Applied Geophysics最新文献

Rapid societal development and increased engineering construction have exacerbated the disturbance of the geological environment. The impact of extreme climatic factors has grown, leading to a surge in geological disasters, with landslides emerging as particularly significant. Consequently, fundamental research in geological disaster detection or monitoring necessitates an in-depth study of the physical phenomena accompanying landslides’ development, evolution, and occurrence. Exploring the signal characteristics associated with landslides is crucial to indirectly understanding their development and change processes—a scientific question deserving thorough exploration. Despite this research’s importance, there is a notable gap in the investigation of the key design and specific implementation of electromagnetic instruments tailored for landslide detection. This gap is particularly pronounced in designing and implementing data acquisition software for electromagnetic instruments. This interdisciplinary research draws on theoretical frameworks from embedded computer science, software engineering, digital signal processing technology, geophysics, and engineering geology. It focuses on developing specialized data acquisition application software for landslide detection or monitoring, contributing to the scientific understanding of landslide development and providing independent intellectual property in the electromagnetic wave signal detection field.

The classification and distinction between nuclear explosions and natural earthquake events are essential to the Comprehensive Nuclear Test Ban Treaty. Nuclear explosion data are lacking; thus, classification problems must be studied in small sample scenarios. The classification problem of the eXtreme Gradient Boosting (XGBoost) model in one small sample scenario is examined using the sparrow search algorithm (SSA) algorithm to optimize the key hyperparameters of the model automatically. The shortcomings of SSA are addressed by using a Gaussian chaotic mapping method, introducing a population proportion dynamic adjustment strategy, and proposing a step-size adjustment factor for modification. The problem of the uneven initial population distribution is addressed by constructing the (modified SSA) MSSA–XGBoost classification model, thereby reducing population diversity and affecting the convergence speed of the algorithm. The fixed proportion problem of the sparrow population, which easily falls into the local optimal solution, is solved using the aforementioned approach. The fixed update step position of the discoverer is also resolved, thus limiting the global search capability and optimization efficiency of the algorithm and realizing the independent optimization of three important hyperparameters. Furthermore, artificial feature extraction can be avoided using this approach, and the number of iterations, maximum tree depth, and learning rate can be automatically optimized, achieving excellent results in small sample seismic event classification. Experimental results reveal that the classification accuracy of the MSSA–XGBoost model is 96.37%, demonstrating its superiority to the original model (93.47%) as well as to the support vector machine and convolutional neural network. Meanwhile, a nearly 30% improvement is observed in computational efficiency.

Accurately estimated quarry blast equivalents can be compared with the quantity of initiated explosives to detect misfires or undetonated explosive remnants, thereby ensuring safe mining operations. Seismic waves are commonly used to estimate the equivalent; however, the ability of permanent seismic networks to detect low-magnitude events is limited. Therefore, we conducted experiments in the Minglongshan–Shangyao fault segment of the Tan–Lu fault zone in the Chuzhou area of Anhui Province in China, deploying six shallow-borehole short-period seismometers at a burial depth of 1 m for continuous monitoring for 20 days. Forty-two earthquakes were detected using a convolutional neural network, and the detected earthquakes were processed based on the source location, frequency spectrum analysis, time of occurrence, equivalent, and magnitude statistics. Through comparison, we found that one of the earthquakes was the M3.1 Suqian earthquake on March 19, 2022. Unlike this natural earthquake, the remaining 41 earthquakes have not been reported by any seismic network. The source location is concentrated, the frequency spectrum is simple, and the time of occurrence is concentrated in the daytime. Based on these results and the widespread quarries in this area, we speculate that these 41 earthquakes were caused by artificial blasting. Through seismic array monitoring, the precise locations of quarry blasts can be determined. Furthermore, the seismic wave energy-blast equivalent algorithm can be used to accurately estimate the quarry blast equivalent.

