Pub Date : 2026-04-01Epub Date: 2026-01-26DOI: 10.1016/j.jappgeo.2026.106136
Chaoqiang Xi , Hao Zhang , Ya Liu , Ling Ning
Ambient-noise interferometry from dense seismic arrays provides rich surface-wave information for dispersion imaging and multiscale shear-wave velocity Vs inversion, but it also produces very large virtual shot gathers that make conventional frequency-domain slant stacking computationally expensive. We present a frequency-domain fast slant-stacking method based on the non-uniform fast Fourier transform NUFFT for dispersion imaging from dense-array ambient seismic noise. Synthetic and field tests show that the NUFFT-based formulation yields same dispersion spectra and picked dispersion curves that are identical to those obtained by conventional direct summation, with an RMS misfit of 0. Benchmarks on three field datasets demonstrate substantial runtime reductions, requiring 1.44–2.23 s and 0.33–0.49 s with parallelization for NUFFT, whereas Numba, PyTorch, and NumPy implementations typically require tens to hundreds of seconds under the same settings, achieving speedups of up to about 60 times relative to the fastest conventional baseline. Comparisons with the widely used CC-FJpy package show consistent dispersion trends while requiring significantly less computation time.
{"title":"Frequency-domain fast slant stacking method based on non-uniform fast Fourier transform for dispersion imaging from dense array ambient seismic noise","authors":"Chaoqiang Xi , Hao Zhang , Ya Liu , Ling Ning","doi":"10.1016/j.jappgeo.2026.106136","DOIUrl":"10.1016/j.jappgeo.2026.106136","url":null,"abstract":"<div><div>Ambient-noise interferometry from dense seismic arrays provides rich surface-wave information for dispersion imaging and multiscale shear-wave velocity Vs inversion, but it also produces very large virtual shot gathers that make conventional frequency-domain slant stacking computationally expensive. We present a frequency-domain fast slant-stacking method based on the non-uniform fast Fourier transform NUFFT for dispersion imaging from dense-array ambient seismic noise. Synthetic and field tests show that the NUFFT-based formulation yields same dispersion spectra and picked dispersion curves that are identical to those obtained by conventional direct summation, with an RMS misfit of 0. Benchmarks on three field datasets demonstrate substantial runtime reductions, requiring 1.44–2.23 s and 0.33–0.49 s with parallelization for NUFFT, whereas Numba, PyTorch, and NumPy implementations typically require tens to hundreds of seconds under the same settings, achieving speedups of up to about 60 times relative to the fastest conventional baseline. Comparisons with the widely used CC-FJpy package show consistent dispersion trends while requiring significantly less computation time.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106136"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081613","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-04-01Epub Date: 2026-01-25DOI: 10.1016/j.jappgeo.2026.106121
Wenda Sun, Jing Zheng, Suping Peng
Distributed Acoustic Sensing (DAS) is an emerging seismic acquisition technology that offers high spatial sampling density, continuous recording capability, and flexible deployment. These characteristics make it particularly suitable for shallow subsurface exploration in urban environments. However, DAS exhibits limited sensitivity to weak seismic signals and typically captures only the axial component of ground motion. In contrast, conventional geophones offer high signal fidelity and multi-component recordings, but are limited by lower spatial resolution, greater deployment costs, and reduced adaptability in complex terrain conditions. To enhance surface wavefields reconstructed from DAS ambient noise, we propose a method that integrates the complementary strengths of DAS and conventional geophones. The method employs a linear array configuration, where the vertical component of a geophone deployed at the front of the array serves as a virtual source, and the DAS system is deployed along the subsequent positions of the array to serve as receivers. These recordings are combined to form a hybrid ambient noise dataset, which is processed in the frequency domain through normalization and cross-correlation. The acausal parts of the cross-correlation functions (CCFs) are taken as virtual shot gathers (VSGs). This method not only preserves the high spatial resolution of DAS but also incorporates the high signal fidelity of geophones, thereby significantly enhancing the surface wavefields in passive surface wave imaging. By evaluating the signal-to-noise ratio (SNR) of randomly selected traces from the CCFs obtained under various ambient noise stacking durations, the proposed method achieves average SNR improvements of 1.54 dB and 1.00 dB compared to the case where both the virtual source and receivers are derived from DAS data. Under the optimal stacking duration, the extracted dispersion energy shows clearer and more concentrated patterns, with an extended frequency range.
