Pierre-Luc St-Charles, Bruno Rousseau, Joumana Ghosn, Gilles Bellefleur, Ernst Schetselaar
Deep learning techniques are used to tackle a variety of tasks related to seismic data processing and interpretation. While many works have shown the benefits of deep learning, assessing the generalization capabilities of proposed methods to data acquired in different conditions and geological environments remains challenging. This is especially true for applications in hardrock environments where seismic surveys are still relatively rare. The primary factors that impede the adoption of machine learning in geosciences include the lack of publicly available and labeled datasets, and the use of inadequate evaluation methodologies. Since machine learning models are prone to overfit and underperform when the data used to train them is site-specific, the applicability of these models on new survey data that could be considered “out-of-distribution” is rarely addressed. This is unfortunate, as evaluating predictive models in out-of-distribution settings can provide a good insight into their usefulness in real-world use cases. To tackle these issues, we propose a simple benchmarking methodology for first break picking to evaluate the transferability of deep learning models that are trained across different environments and acquisition conditions. For this, we consider a reflection seismic survey dataset acquired at five distinct hardrock mining sites combined with annotations for first break picking. We train and evaluate a baseline deep learning solution based on a U-Net for future comparisons, and discuss potential improvements to this approach.
{"title":"Deep Learning Benchmark for First Break Detection from Hardrock Seismic Reflection Data","authors":"Pierre-Luc St-Charles, Bruno Rousseau, Joumana Ghosn, Gilles Bellefleur, Ernst Schetselaar","doi":"10.1190/geo2022-0741.1","DOIUrl":"https://doi.org/10.1190/geo2022-0741.1","url":null,"abstract":"Deep learning techniques are used to tackle a variety of tasks related to seismic data processing and interpretation. While many works have shown the benefits of deep learning, assessing the generalization capabilities of proposed methods to data acquired in different conditions and geological environments remains challenging. This is especially true for applications in hardrock environments where seismic surveys are still relatively rare. The primary factors that impede the adoption of machine learning in geosciences include the lack of publicly available and labeled datasets, and the use of inadequate evaluation methodologies. Since machine learning models are prone to overfit and underperform when the data used to train them is site-specific, the applicability of these models on new survey data that could be considered “out-of-distribution” is rarely addressed. This is unfortunate, as evaluating predictive models in out-of-distribution settings can provide a good insight into their usefulness in real-world use cases. To tackle these issues, we propose a simple benchmarking methodology for first break picking to evaluate the transferability of deep learning models that are trained across different environments and acquisition conditions. For this, we consider a reflection seismic survey dataset acquired at five distinct hardrock mining sites combined with annotations for first break picking. We train and evaluate a baseline deep learning solution based on a U-Net for future comparisons, and discuss potential improvements to this approach.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"5 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":"136353105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Elastic parameters such as P- and S-wave velocity and density are of great significance for subsurface quantitative interpretation and reservoir prediction. Current pre-stack amplitude-versus-angle (AVA) inversion methods have been widely used in industry to obtain subsurface elastic parameters. Conventional AVA inversion methods are theoretically based on a linearized physical model formulating the relationship between pre-stack seismic reflection coefficients and subsurface model elastic parameters, called physical model-driven inversion. However, the linearized physical models lead to low accuracy and high uncertainty of inversion results. In recent years, several neural network-based pre-stack AVA inversion methods, called data-driven inversion, have been developed to address this issue. But these methods typically require a large amount of labeled data for training network, and the process does not have a clear physical mechanism. So the data-driven inversion results lack physical interpretability. To address these issues, a joint data- and physics-driven inversion of pre-stack AVA elastic parameters is proposed. Under the framework of semi-supervised learning, a two-dimensional convolutional neural network and a recurrent neural network are used to establish the mapping between several adjacent pre-stack AVA gathers and one-dimensional elastic parameters in time domain. The full Zoeppritz equation is used as a physical model constraint to the neural network, and loss functions are constructed using both well-logging data and pre-stack AVA seismic data. This approach can perform training network using small labeled data and increase physical interpretability of the inversion process. The inverse distance weighted correlation coefficient of seismic data is proposed to weight the loss function of seismic data and well-logging data. Synthetic and field data examples show that the joint data- and physics-driven pre-stack AVA elastic parameters inversion improves the accuracy and resolution, and provides an estimation of uncertainty of the inversion results.