Abstract

Microseismic monitoring technology is widely used in tunnel and coal mine safety production. For signals generated by ultra-weak microseismic events, traditional sensors encounter limitations in terms of detection sensitivity. Given the complex engineering environment, automatic multi-classification of microseismic data is highly required. In this study, we use acceleration sensors to collect signals and combine the improved Visual Geometry Group with a convolutional block attention module to obtain a new network structure, termed CNN_BAM, for automatic classification and identification of microseismic events. We use the dataset collected from the Hanjiang-to-Weihe River Diversion Project to train and validate the network model. Results show that the CNN_BAM model exhibits good feature extraction ability, achieving a recognition accuracy of 99.29%, surpassing all its counterparts. The stability and accuracy of the classification algorithm improve remarkably. In addition, through fine-tuning and migration to the Pan II Mine Project, the network demonstrates reliable generalization performance. This outcome reflects its adaptability across different projects and promising application prospects.

In January 2020, two moderate earthquakes with magnitudes of M_S 5.4 and M_S 6.4 occurred in the nappe structure of the boundary between the Southern Tianshan Mountains and Tarim Basin. To investigate the seismogenic structure of these two events, we systematically analyzed the earthquake sequence locations, focal mechanisms, and stress field in the source region. Using the P and S arrival-time data from January 1, 2009, to July 31, 2021, recorded by 18 seismic stations of the Xinjiang network, we obtained precise seismic event locations. Results show that the temporal and spatial distribution of the foreshock and aftershock sequences displays obvious differences in migration behaviors. The former is mainly distributed along the NNW direction, whereas the latter is distributed along the Ozgertau fault in the EW direction and characterized by a double-layer feature. Furthermore, we derived the focal mechanism solutions of the M_S ≥ 4.0 events, including the M_S 5.4 foreshock and the M_S 6.4 main shock. The inversion results illustrate that the M_S 5.4 foreshock is a strike-slip event, whereas the main shock and seven aftershocks are thrust events. Based on our results and the regional geological background, we conclude that the seismogenic structures of the M_S 5.4 foreshock and M_S 6.4 main shock are related to different faults. The M_S 5.4 event is located in a blind “quasi-transformation” fault with the NNW direction, and the M_S 6.4 earthquake occurs on a blind thrust nappe fault in the EW direction on the Kepingtage (Kalpintag) nappe. Our results indicate that the Tienshan orogenic belt has a complex tectonic environment. The difference in the strikes and dips of the M_S 5.4 foreshock and M_S 6.4 main shock reflects the stress compression in the near-NS direction in the source region.

Using the repeated survey data from field stations and the weighted least-squares method, this study obtained the long-term correction results of the measurement data and compared these results with those of current methods. The results were as follows: 1. The new method substantially improved long-term spatial distortion compared to the old method. The secular variation (SV) results of the main geomagnetic field produced by the new method have a higher linear correlation to those of the international geomagnetic reference field (IGRF)_SV model. The mean square error (MSE) of the difference in the three elements F, D, and I between the new method and the IGRF_SV model is 10.7%, 47.0%, and 14.5% of that of the original method, respectively. 2. Applying the new SV correction method obtains more stable and reasonable variations in Earth’s crustal magnetic field. The average amplitude of the Earth’s crustal magnetic field variation in F, D, and I is 28.5%, 55.4%, and 34.4 of the original results, the MSE is 59.1%, 56.5%, and 40.3% of the original results, and the mean gradient is 93.6%, 91.9%, and 97.0%, respectively. 3. In the processed results of the new method, the seismomagnetic information is clearly optimized, and the epicenter location is more consistent with the 0 value line of the Earth’s crustal magnetic field. The processed results of the new method are considerably better than those of the original method and have a higher application value.

The Qiangtang Basin, located in the Tibetan Plateau, is an appropriate area to verify the Lhasa–Qiangtang collision, which was recorded by the middle-upper part of the Yanshiping Group (the Xiali and Suowa Fms) in the basin. However, the chronology of the Xiali and Suowa Fms is HYPERLINK “javascript:;” controversial, which limits comprehending the timing of the Lhasa–Qiangtang collision. More importantly, HYPERLINK “javascript:;” oil HYPERLINK “javascript:;” shale and salt springs were exposed in the Xiali and Suowa Fms in the basin. 544 paleomagnetic samples were collected from the Yanshiping section in the basin in order to reveal the timing of the Lhasa–Qiangtang collision from the view of paleomagnetic ages of the two formations. However, we did not give credible magnetostratigraphic ages of the two formations because of ammonite fossils, a global primary standard for the Jurassic strata correlation, without being found in the last study. Yin (2016) revised the long-term HYPERLINK “javascript:;” controversial paleontological age of the Suowa Fm. from a Tithonian age of the Late Jurassic or a Berriasian age of the Early Cretaceous, to a Middle Bathonian–Middle Callovian age of the Middle Jurassic based on new ammonite fossils. Considering ammonite fossils as a powerful tool and a global primary standard for the Jurassic strata correlation, we attempted to correlate the last magnetostratigraphy with the GPTS 2012 again. Magnetostratigraphic ages of 164.0–160.2 Ma and 160.2–156.8 Ma for the Xiali and Suowa Fms are suggested, respectively. The timing of the Lhasa–Qiangtang collision (156.8–154.9 Ma) is revealed from the magnetostratigraphic ages and the zircon U–Pb age of the Xueshan Fm, 154.9 (+6.8/−1.6) Ma, overlying the Suowa Fm in the Yanshiping section.