{"title":"A new method for reconstructing surface wavefields via cross-correlation of ambient noise from DAS and geophone records","authors":"Wenda Sun, Jing Zheng, Suping Peng","doi":"10.1016/j.jappgeo.2026.106121","DOIUrl":"10.1016/j.jappgeo.2026.106121","url":null,"abstract":"<div><div>Distributed Acoustic Sensing (DAS) is an emerging seismic acquisition technology that offers high spatial sampling density, continuous recording capability, and flexible deployment. These characteristics make it particularly suitable for shallow subsurface exploration in urban environments. However, DAS exhibits limited sensitivity to weak seismic signals and typically captures only the axial component of ground motion. In contrast, conventional geophones offer high signal fidelity and multi-component recordings, but are limited by lower spatial resolution, greater deployment costs, and reduced adaptability in complex terrain conditions. To enhance surface wavefields reconstructed from DAS ambient noise, we propose a method that integrates the complementary strengths of DAS and conventional geophones. The method employs a linear array configuration, where the vertical component of a geophone deployed at the front of the array serves as a virtual source, and the DAS system is deployed along the subsequent positions of the array to serve as receivers. These recordings are combined to form a hybrid ambient noise dataset, which is processed in the frequency domain through normalization and cross-correlation. The acausal parts of the cross-correlation functions (CCFs) are taken as virtual shot gathers (VSGs). This method not only preserves the high spatial resolution of DAS but also incorporates the high signal fidelity of geophones, thereby significantly enhancing the surface wavefields in passive surface wave imaging. By evaluating the signal-to-noise ratio (SNR) of randomly selected traces from the CCFs obtained under various ambient noise stacking durations, the proposed method achieves average SNR improvements of 1.54 dB and 1.00 dB compared to the case where both the virtual source and receivers are derived from DAS data. Under the optimal stacking duration, the extracted dispersion energy shows clearer and more concentrated patterns, with an extended frequency range.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106121"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-04-01Epub Date: 2026-01-27DOI: 10.1016/j.jappgeo.2026.106135
Lei Chen , Zhifei Gong , Long Chen , Chuanwu Wang
Currently, more and more tunnels are being built and designed in urban and mountain areas. To ensure the safe construction of a tunnel, the seismic method is widely applied to estimate the geological conditions ahead of the tunnel face. However, due to the complex environment in the tunnel (narrow observation aperture, strong noise, etc.), it is hard to extract the effective reflected waves with poor energy for imaging geology, especially for the target far away from the tunnel face. In this paper, for the demand of geological detection in the Water Diversion Project from the Songhua River, the tunnel seismic ahead-prospecting was improved by using the adaptive beamforming approach, supporting the extraction of reflected waves with poor energy. As the core of beamforming, the delay time is an important parameter to control the reliability of stacking, especially for the various strata. The relevance analysis-based adaptive beamforming is proposed for the delay-time calculation to obtain the correlation coefficient. Among them, the reflection group composed of several reflections is used to obtain an accurate delay time first. Then, the seismic data are reconstructed and overlapped to improve the SNR based on the obtained delay time. Numerical simulation denotes that the energy of reflected waves can be enhanced significantly. Finally, the improved seismic ahead-prospecting was applied in the Water Diversion Project from the Songhua River; the method successfully predicted the fractured zones ahead of the tunnel face and guaranteed tunnel construction safety.