{"title":"Joint Data- and Physics-driven Pre-stack AVA Elastic Parameters Inversion","authors":"Shuliang Wu, Yingying Wang, Qingping Li, Zhiliang He, Jianhua Geng","doi":"10.1190/geo2023-0135.1","DOIUrl":"https://doi.org/10.1190/geo2023-0135.1","url":null,"abstract":"Elastic parameters such as P- and S-wave velocity and density are of great significance for subsurface quantitative interpretation and reservoir prediction. Current pre-stack amplitude-versus-angle (AVA) inversion methods have been widely used in industry to obtain subsurface elastic parameters. Conventional AVA inversion methods are theoretically based on a linearized physical model formulating the relationship between pre-stack seismic reflection coefficients and subsurface model elastic parameters, called physical model-driven inversion. However, the linearized physical models lead to low accuracy and high uncertainty of inversion results. In recent years, several neural network-based pre-stack AVA inversion methods, called data-driven inversion, have been developed to address this issue. But these methods typically require a large amount of labeled data for training network, and the process does not have a clear physical mechanism. So the data-driven inversion results lack physical interpretability. To address these issues, a joint data- and physics-driven inversion of pre-stack AVA elastic parameters is proposed. Under the framework of semi-supervised learning, a two-dimensional convolutional neural network and a recurrent neural network are used to establish the mapping between several adjacent pre-stack AVA gathers and one-dimensional elastic parameters in time domain. The full Zoeppritz equation is used as a physical model constraint to the neural network, and loss functions are constructed using both well-logging data and pre-stack AVA seismic data. This approach can perform training network using small labeled data and increase physical interpretability of the inversion process. The inverse distance weighted correlation coefficient of seismic data is proposed to weight the loss function of seismic data and well-logging data. Synthetic and field data examples show that the joint data- and physics-driven pre-stack AVA elastic parameters inversion improves the accuracy and resolution, and provides an estimation of uncertainty of the inversion results.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"76 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":"136352780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Paleokarst systems, found in carbonate rock formations worldwide, have potential for creating vast reservoirs and facilitating hydrocarbon migration. Thus, studying these systems is essential for the exploration and development of carbonate reservoirs. Our proposed approach is to use a convolutional neural network (CNN) based method to automatically and precisely identify cave features within 3D seismic data. We present an efficient method to produce ample amounts of 3D training data, which is comprised of synthetic seismic data and labels for cave features contained in the seismic data, as a solution to bypass the labeling task for training the CNN. This workflow uses point-spread functions (PSFs) to simulate cave response in the seismic data and allows us to easily generate realistic and diverse synthetic training datasets with different geological structures and cave features. By training the CNN with these synthetic datasets, it can effectively learn to detect cave features in field seismic volumes. We have evaluated the effectiveness of our method using multiple examples and found that it performs more accurately than previous methods, including seismic attributes and other CNN-based paleokarst characterization methods.
{"title":"Paleokarst caves recognition from seismic response simulation to CNN detection","authors":"Donglin Zhu, Rui Guo, Xiangwen Li, Lei Li, Shifan Zhan, Chunfeng Tao, Yingnan Gao","doi":"10.1190/geo2023-0133.1","DOIUrl":"https://doi.org/10.1190/geo2023-0133.1","url":null,"abstract":"Paleokarst systems, found in carbonate rock formations worldwide, have potential for creating vast reservoirs and facilitating hydrocarbon migration. Thus, studying these systems is essential for the exploration and development of carbonate reservoirs. Our proposed approach is to use a convolutional neural network (CNN) based method to automatically and precisely identify cave features within 3D seismic data. We present an efficient method to produce ample amounts of 3D training data, which is comprised of synthetic seismic data and labels for cave features contained in the seismic data, as a solution to bypass the labeling task for training the CNN. This workflow uses point-spread functions (PSFs) to simulate cave response in the seismic data and allows us to easily generate realistic and diverse synthetic training datasets with different geological structures and cave features. By training the CNN with these synthetic datasets, it can effectively learn to detect cave features in field seismic volumes. We have evaluated the effectiveness of our method using multiple examples and found that it performs more accurately than previous methods, including seismic attributes and other CNN-based paleokarst characterization methods.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135093627","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Katerina Polychronopoulou, Michal Malinowski, Marta Cyz, Nikos Martakis, George Apostolopoulos, Deyan Draganov
