi-task learning network architecture models in the future.
And download the PDF file for free
24
Zhao Guangdong et al.: Three-dimensional gravity field inversion method in the frequency domain based on L1+L2 mixed norm
Highlights:
Three-dimensional gravity inversion is widely used in mineral resource exploration and deep density structure research, but the traditional three-dimensional gravity field inversion method in space domain has the problems of low depth resolution, strong multi-solution and low computational efficiency, which affects the accuracy and credibility of geological interpretation. In view of the above shortcomings, this paper proposes a frequency-domain three-dimensional gravity field inversion method based on the regularization constraint of mixed norms. Firstly, the objective function of the model based on depth weighting and L1+L2 mixed norm is constructed. In the process of inversion iteration, the frequency-domain 3D gravity field forward modeling method is used to update the model, and the storage and calculation of dense Yacobian matrices in the traditional 3D gravity field inversion in the space domain are transformed into frequency-domain forward modeling, which greatly reduces the memory occupation. In addition, the high-precision Gaussian numerical integration is used to replace the rectangular integration in the traditional Fast Fourier Transform (FFT) algorithm to ensure the calculation accuracy of the frequency-domain forward method. The model test results show that compared with the traditional spatial domain based on L2 norm smooth constraint inversion method and L1 norm focused inversion method, the proposed inversion method can effectively reduce the "skin effect" and tailing phenomenon of 3D gravity inversion, and can recover more complex geological models, and the inversion results are closer to the real model. Finally, the method is applied to the three-dimensional density imaging of the Mobrun sulphide orebody in Noranda, Quebec, Canada, and the results show that the depth range of the orebody is about 15~170 m, which is consistent with the depth range obtained from the previous drilling data, which proves the effectiveness of the method.
And download the PDF file for free
25
Hu Shuanggui et al.: Geological body positioning using gravity gradient tensor isoplane curvature
Highlights:
For the study of gravity gradient tensor curvature, the previous work mainly focused on the interpretation of gravity tensor curvature and edge detection, and almost did not involve the location of subsurface density anomalies. In this paper, a set of subsurface density anomaly position estimation strategies based on gravity gradient tensor isotope curvature based on gravity gradient tensor and gravity gradient tensor are proposed. Firstly, starting from the basic definition of the curvature of the gravitational gradient tensor isotope, the aritope curvature of the gravitational gradient tensor is calculated. Then, by finding the gravitational gradient tensor curvature of the spherical or circular isotope, a source parameter estimation method for locating the location of the subsurface density anomaly by using the maximum principal curvature is proposed, and the analytical expression for estimating the source position information of the 3D sphere (particle) and the 2D horizontal line is derived in detail. Furthermore, in view of the existence of noise and multiple sources, a robust estimation process for obtaining the position information of density anomalies by using gravity gradient tensor isotope curvature is proposed, and the fuzzy C-means clustering algorithm is used to further determine the center position of underground density anomalies. Finally, the feasibility and reliability of the proposed algorithm are verified by the theoretical model test and the measured airborne gravity gradient data of Winton Salt Hills. The results show that the proposed method can locate the spatial location of single or multiple underground 3D and 2D density anomalies under the condition that the radius of curvature definition condition is satisfied, and has good robustness and anti-noise ability. This method expands the application range of gravity gradient tensor curvature, and can provide a priori spatial position information for the three-dimensional inversion of gravity gradient tensor.
And download the PDF file for free
26
Tao Mengli et al.: Three-dimensional forward modeling of airborne electromagnetism in the frequency domain based on multi-scale finite element method
Highlights:
Forward simulation is the basis of electromagnetic data inversion, and its calculation speed and accuracy have always been the two core problems restricting electromagnetic inversion. In the 3D electromagnetic forward and inversion, the traditional method improves the calculation accuracy by encrypting the grid or increasing the interpolation basis function order, but it also reduces the calculation efficiency and restricts the practical application of 3D electromagnetic inversion. Therefore, how to achieve high-precision and fast forward modeling of large-scale models is an urgent problem to be solved in the current electromagnetic 3D forward and inversion modeling. In this paper, the multi-scale finite element method is applied to the solution of Maxwell's equations. Firstly, we construct a multi-scale basis function satisfying the local characteristic differential operator on the coarse grid scale, and then solve the original problem on the coarse grid scale, by establishing the mapping relationship between the two sets of thick and thin grids, and then solve the electromagnetic problem on the coarse grid with few unknowns, and then use the mapping relationship between the two sets of thick and thin grids to obtain the electromagnetic field response on the fine grid, so that the forward response of the electromagnetic field at different scales can be quickly obtained under the premise of ensuring the calculation accuracy, and the calculation speed is greatly improved. In addition, this paper also optimizes the mesh based on the octree idea to further improve the efficiency of 3D forward modeling. The effectiveness of the algorithm is verified by multi-scale finite element forward simulation of typical geoelectric structures and compared with the traditional finite element results. Finally, we test the ability of the proposed algorithm to simulate complex underground anomalies by simulating the airborne electromagnetic response of the ovoid nickel-copper sulfide mining area in Voisey's Bay, Canada.
And download the PDF file for free
27
Yang Yongyou et al.: Research on high-temperature directional inclinometer while drilling based on MEMS gyroscope
Highlights:
In view of the challenges faced by oil and gas wells in the measurement of oil and gas wells, such as small aperture, ultra-deep wells, wide temperature range, high temperature change rate and long working time, a directional inclinometer based on MEMS gyroscope was studied, and the software and hardware system of MEMS gyroscope directional inclinometer was designed. In order to ensure the measurement accuracy in this environment, a dynamic temperature compensation scheme was proposed, and the method based on wavelet filtering multi-scale analysis was used to denoise and filter the gyroscope output signal in the dynamic temperature change environment to achieve the measurement accuracy of the directional inclinometer system. In this paper, the hardware system scheme of directional inclinometer, directional inclinometer algorithm, and wavelet multi-scale filtering algorithm are given. The test results show that the accuracy of the well inclination angle is better than ±2°, the accuracy of the well inclination angle is better than ±0.1°, and the maximum working temperature is 150°C under the normal temperature and dynamic temperature change environment, which can meet the application scenario requirements of exploration and development of deep oil and gas wells in mainland China.
And download the PDF file for free
THE
END
Typesetting: Xu Bingbing (Institute of Geology and Geosciences, Chinese Academy of Sciences)
Proofreader: Qiu Zihui (School of Earth and Space Sciences, University of Science and Technology of China)