{"title":"Tunnel geological forward-prospecting based on an optimized beamforming seismic method","authors":"Lei Chen , Zhifei Gong , Long Chen , Chuanwu Wang","doi":"10.1016/j.jappgeo.2026.106135","DOIUrl":"10.1016/j.jappgeo.2026.106135","url":null,"abstract":"<div><div>Currently, more and more tunnels are being built and designed in urban and mountain areas. To ensure the safe construction of a tunnel, the seismic method is widely applied to estimate the geological conditions ahead of the tunnel face. However, due to the complex environment in the tunnel (narrow observation aperture, strong noise, etc.), it is hard to extract the effective reflected waves with poor energy for imaging geology, especially for the target far away from the tunnel face. In this paper, for the demand of geological detection in the Water Diversion Project from the Songhua River, the tunnel seismic ahead-prospecting was improved by using the adaptive beamforming approach, supporting the extraction of reflected waves with poor energy. As the core of beamforming, the delay time is an important parameter to control the reliability of stacking, especially for the various strata. The relevance analysis-based adaptive beamforming is proposed for the delay-time calculation to obtain the correlation coefficient. Among them, the reflection group composed of several reflections is used to obtain an accurate delay time first. Then, the seismic data are reconstructed and overlapped to improve the SNR based on the obtained delay time. Numerical simulation denotes that the energy of reflected waves can be enhanced significantly. Finally, the improved seismic ahead-prospecting was applied in the Water Diversion Project from the Songhua River; the method successfully predicted the fractured zones ahead of the tunnel face and guaranteed tunnel construction safety.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106135"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-04-01Epub Date: 2026-01-24DOI: 10.1016/j.jappgeo.2026.106125
Feng Qiu , Zhengwang Hu , Jinsong Du , Songtao Hu
With the same distribution of the observation points, the third-order gravitational gradient tensor is capable of acquiring more comprehensive information pertaining to the geophysical field and its corresponding sources, demonstrates superior sensitivity to changes in the burial depth of field sources, and features enhanced horizontal resolving power. But the current instrumental measurement technology has not yet reached the level of measuring the third-order gradient tensor of gravitational potential directly in the field. In this paper, we provide an alternative technique for calculating the complete third-order gradient tensor of gravitational potential from the pre-existing vertical gravity gradient data by using the fast Fourier transform (FFT). In order to verify the correctness of the transform calculation method, the complete third-order gradient tensor components of gravitational potential are computed from the synthetic vertical gravity gradient data by two different three-dimensional (3-D) theoretical density models. Comparing the FFT-based calculation results with theoretically forward calculated data shows that the root-mean-square (RMS) error for each component, is at most 1.5 pMKS (1 pMKS = 10−12 m−1·s−2). In addition, as a real example, based on the practically measured vertical gravity gradient data over the Vinton salt dome area, the complete third-order gradient tensor is obtained by using the FFT-based calculation method and the results illustrate a better resolution and a richer information for understanding the underground density structures. Then, the determined tensor data is quantitatively interpreted by using the DEXP (i.e., Depth from EXtreme Point) imaging technique. The imaging results show that, the depth and boundaries of anomalous density sources by the DEXP imaging method are consistent with the previous research results, and the third-order gradient tensor-based imaging has a weaker influence of trend component than the traditional gravity and second-order gravity gradient tensor. Both synthetic examples and practical application suggest that our proposed method is not only valid and reliable but also has a high computational efficiency due to the FFT algorithm, and meanwhile the calculated third-order gradient tensor also provides a novel way for exploring the geological structures, ore bodies and hydrocarbon reservoirs, etc.
{"title":"A FFT-based calculation method for third-order gradient tensor of gravity potential from vertical gravity gradient data","authors":"Feng Qiu , Zhengwang Hu , Jinsong Du , Songtao Hu","doi":"10.1016/j.jappgeo.2026.106125","DOIUrl":"10.1016/j.jappgeo.2026.106125","url":null,"abstract":"<div><div>With the same distribution of the observation points, the third-order gravitational gradient tensor is capable of acquiring more comprehensive information pertaining to the geophysical field and its corresponding sources, demonstrates superior sensitivity to changes in the burial depth of field sources, and features enhanced horizontal resolving power. But the current instrumental measurement technology has not yet reached the level of measuring the third-order gradient tensor of gravitational potential directly in the field. In this paper, we provide an alternative technique for calculating the complete third-order gradient tensor of gravitational potential from the pre-existing vertical gravity gradient data by using the fast Fourier transform (FFT). In order to verify the correctness of the transform calculation method, the complete third-order gradient tensor components of gravitational potential are computed from the synthetic vertical gravity gradient data by two different three-dimensional (3-D) theoretical density models. Comparing the FFT-based calculation results with theoretically forward calculated data shows that the root-mean-square (RMS) error for each component, is at