As the global need for aluminum constantly rises, bauxite is considered to be a critical mineral, and the mining industry is in search of new and effective exploration solutions. In this context, we designed and implemented a purely earthquake-based passive seismic survey at the Gerolekas bauxite mining site, in Greece. It is a very difficult exploration setting, characterized by rough topography, limited accessibility, and a very complex geotectonic regime. We gather a passive seismic dataset consisting of 4 months of continuous recordings (May-August 2018) from 129 stand-alone three-component seismological stations. We then analyze this dataset and extract 848 microearthquakes that will serve as sources for the application of local earthquake tomography (LET) and transient-source seismic interferometry (TSI) by autocorrelation. We apply LET to estimate 3D P- and S-wave velocity models of the subsurface below the study area and TSI by autocorrelation to retrieve the zero-offset virtual reflection responses below each of the recording stations. The velocity models provide a relatively coarse image of a previously completely unexplored part of the mining concession, while the higher-resolution virtual reflection imaging illuminates in detail the different interfaces. We also reprocess three lines of legacy active seismic data that were shot in 2003, using the LET P-wave velocity model for depth migration, and confirm the improvement of seismic imaging. Finally, we evaluate the obtained results using well data and jointly interpret them, extracting useful information on the expected target depths and showing that earthquake-based passive seismic techniques can be an innovative and environmentally friendly option for mineral exploration.
{"title":"Integrating earthquake-based passive seismic in mineral exploration: case study from the Gerolekas bauxite mining area, Greece","authors":"Katerina Polychronopoulou, Michal Malinowski, Marta Cyz, Nikos Martakis, George Apostolopoulos, Deyan Draganov","doi":"10.1190/geo2023-0213.1","DOIUrl":"https://doi.org/10.1190/geo2023-0213.1","url":null,"abstract":"As the global need for aluminum constantly rises, bauxite is considered to be a critical mineral, and the mining industry is in search of new and effective exploration solutions. In this context, we designed and implemented a purely earthquake-based passive seismic survey at the Gerolekas bauxite mining site, in Greece. It is a very difficult exploration setting, characterized by rough topography, limited accessibility, and a very complex geotectonic regime. We gather a passive seismic dataset consisting of 4 months of continuous recordings (May-August 2018) from 129 stand-alone three-component seismological stations. We then analyze this dataset and extract 848 microearthquakes that will serve as sources for the application of local earthquake tomography (LET) and transient-source seismic interferometry (TSI) by autocorrelation. We apply LET to estimate 3D P- and S-wave velocity models of the subsurface below the study area and TSI by autocorrelation to retrieve the zero-offset virtual reflection responses below each of the recording stations. The velocity models provide a relatively coarse image of a previously completely unexplored part of the mining concession, while the higher-resolution virtual reflection imaging illuminates in detail the different interfaces. We also reprocess three lines of legacy active seismic data that were shot in 2003, using the LET P-wave velocity model for depth migration, and confirm the improvement of seismic imaging. Finally, we evaluate the obtained results using well data and jointly interpret them, extracting useful information on the expected target depths and showing that earthquake-based passive seismic techniques can be an innovative and environmentally friendly option for mineral exploration.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"2016 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135094044","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Gauss-Newton method has good convergence properties when employed for the solution of both seismic and electromagnetic inversion problems. One main issue is high numerical cost. The numerical cost can be reduced if the optimization domain can be decoupled from the simulation domain and such that the number of optimization parameters is much smaller than the number of grid nodes required for accurate simulation results. Overparameterization can be avoided. The decoupling can be achieved in a rigorous manner with the use of node-based basis functions. We provide a generic derivation of the method that is easily specialized to seismic and electromagnetic problems. The transformations between the optimization domain and the simulation domain are most effective if both domains can be described by rectilinear grids. A variable seabed depth causes a difficulty. We introduce a transform from the true bathymetry to a flat seabed that solves this problem. The method is validated by application to both synthetic and real electromagnetic data sets. The real data was acquired at the slow spreading Mohns ridge located east of Greenland and southwest of Svalbard. We provide a discussion on the interpretation of these data for an inverse scheme using the VTI (Transverse Isotropy with a Vertical symmetry axis) approximation. We offer some insights on how to interpret inversion results in the case of exploration for marine minerals. The interpretation differs from a hydrocarbon exploration setting owing to the presence of vertical conductors due to formation water circulation and vertical resistors due to volcanic intrusions.