Seismic trace editing is a tedious process in data preprocessing that can incur high time costs, especially when handling large 3D datasets. In addition, existing methods to edit seismic traces may miss vital information when killing noisy traces simply. Thus, in this paper, we propose an automated method to edit seismic traces based on machine learning. The proposed method combines the Hough transform technique and a convolutional neural network (CNN) to improve the feasibility of the scheme. The Hough transform is a feature extraction technique that helps identify anomaly lines in images, and we employ it in the proposed method to ascertain the prospective positions of noisy and bad traces. We then implement a bandpass filter and the trained CNN model to identify the precise noisy traces in the target region indicated by the Hough transform process. Upon identification, automated processing is applied to determine whether the processed traces can be useful or should be discarded. This comprehensive framework includes four main steps, i.e., data preprocessing, Hough transform detection, network training, and network prediction. Experiments conducted on real-world data yielded 98% accuracy, which indicates the potential efficacy of the proposed automated trace editing method in practical applications.

Applying reverse time migration (RTM) to seismic data often results in wavefield propagation fraught with migration artifacts. To overcome this, we introduce least-squares RTM (LSRTM), which is applied to the migrated section via the Born approximation and the conjugate gradient algorithm. LSRTM extrapolates the reconstructed wavefield using a wave equation that has been transformed into the Riemannian domain. This approach addresses the oversampling effect of seismic signals by ensuring even sampling and allows for the recovery of greater amplitude in the final migrated image. For each point in the Cartesian coordinate system, there is a corresponding vertical time point. Consequently, we can interpolate the reconstructed source wavefield in the new ray coordinates by drawing a Cartesian–Riemannian mapping function. The specific finite difference (FD) scheme and boundary conditions notwithstanding, the Riemannian wavefield extrapolator operates via two formulas depending on the type of wave equation used. In vertical transversely isotropic (VTI) media, velocity tends to decrease with depth, significantly distorting the migration results. This issue can be resolved by applying the LSRTM in either the Cartesian or pseudodepth domain, supported by a proper wavefield extrapolator. The finite-difference Riemannian wavefield extrapolator, when applied to the Born modeled seismic data, produces results strikingly similar to the classical LSRTM, albeit with some amplitude differences owing to various implementation issues and the oversampling effect. Our results strongly indicate that the domain transformation strategy effectively reduces computational time without compromising the accuracy of the Cartesian-mesh-typed LSRTM results.

As a significant detection task, underwater explosion events have piqued the interest of geophysicists. On September 26, 2022, two underwater explosions occurred in the Baltic Sea region involving the Nord Stream pipeline. Subsequent reports from the media and the European-Mediterranean Seismological Centre deduced that one event was detected at 17:03:49 GMT, exhibiting a magnitude of 3.1 on the Richter scale, whereas the second event lacked comprehensive information. To meticulously investigate the details of the two Nord Stream incidents, we employed 14 proximate stations for subsequent analysis. Using a linear inversion technique, we determined that the two events occurred at 00:03:30 and 17:03:49, with corresponding coordinates of 54.87°N, 15.52°E and 55.59°N, 15.80°E, respectively, and Richter magnitudes of 2.3 and 3.1, respectively. Spectral analysis corroborates that both events were underwater explosions, and the presence of anomalous signals within the frequency domain may provide valuable insights into source parameters and characteristics, particularly for the first event.