most 1.5 pMKS (1 pMKS = 10<sup>−12</sup> m<sup>−1</sup>·s<sup>−2</sup>). In addition, as a real example, based on the practically measured vertical gravity gradient data over the Vinton salt dome area, the complete third-order gradient tensor is obtained by using the FFT-based calculation method and the results illustrate a better resolution and a richer information for understanding the underground density structures. Then, the determined tensor data is quantitatively interpreted by using the DEXP (i.e., Depth from EXtreme Point) imaging technique. The imaging results show that, the depth and boundaries of anomalous density sources by the DEXP imaging method are consistent with the previous research results, and the third-order gradient tensor-based imaging has a weaker influence of trend component than the traditional gravity and second-order gravity gradient tensor. Both synthetic examples and practical application suggest that our proposed method is not only valid and reliable but also has a high computational efficiency due to the FFT algorithm, and meanwhile the calculated third-order gradient tensor also provides a novel way for exploring the geological structures, ore bodies and hydrocarbon reservoirs, etc.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106125"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-04-01Epub Date: 2026-01-22DOI: 10.1016/j.jappgeo.2026.106127
Luiz Filipe Caríssimo Soares , Eduardo Antonio Gomes Marques , Cibele Cláuver
Assessing safety and conservation conditions in earth dams involves the investigation of phenomena that compromise their stability. In this regard, geophysical methods can be combined with direct investigations to enhance the efficiency of geotechnical studies, allowing faster data acquisition at relatively low cost. Therefore, this study aims to evaluate the conservation conditions of a small earth dam through the integrated interpretation of results obtained using the geophysical methods of Electrical Resistivity and Ground-Penetrating Radar, coupled with the results of grain size analyses of samples collected in situ. The results confirmed the efficiency of the proposed methodology, enabling assessment of features associated with the prolonged biological activity and surface water runoff. Among the identified geotechnical properties, a zone affected by the intense presence of ant nests and tree roots was detected, highlighting the need for corrective actions to maintain the structural integrity of the embankment.
{"title":"Application of geophysical methods in the assessment of conservation conditions in a small earth dam","authors":"Luiz Filipe Caríssimo Soares , Eduardo Antonio Gomes Marques , Cibele Cláuver","doi":"10.1016/j.jappgeo.2026.106127","DOIUrl":"10.1016/j.jappgeo.2026.106127","url":null,"abstract":"<div><div>Assessing safety and conservation conditions in earth dams involves the investigation of phenomena that compromise their stability. In this regard, geophysical methods can be combined with direct investigations to enhance the efficiency of geotechnical studies, allowing faster data acquisition at relatively low cost. Therefore, this study aims to evaluate the conservation conditions of a small earth dam through the integrated interpretation of results obtained using the geophysical methods of Electrical Resistivity and Ground-Penetrating Radar, coupled with the results of grain size analyses of samples collected in situ. The results confirmed the efficiency of the proposed methodology, enabling assessment of features associated with the prolonged biological activity and surface water runoff. Among the identified geotechnical properties, a zone affected by the intense presence of ant nests and tree roots was detected, highlighting the need for corrective actions to maintain the structural integrity of the embankment.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106127"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081612","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-04-01Epub Date: 2026-01-23DOI: 10.1016/j.jappgeo.2026.106128
Jiwei Zhang , Xiaoyi Ji , Mingzhe Zhao , Yaxiu Li , Haifeng Wang , Ming Zhong , Shuai Li
Underground defects in urban roads endanger driving safety and hinder road usability. These defects are primarily identified using technologies such as ground penetrating radar. The current intelligent algorithms used for identifying underground road defects rely heavily on large datasets of on-site road images. However, the automatic detection of defects remains challenging due to small datasets, limited image availability, and inconsistent on-field image quality. This paper proposes a novel approach to address these challenges through a model based on actual road conditions and forward simulations of road defect images. To improve the quality of both real and simulated field images, we apply a joint denoising method that combines wavelet transform, the K-SVD algorithm, and bilateral filtering. This denoising process enhances both real and simulated field images and expands the image dataset, transforming it into a mixed database, and strengthens the distinctive features of each defect, facilitating more accurate algorithm-based detection. In the first and second stages of the study, we conduct a comparative analysis of various deep learning-based object detection models. We then propose a deep learning model, optimized with the joint denoising model, that is best suited for practical road evaluation projects. The model was trained and validated across 100 km of high-quality field measurement data collected from various districts and counties in Beijing. Experimental results showed that the model can achieve a prediction accuracy of 82.3% for Looseness, 92.6% for Cavities, and 50.9% for Voids, with an overall Mean Average Precision of 75.3%. These results demonstrate that the method proposed in this study can enhance the detection accuracy for various subsurface defects.