{"title":"Gauss-Newton Inversion with Node-Based Basis Functions: Application#xD;to Imaging of Seabed Minerals in an Area with Rough Bathymetry#xD;","authors":"Rune Mittet, Anna Avdeeva","doi":"10.1190/geo2022-0763.1","DOIUrl":"https://doi.org/10.1190/geo2022-0763.1","url":null,"abstract":"The Gauss-Newton method has good convergence properties when employed for the solution of both seismic and electromagnetic inversion problems. One main issue is high numerical cost. The numerical cost can be reduced if the optimization domain can be decoupled from the simulation domain and such that the number of optimization parameters is much smaller than the number of grid nodes required for accurate simulation results. Overparameterization can be avoided. The decoupling can be achieved in a rigorous manner with the use of node-based basis functions. We provide a generic derivation of the method that is easily specialized to seismic and electromagnetic problems. The transformations between the optimization domain and the simulation domain are most effective if both domains can be described by rectilinear grids. A variable seabed depth causes a difficulty. We introduce a transform from the true bathymetry to a flat seabed that solves this problem. The method is validated by application to both synthetic and real electromagnetic data sets. The real data was acquired at the slow spreading Mohns ridge located east of Greenland and southwest of Svalbard. We provide a discussion on the interpretation of these data for an inverse scheme using the VTI (Transverse Isotropy with a Vertical symmetry axis) approximation. We offer some insights on how to interpret inversion results in the case of exploration for marine minerals. The interpretation differs from a hydrocarbon exploration setting owing to the presence of vertical conductors due to formation water circulation and vertical resistors due to volcanic intrusions.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135347479","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In a wellbore, any change in flow rate will result in a change in flow pattern and velocity. The flow pattern and velocity are the key parameters that determine the pressure gradient and liquid holdup. To study the effect of the flow pattern and velocity field on the flow rate of oil-water flow in horizontal wells, we apply the commercial software package ANSYS Fluent 2020 R2 to predict the flow patterns, water holdups, pressure gradients, flow rates, and velocity fields of horizontal wells. Trallero’s flow pattern chart and existing experimental data are used to verify the reliability of the model. We develop a simplified mathematical model of water holdup and compare it with existing models. This mathematical model may be limited to the range of fluid properties in the simulated method. The water holdup of the numerical simulation has a definite correlation with the experimental data. By comparing the numerical simulation results of the Nicolas model, the relationship between the slip velocity and water holdup is verified, and the reliability of the simulation results is verified. The simulation results demonstrate that the change in flow pattern is highly sensitive to the change in flow rate. When the flow pattern is stratified flow, the relative error of the simulated flow is small. When the flow pattern is dispersed flow, the relative error of the simulated flow is slightly larger. The oil is mainly concentrated in the high-velocity core area. At a higher total mixing velocity, the flow pattern is that of dispersed flow, with one phase uniformly mixed in the other phase. The simulation results have good qualitative and quantitative agreement with the experimental results.
{"title":"Effect of flow patterns and velocity field on oil-water two-phase flow rate in horizontal wells","authors":"Yuyan Wu, Haimin Guo, Rui Deng, Hongwei Song","doi":"10.1190/geo2023-0061.1","DOIUrl":"https://doi.org/10.1190/geo2023-0061.1","url":null,"abstract":"In a wellbore, any change in flow rate will result in a change in flow pattern and velocity. The flow pattern and velocity are the key parameters that determine the pressure gradient and liquid holdup. To study the effect of the flow pattern and velocity field on the flow rate of oil-water flow in horizontal wells, we apply the commercial software package ANSYS Fluent 2020 R2 to predict the flow patterns, water holdups, pressure gradients, flow rates, and velocity fields of horizontal wells. Trallero’s flow pattern chart and existing experimental data are used to verify the reliability of the model. We develop a simplified mathematical model of water holdup and compare it with existing models. This mathematical model may be limited to the range of fluid properties in the simulated method. The water holdup of the numerical simulation has a definite correlation with the experimental data. By comparing the numerical simulation results of the Nicolas model, the relationship between the slip velocity and water holdup is verified, and the reliability of the simulation results is verified. The simulation results demonstrate that the change in flow pattern is highly sensitive to the change in flow rate. When the flow pattern is stratified flow, the relative error of the simulated flow is small. When the flow pattern is dispersed flow, the relative error of the simulated flow is slightly larger. The oil is mainly concentrated in the high-velocity core area. At a higher total mixing velocity, the flow pattern is that of dispersed flow, with one phase uniformly mixed in the other phase. The simulation results have good qualitative and quantitative agreement with the experimental results.