{"title":"Automatic identification and location of underground defects in urban roads via ground penetrating radar and deep learning approaches","authors":"Jiwei Zhang , Xiaoyi Ji , Mingzhe Zhao , Yaxiu Li , Haifeng Wang , Ming Zhong , Shuai Li","doi":"10.1016/j.jappgeo.2026.106128","DOIUrl":"10.1016/j.jappgeo.2026.106128","url":null,"abstract":"<div><div>Underground defects in urban roads endanger driving safety and hinder road usability. These defects are primarily identified using technologies such as ground penetrating radar. The current intelligent algorithms used for identifying underground road defects rely heavily on large datasets of on-site road images. However, the automatic detection of defects remains challenging due to small datasets, limited image availability, and inconsistent on-field image quality. This paper proposes a novel approach to address these challenges through a model based on actual road conditions and forward simulations of road defect images. To improve the quality of both real and simulated field images, we apply a joint denoising method that combines wavelet transform, the <em>K-SVD</em> algorithm, and bilateral filtering. This denoising process enhances both real and simulated field images and expands the image dataset, transforming it into a mixed database, and strengthens the distinctive features of each defect, facilitating more accurate algorithm-based detection. In the first and second stages of the study, we conduct a comparative analysis of various deep learning-based object detection models. We then propose a deep learning model, optimized with the joint denoising model, that is best suited for practical road evaluation projects. The model was trained and validated across 100 km of high-quality field measurement data collected from various districts and counties in Beijing. Experimental results showed that the model can achieve a prediction accuracy of 82.3% for Looseness, 92.6% for Cavities, and 50.9% for Voids, with an overall Mean Average Precision of 75.3%. These results demonstrate that the method proposed in this study can enhance the detection accuracy for various subsurface defects.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"247 ","pages":"Article 106128"},"PeriodicalIF":2.1,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146071109","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Conventional total magnetic intensity (TMI) anomaly continuation typically approximate the difference between the measured total magnetic field and the main field modulus as the projection of the magnetic vector in the direction of the main field. However, in regions with high TMI anomaly amplitudes, using conventional linear approximation multi-layer equivalent source models results in significant continuation errors. This paper presents a nonlinear multi-layer equivalent source method for magnetic anomaly continuation, which utilizes an iterative technique. The technique begins by analyzing the physical significance of the magnetic anomaly modulus and subsequently constructs the model based on this interpretation. The iterative process is optimized using an adaptive conjugate gradient method. The findings demonstrate that, in high-amplitude areas, the differences in TMI anomaly modulus and approximate projections need to be taken into account. Moreover, the method proposed in this paper enhances continuation accuracy, with its accuracy advantage becoming more pronounced as the continuation distance increases.
{"title":"Nonlinear multi-layer equivalent source continuation method in strong magnetic fields","authors":"Jinkai Feng, Shanshan Li, Xu Feng, Haopeng Fan, Xinxing Li, Diao Fan","doi":"10.1016/j.jappgeo.2025.106089","DOIUrl":"10.1016/j.jappgeo.2025.106089","url":null,"abstract":"<div><div>Conventional total magnetic intensity (TMI) anomaly continuation typically approximate the difference between the measured total magnetic field and the main field modulus as the projection of the magnetic vector in the direction of the main field. However, in regions with high TMI anomaly amplitudes, using conventional linear approximation multi-layer equivalent source models results in significant continuation errors. This paper presents a nonlinear multi-layer equivalent source method for magnetic anomaly continuation, which utilizes an iterative technique. The technique begins by analyzing the physical significance of the magnetic anomaly modulus and subsequently constructs the model based on this interpretation. The iterative process is optimized using an adaptive conjugate gradient method. The findings demonstrate that, in high-amplitude areas, the differences in TMI anomaly modulus and approximate projections need to be taken into account. Moreover, the method proposed in this paper enhances continuation accuracy, with its accuracy advantage becoming more pronounced as the continuation distance increases.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"246 ","pages":"Article 106089"},"PeriodicalIF":2.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-08DOI: 10.1016/j.jappgeo.2026.106106
Zhimei Kang , Yukun Liu , Xiaolong Wang , Yulin Du
Well logging data serve as a crucial bridge connecting seismic data and geological interpretation, playing an indispensable role in hydrocarbon exploration and resource evaluation. However, the complex heterogeneity of reservoirs and the high-temperature, high-pressure conditions in deep and ultradeep often lead to wellbore instability, formation loss, and technical constraints, which in turn cause severe data gaps and distortions during log acquisition. Traditional well log reconstruction methods, such as empirical models and multivariate fitting, generally suffer from reliance on subjective experience, low prediction accuracy, and poor model adaptability when dealing with such complex reservoirs. To address these challenges, this study proposes a novel log reconstruction model (Bi-LSTM-SAM-TTT) that integrates Bi-directional Long Short-Term Memory (Bi-LSTM) networks, a Self-Attention Mechanism (SAM), and Test-Time Training (TTT) algorithms. Geological stratification and depth information are incorporated as prior knowledge during model training, effectively strengthening the sequential correlation between geological features and logging data, and significantly improving reconstruction accuracy. By comparing multi-variate fitting, LSTM, Bi-LSTM, and Bi-LSTM-SAM methods, the results demonstrate that the Bi-LSTM-SAM-TTT model achieves the best performance in reconstructing three key logging curves: resistivity (RD), density (DEN), and acoustic interval transit time (DTC). Compared with the LSTM model, the proposed model reduces the RMSE by 46.1% (RD), 39.6% (DEN), and 39.5% (DTC), respectively, while the coefficient of determination (R2) increases to above 0.92 for all three curves. In a case study, the R2 values for DTC prediction using the four models were 0.8003, 0.8146, 0.8523, and 0.8843, respectively, with the Bi-LSTM-SAM-TTT model clearly outperforming the others. Moreover, prediction interval analysis under different confidence levels shows that the coverage of the 95% confidence interval exceeds 98%, indicating high predictive reliability of the proposed model. In summary, the Bi-LSTM-SAM-TTT model not only effectively mitigates the problem of missing well log data in ultradeep formations but also exhibits strong robustness and generalization capability, providing a new approach for high-precision well log reconstruction in deep and ultradeep hydrocarbon exploration.
测井资料是连接地震资料和地质解释的重要桥梁，在油气勘探和资源评价中发挥着不可或缺的作用。然而，储层复杂的非均质性，以及深层和超深层的高温高压条件，往往会导致井筒不稳定、地层漏失和技术限制，从而在测井采集过程中造成严重的数据缺口和失真。传统的测井重建方法，如经验模型和多元拟合，在处理此类复杂储层时，普遍存在依赖主观经验、预测精度低、模型适应性差的问题。为了解决这些挑战，本研究提出了一种新的日志重建模型（Bi-LSTM-SAM-TTT），该模型集成了双向长短期记忆（Bi-LSTM）网络、自注意机制（SAM）和测试时间训练（TTT）算法。在模型训练过程中，将地质分层和深度信息作为先验知识，有效加强了地质特征与测井资料的序列相关性，显著提高了重建精度。通过对比多元拟合、LSTM、Bi-LSTM和Bi-LSTM- sam方法，结果表明，Bi-LSTM- sam - ttt模型在重建电阻率（RD）、密度（DEN）和声波间隔透射时间（DTC）三条关键测井曲线方面表现最佳。与LSTM模型相比，该模型的RMSE分别降低了46.1% （RD）、39.6% （DEN）和39.5% (DTC)，三条曲线的决定系数（R2）均提高到0.92以上。在案例研究中，4种模型预测DTC的R2值分别为0.8003、0.8146、0.8523和0.8843，其中Bi-LSTM-SAM-TTT模型明显优于其他模型。此外，不同置信水平下的预测区间分析表明，95%置信区间的覆盖率超过98%，表明所提模型具有较高的预测可靠性。综上所述，Bi-LSTM-SAM-TTT模型不仅有效缓解了超深层地层测井资料缺失的问题，而且具有较强的鲁棒性和泛化能力，为深、超深层油气勘探的高精度测井重建提供了新的途径。