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135352301","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Conventional borehole acoustic measurements deliver compressional and shear wave slowness logs that inherently average in-situ rock properties along the receiver array of the acoustic instrument. These acquisition and processing conditions often limit the accuracy and resolution of the estimated rock elastic properties across heterolithic sedimentary sequences. We introduce an inversion-based interpretation method for borehole acoustic measurements that improves their vertical resolution by complementing them with ultrasonic borehole images. Results consist of high-resolution, layer-by-layer compressional and shear wave slownesses. The combination of borehole acoustic measurements with borehole ultrasonic images enhances the definition of small rock features such as thin beds or vugs. We verify the new inversion-based interpretation method with synthetic borehole measurements and field acoustic logs acquired across sandstone-shale laminated formations and spatially heterogeneous carbonates. High-resolution layer-by-layer compressional and shear slownesses obtained with the new inversion method give rise to wider variations of calculated elastic properties than with standard acoustic logs for improved petrophysical and geomechanical evaluation. It is also found that implementing a common set of layers for the estimation of layer-by-layer rock elastic properties mitigates biases due to discrepancies in the intrinsic resolution of the various input measurements.
{"title":"Axial Resolution Enhancement of Borehole Acoustic Measurements via Inversion-Based interpretation Supported with Ultrasonic data","authors":"Jingxuan Liu, Ali Eghbali, Carlos Torres-Verdín","doi":"10.1190/geo2023-0313.1","DOIUrl":"https://doi.org/10.1190/geo2023-0313.1","url":null,"abstract":"Conventional borehole acoustic measurements deliver compressional and shear wave slowness logs that inherently average in-situ rock properties along the receiver array of the acoustic instrument. These acquisition and processing conditions often limit the accuracy and resolution of the estimated rock elastic properties across heterolithic sedimentary sequences. We introduce an inversion-based interpretation method for borehole acoustic measurements that improves their vertical resolution by complementing them with ultrasonic borehole images. Results consist of high-resolution, layer-by-layer compressional and shear wave slownesses. The combination of borehole acoustic measurements with borehole ultrasonic images enhances the definition of small rock features such as thin beds or vugs. We verify the new inversion-based interpretation method with synthetic borehole measurements and field acoustic logs acquired across sandstone-shale laminated formations and spatially heterogeneous carbonates. High-resolution layer-by-layer compressional and shear slownesses obtained with the new inversion method give rise to wider variations of calculated elastic properties than with standard acoustic logs for improved petrophysical and geomechanical evaluation. It is also found that implementing a common set of layers for the estimation of layer-by-layer rock elastic properties mitigates biases due to discrepancies in the intrinsic resolution of the various input measurements.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"121 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135352288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Théo Rebert, Caifang Cai, Amélie Hallier, Thomas Bardainne
Rockfalls pose a threat to human infrastructure below cliffs. Sensitive and reactive alarm systems are needed for rail traffic safety, as small rockfalls (≈ 0.01 m 3 ) impacting the rail may cause train derailment. We propose to use seismic processing for rockfall early warning, powered by dense arrays deployed along the track. The method is evaluated by dropping rocks from a controlled height and triggering rockfalls on a cliff. We show that seismic arrays are highly sensitive to small impacts, and are able to detect them, locate them, and estimate their magnitude. The detection can be performed in near real-time with a simple algorithm, as small-scale rockfalls produce impulsive waveforms near the impact. Precise localization with Matched Field Processing is able to track the trajectory of a rockfall. Impacts against a rail might be recognized by their source signature. The seismic amplitudes are related to the rockfall volume by the Hertz law, which may be used to estimate their volume. These results show the potential of seismic-driven near real-time rockfall alarm systems.