{"title":"A deep learning-based method for well log data reconstruction in marine carbonate reservoirs","authors":"Zhimei Kang , Yukun Liu , Xiaolong Wang , Yulin Du","doi":"10.1016/j.jappgeo.2026.106106","DOIUrl":"10.1016/j.jappgeo.2026.106106","url":null,"abstract":"<div><div>Well logging data serve as a crucial bridge connecting seismic data and geological interpretation, playing an indispensable role in hydrocarbon exploration and resource evaluation. However, the complex heterogeneity of reservoirs and the high-temperature, high-pressure conditions in deep and ultradeep often lead to wellbore instability, formation loss, and technical constraints, which in turn cause severe data gaps and distortions during log acquisition. Traditional well log reconstruction methods, such as empirical models and multivariate fitting, generally suffer from reliance on subjective experience, low prediction accuracy, and poor model adaptability when dealing with such complex reservoirs. To address these challenges, this study proposes a novel log reconstruction model (Bi-LSTM-SAM-TTT) that integrates Bi-directional Long Short-Term Memory (Bi-LSTM) networks, a Self-Attention Mechanism (SAM), and Test-Time Training (TTT) algorithms. Geological stratification and depth information are incorporated as prior knowledge during model training, effectively strengthening the sequential correlation between geological features and logging data, and significantly improving reconstruction accuracy. By comparing multi-variate fitting, LSTM, Bi-LSTM, and Bi-LSTM-SAM methods, the results demonstrate that the Bi-LSTM-SAM-TTT model achieves the best performance in reconstructing three key logging curves: resistivity (RD), density (DEN), and acoustic interval transit time (DTC). Compared with the LSTM model, the proposed model reduces the RMSE by 46.1% (RD), 39.6% (DEN), and 39.5% (DTC), respectively, while the coefficient of determination (R<sup>2</sup>) increases to above 0.92 for all three curves. In a case study, the R<sup>2</sup> values for DTC prediction using the four models were 0.8003, 0.8146, 0.8523, and 0.8843, respectively, with the Bi-LSTM-SAM-TTT model clearly outperforming the others. Moreover, prediction interval analysis under different confidence levels shows that the coverage of the 95% confidence interval exceeds 98%, indicating high predictive reliability of the proposed model. In summary, the Bi-LSTM-SAM-TTT model not only effectively mitigates the problem of missing well log data in ultradeep formations but also exhibits strong robustness and generalization capability, providing a new approach for high-precision well log reconstruction in deep and ultradeep hydrocarbon exploration.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"246 ","pages":"Article 106106"},"PeriodicalIF":2.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-15DOI: 10.1016/j.jappgeo.2026.106112
Longxiang Han , Chengliang Wu , Huazhong Wang
Consistent-phase stacking of band-limited imaging wavelet from the same subsurface reflection point but different source-receiver pairs is a fundamental requirement for achieving high-fidelity and high-resolution seismic imaging. The final image is typically obtained by stacking common-image gathers (CIGs). However, inconsistent wavelet phases across different angles or offsets in CIGs can lead to destructive interference, waveform distortion, and amplitude loss, ultimately degrading image resolution. Most conventional phase correction methods assume a constant phase shift across all frequencies, which fails to account for frequency-dependent phase variations introduced by source signatures, absorption, and other real-field factors. Neglecting these variations can significantly degrade the fidelity and resolution of the final stacked image. To address this issue, we propose a statistical method for detecting and correcting frequency-dependent phase differences in CIGs. After flattening the CIGs, we perform multi-scale Gaussian filtering to divide the data into narrow frequency bands, effectively reducing noise and ensuring more stable phase estimation. Then, the phase differences between the original and a reference CIG—formed by averaging multiple traces within the effective illumination range—are estimated for each narrow frequency band using a particle swarm optimization (PSO) algorithm. Treating the measured phase shift in each band as corresponding to its center frequency, we employ spline interpolation to construct a smooth, continuous phase correction curve. This curve is then applied to correct the wavelet phase across the full bandwidth. Both synthetic and field data are used to demonstrate the effectiveness of the proposed method. Experimental results show that the method effectively corrects residual phase differences in CIGs, significantly enhancing the amplitude fidelity and resolution of the final seismic image.
{"title":"Residual phase correction for common imaging gathers and its application in fidelity high-resolution imaging","authors":"Longxiang Han , Chengliang Wu , Huazhong Wang","doi":"10.1016/j.jappgeo.2026.106112","DOIUrl":"10.1016/j.jappgeo.2026.106112","url":null,"abstract":"<div><div>Consistent-phase stacking of band-limited imaging wavelet from the same subsurface reflection point but different source-receiver pairs is a fundamental requirement for achieving high-fidelity and high-resolution seismic imaging. The final image is typically obtained by stacking common-image gathers (CIGs). However, inconsistent wavelet phases across different angles or offsets in CIGs can lead to destructive interference, waveform distortion, and amplitude loss, ultimately degrading image resolution. Most conventional phase correction methods assume a constant phase shift across all frequencies, which fails to account for frequency-dependent phase variations introduced by source signatures, absorption, and other real-field factors. Neglecting these variations can significantly degrade the fidelity and resolution of the final stacked image. To address this issue, we propose a statistical method for detecting and correcting frequency-dependent phase differences in CIGs. After flattening the CIGs, we perform multi-scale Gaussian filtering to divide the data into narrow frequency bands, effectively reducing noise and ensuring more stable phase estimation. Then, the phase differences between the original and a reference CIG—formed by averaging multiple traces within the effective illumination range—are estimated for each narrow frequency band using a particle swarm optimization (PSO) algorithm. Treating the measured phase shift in each band as corresponding to its center frequency, we employ spline interpolation to construct a smooth, continuous phase correction curve. This curve is then applied to correct the wavelet phase across the full bandwidth. Both synthetic and field data are used to demonstrate the effectiveness of the proposed method. Experimental results show that the method effectively corrects residual phase differences in CIGs, significantly enhancing the amplitude fidelity and resolution of the final seismic image.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"246 ","pages":"Article 106112"},"PeriodicalIF":2.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146038841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-03-01Epub Date: 2026-01-07DOI: 10.1016/j.jappgeo.2026.106094
Jianguang Han , Hao Zhang
Adaptive focused beam migration represents an advanced imaging technique for seismic wave fields. This method enhances beam energy focus by dynamically adjusting beam width based on velocity information, offering superior imaging capabilities for complex geological structures. Based on previous research on isotropic adaptive focused beam pre-stack depth migration (FB-PSDM), this study introduces anisotropic ray tracing equations to improve seismic wave field imaging in anisotropic media. Furthermore, the proposed method integrates a pre-stack wave field separation technique for multi-component seismic data, resulting in the development of an anisotropic multi-component adaptive FB-PSDM approach. Comparative analysis of single-shot PP-wave and PS-wave migration results in horizontal transversely isotropic with a vertical symmetry axis (VTI) media models demonstrates that the proposed method yields more accurate imaging outcomes compared to conventional isotropic migration methods. Additional validation through PP-wave and PS-wave migration tests on complex-fault transversely isotorpic with a tilted symmetry axis (TTI) medium model and the Marmousi-2 TTI medium model further confirms the superior performance of the proposed method. These results consistently indicate that the anisotropic multi-component adaptive FB-PSDM method significantly outperforms isotropic migration methods in imaging quality for complex anisotropic geological structures.
{"title":"Multi-component seismic imaging using adaptive focused beam migration in transversely isotropic media","authors":"Jianguang Han , Hao Zhang","doi":"10.1016/j.jappgeo.2026.106094","DOIUrl":"10.1016/j.jappgeo.2026.106094","url":null,"abstract":"<div><div>Adaptive focused beam migration represents an advanced imaging technique for seismic wave fields. This method enhances beam energy focus by dynamically adjusting beam width based on velocity information, offering superior imaging capabilities for complex geological structures. Based on previous research on isotropic adaptive focused beam pre-stack depth migration (FB-PSDM), this study introduces anisotropic ray tracing equations to improve seismic wave field imaging in anisotropic media. Furthermore, the proposed method integrates a pre-stack wave field separation technique for multi-component seismic data, resulting in the development of an anisotropic multi-component adaptive FB-PSDM approach. Comparative analysis of single-shot PP-wave and PS-wave migration results in horizontal transversely isotropic with a vertical symmetry axis (VTI) media models demonstrates that the proposed method yields more accurate imaging outcomes compared to conventional isotropic migration methods. Additional validation through PP-wave and PS-wave migration tests on complex-fault transversely isotorpic with a tilted symmetry axis (TTI) medium model and the Marmousi-2 TTI medium model further confirms the superior performance of the proposed method. These results consistently indicate that the anisotropic multi-component adaptive FB-PSDM method significantly outperforms isotropic migration methods in imaging quality for complex anisotropic geological structures.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"246 ","pages":"Article 106094"},"PeriodicalIF":2.1,"publicationDate":"2026-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145928698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号