岩崩对悬崖下的人类基础设施构成威胁。铁路交通安全需要敏感和反应报警系统，因为撞击轨道的小落石(≈0.01 m 3)可能导致列车脱轨。我们建议利用地震处理技术进行岩崩早期预警，由沿轨道部署的密集阵列提供动力。该方法是通过从控制的高度投掷岩石并触发悬崖上的岩崩来评估的。我们表明，地震阵列对小的冲击高度敏感，并且能够检测到它们，定位它们，并估计它们的震级。这种检测可以用一种简单的算法近乎实时地进行，因为小规模的岩崩会在撞击点附近产生脉冲波形。通过匹配场处理的精确定位能够跟踪岩崩的轨迹。对轨道的冲击可以通过其源特征来识别。地震振幅根据赫兹定律与岩崩体积相关，赫兹定律可用于估计岩崩体积。这些结果显示了地震驱动的近实时岩崩报警系统的潜力。
{"title":"Rockfall alarm system for railway monitoring: integrating seismic detection, localization and characterization","authors":"Théo Rebert, Caifang Cai, Amélie Hallier, Thomas Bardainne","doi":"10.1190/geo2023-0058.1","DOIUrl":"https://doi.org/10.1190/geo2023-0058.1","url":null,"abstract":"Rockfalls pose a threat to human infrastructure below cliffs. Sensitive and reactive alarm systems are needed for rail traffic safety, as small rockfalls (≈ 0.01 m 3 ) impacting the rail may cause train derailment. We propose to use seismic processing for rockfall early warning, powered by dense arrays deployed along the track. The method is evaluated by dropping rocks from a controlled height and triggering rockfalls on a cliff. We show that seismic arrays are highly sensitive to small impacts, and are able to detect them, locate them, and estimate their magnitude. The detection can be performed in near real-time with a simple algorithm, as small-scale rockfalls produce impulsive waveforms near the impact. Precise localization with Matched Field Processing is able to track the trajectory of a rockfall. Impacts against a rail might be recognized by their source signature. The seismic amplitudes are related to the rockfall volume by the Hertz law, which may be used to estimate their volume. These results show the potential of seismic-driven near real-time rockfall alarm systems.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"437 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":"134975328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Carlos Cunha, Gerson Ritter, Alexandre Sardinha, Bruno Pereira Dias, Claudio Guerra, Fernanda Thedy, Nelson Hargreaves, Rodrigo Coacci
We define a common framework for reverse-time migration (RTM) and Kirchhoff migration based on compact representations of Green’s functions. These compact Green’s functions (CGF) are 3D volumes containing traveltimes and amplitudes for the N most representative events in the upcoming/downgoing decomposed 4D wavefields originating from a point source. Within this framework, we implement an RTM algorithm using a multivalued excitation time/amplitude imaging condition. This new approach produces four complementary imaging volumes (different combinations of source and receiver decomposed wavefields) and angle/azimuth gathers with computational effort less than 15% greater than that of plain (one image, no gathers) RTM algorithms. The advantages of separating the image volume into four complementary volumes are well-established in the literature (low frequency noise separation and turning-wave imaging); however, its use has been limited by the computational cost. Despite using two source propagations to decompose the source wavefield, we reduce the computations to less than 20% of a single source propagation by performing finite-difference propagation with half the frequency limit used in the receiver wavefield propagation. The combination of CGF and an excitation time/amplitude imaging condition allows receiver wavefield decomposition with only one wavefield propagation. Our RTM algorithm constructs angle/azimuth gathers using a post-migration computation of the source and receiver wavefield’s propagation directions. To compute the propagation directions after migration, we use a new concept: the cumulative wavefield volumes, which are 3D, imaging-condition-guided compressions, of the 4D source and receiver wavefields. We also use CGF to implement a Kirchhoff migration algorithm that produces four complementary image volumes with RTM-like quality. Further, we present synthetic and field data examples to clarify the new concepts and illustrate the results obtained using both methods.
{"title":"Multi-image, reverse-time and Kirchhoff migrations with compact Green′s functions","authors":"Carlos Cunha, Gerson Ritter, Alexandre Sardinha, Bruno Pereira Dias, Claudio Guerra, Fernanda Thedy, Nelson Hargreaves, Rodrigo Coacci","doi":"10.1190/geo2023-0106.1","DOIUrl":"https://doi.org/10.1190/geo2023-0106.1","url":null,"abstract":"We define a common framework for reverse-time migration (RTM) and Kirchhoff migration based on compact representations of Green’s functions. These compact Green’s functions (CGF) are 3D volumes containing traveltimes and amplitudes for the N most representative events in the upcoming/downgoing decomposed 4D wavefields originating from a point source. Within this framework, we implement an RTM algorithm using a multivalued excitation time/amplitude imaging condition. This new approach produces four complementary imaging volumes (different combinations of source and receiver decomposed wavefields) and angle/azimuth gathers with computational effort less than 15% greater than that of plain (one image, no gathers) RTM algorithms. The advantages of separating the image volume into four complementary volumes are well-established in the literature (low frequency noise separation and turning-wave imaging); however, its use has been limited by the computational cost. Despite using two source propagations to decompose the source wavefield, we reduce the computations to less than 20% of a single source propagation by performing finite-difference propagation with half the frequency limit used in the receiver wavefield propagation. The combination of CGF and an excitation time/amplitude imaging condition allows receiver wavefield decomposition with only one wavefield propagation. Our RTM algorithm constructs angle/azimuth gathers using a post-migration computation of the source and receiver wavefield’s propagation directions. To compute the propagation directions after migration, we use a new concept: the cumulative wavefield volumes, which are 3D, imaging-condition-guided compressions, of the 4D source and receiver wavefields. We also use CGF to implement a Kirchhoff migration algorithm that produces four complementary image volumes with RTM-like quality. Further, we present synthetic and field data examples to clarify the new concepts and illustrate the results obtained using both methods.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"16 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":"134975490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Seismic attenuation is a basic physical property of the Earth, which significantly affects the characteristics of seismic wavefields. Accurately simulating wave propagation in the Earth is essential to image subsurface structures. Some prevailing methods (e.g., the standard linear solid and fractional Laplacian equation) to describe seismic wave propagation in attenuating media are mainly based on the constant- Q model (CQM), which is valid at room temperature and pressure. However, laboratory measurements suggest that the quality factor Q is a function of frequencies in some regions. To simulate the frequency-dependent Q effect, we derive a viscoacoustic wave equation from the stress-strain relationship of the fractional Zener model (FZM) with variable fractional orders. During the implementation, we separate the real and imaginary parts of the modulus and introduce a low-rank decomposition method to solve the FZM equation. Since the amplitude dissipation and phase dispersion are decoupled, we establish a compensated reverse-time migration ( Q-RTM) algorithm to mitigate adverse effects caused by seismic attenuation and improve the quality of seismic migration in frequency-dependent attenuating media. A two-layer and the BP gas chimney models are used to perform Q-RTM tests. A low-pass filter with a Tukey window function is applied to suppress numerical instability during the compensation. Numerical results demonstrate that the proposed FZM Q-RTM approach can produce high-resolution images with corrected reflector positions and amplitudes. Because the CQM equation ignores the frequency dependence of Q, it may lead to over-compensation in Q-RTM.
{"title":"Frequency-dependent <i>Q</i> simulation and viscoacoustic reverse-time migration based on the fractional Zener model","authors":"Yabing Zhang, Hejun Zhu, Yang Liu, Tongjun Chen","doi":"10.1190/geo2023-0258.1","DOIUrl":"https://doi.org/10.1190/geo2023-0258.1","url":null,"abstract":"Seismic attenuation is a basic physical property of the Earth, which significantly affects the characteristics of seismic wavefields. Accurately simulating wave propagation in the Earth is essential to image subsurface structures. Some prevailing methods (e.g., the standard linear solid and fractional Laplacian equation) to describe seismic wave propagation in attenuating media are mainly based on the constant- Q model (CQM), which is valid at room temperature and pressure. However, laboratory measurements suggest that the quality factor Q is a function of frequencies in some regions. To simulate the frequency-dependent Q effect, we derive a viscoacoustic wave equation from the stress-strain relationship of the fractional Zener model (FZM) with variable fractional orders. During the implementation, we separate the real and imaginary parts of the modulus and introduce a low-rank decomposition method to solve the FZM equation. Since the amplitude dissipation and phase dispersion are decoupled, we establish a compensated reverse-time migration ( Q-RTM) algorithm to mitigate adverse effects caused by seismic attenuation and improve the quality of seismic migration in frequency-dependent attenuating media. A two-layer and the BP gas chimney models are used to perform Q-RTM tests. A low-pass filter with a Tukey window function is applied to suppress numerical instability during the compensation. Numerical results demonstrate that the proposed FZM Q-RTM approach can produce high-resolution images with corrected reflector positions and amplitudes. Because the CQM equation ignores the frequency dependence of Q, it may lead to over-compensation in Q-RTM.","PeriodicalId":55102,"journal":{"name":"Geophysics","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135597595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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：