Journal of Applied Geophysics最新文献

Land uses have a large impact on the magnetic properties of soil. Understanding the changes in such properties caused by different land uses will help to correctly explain and apply magnetic parameters. In this study, the magnetic susceptibility, magnetic mineral types, and magnetic domain state of iron oxide particles in soil after the transformation of a forest into a tea garden and cornfield in a tropical area were compared. 1)The vertical distribution characteristics of soil magnetism under the three land uses in tropical red soil area were similar. The magnetic properties of soil in the 0–20 cm surface layer was significantly different, but there were no significant differences below 20 cm. The low-frequency mass magnetic susceptibility and anhysteretic susceptibility of topsoil were largest in the cornfield, followed by the forest and tea garden, and the magnetic susceptibility of the topsoil in the cornfield was significantly higher than in the tea garden and forest. 2)The magnetic minerals in tropical red soil area were mainly ferrimagnetic minerals (maghemite and magnetite). The content of ferrimagnetic minerals decreased with depth, while the content of antiferromagnetic minerals (hematite) increased. 3)The magnetic particles were mainly composed of superparamagnetic (SP), single domain (SD), and pseudo single domain (PSD) particles. The magnetic differences in the topsoil were mainly caused by changes in the ferrimagnetic mineral concentration of SD particles. Tea garden and cornfield are the two main agricultural land in the study area. Through the correct application of magnetic parameters, it is helpful to monitor the influence of land use change on soil properties, so as to provide the necessary reference for the rational use of soil.

To analyze the hazard-causing modes of landslides, this paper proposes a three-dimensional discrete element model reconstruction method that employs an unmanned aerial vehicle survey and multi-electrode resistivity tomography method. To convert the resistivity profile into a material profile, we adopt the peak of the probability density method for material classification and utilize the Haar wavelet transform for image denoising. Subsequently, inverse distance weighting interpolation and the curtain-point method are used to transform two-dimensional profiles into a 3D visualization model. Similarly, the triangular mesh boundary can be extracted from the 3D visualization model using the curtain-point method. A mapping function f including the macroscopic parameters, was defined to populate the particles within the boundaries. Using the iterative method and defining the loss function L for parameter calibration, the targeted 3D discrete element model was constructed after setting the velocity threshold. This method was applied to the Changhe landslide (September 14, 2019) in Gansu Province, China, which had a typical damaged soil layer due to earthquake and rainfall factors. The results indicate that the lower part first exhibits significant displacement, followed by the upper and middle parts, which is consistent with the on-site inspections and UAV findings.

Uniaxial cyclic loading compression experiments are conducted with synchronous monitoring of acoustic emission (AE) and electromagnetic radiation (EMR). The AE and EMR characteristics of rocks during cyclic loading are analyzed. Subsequently, the Kaiser and Felicity effects of AE and EMR are disclosed. Moreover, the b-value and fractal dimension are calculated to explore the cracking mechanism of rock materials. These results show that the AE and EMR are active during the 1st and 2nd loading cycles, and sparse during the 3rd and 4th loading cycles where the AE and EMR are only generated when the stress exceeds the previous peak stress due to the Kaiser effect. In the 5th loading cycle, AE and EMR accelerate and peak at rock failure. The variation of EMR-based FR is consistent with that for AE under cyclic loading. Hence, the FR associated with EMR is also well correlated with stress memory. The AE and EMR-based FR is greater than 1 in the 2nd-4th loading cycles, and less than 1 in the 5th cycle, indicating that FR can be used to reflect the damage evolution of rocks. The b-value and fractal dimension increase in the 1st-4th loading cycles, and turn to decrease in the 5th loading cycle. This pattern serves as a precursor to identify the rock failure.

Integrated interpretation has been a cutting-edge approach in three-dimensional (3D) mineral potential mapping (MPM) in recent decades. This research presented a multisource geo-data fusion and integrated interpretation approach for 3D MPM. Thereinto, the favorable geological bodies and faults were scored according to their closeness of the relationship with mineralization. The soil geochemical data and Landsat 8 OLI imagery was processed by orthogonal factor analysis (FA) to explore the mineralization-related geochemical factors and to extract hydrothermal alterations, respectively. The aeromagnetic data was denoised by multifractal singular value decomposition (MSVD) and was extracted of the residual anomalies by bi-dimensional empirical mode decomposition (BEMD). In this way, these multisource geo-anomalies were fused by TOPSIS algorithm and the fusion result was segmented by concentration-area (C-A) multifractal model for targets mapping. Then a wavenumber-domain fast 3D-inversion approach of MagFInv3D was applied in inverting the residual magnetic anomalies. At last, the targets, occurrences and regional faults along the inverted underground residual anomalies were integrated together for interpretation. It proved that the proposed approach could show high performance in 3D MPM, which could provide a reference for 3D mineral exploration, especially in a tectonically-controlled and polymetallic metallogenic belt.

We interpret the magnetic anomalies for exploration and characterization of the Galinge iron polymetallic deposit, which is the largest iron polymetallic deposit in Western China. This undeveloped deposit is completely concealed by the thick Quaternary sediments, presenting an exploration challenge. A lot of drillholes have been implemented focusing on the strong magnetic anomaly centers, while the relationship between magnetite-bearing and non-magnetite-bearing drillholes is unclear, and the metallogenic regularity in the whole area needs to be further studied. The edge detection is applied on the downward continued field to resolve individual magnetite ore bodies and enhance possible concealed ore bodies. Three anomaly belts trending NE-SW with a total extension length of ∼11 km are first revealed. Combining the sharper images from the enhanced magnetic anomaly and the drillholes, we infer that the deposits would be related to the syncline structure. Forward modeling confirms the inferred three belts of the ore bodies and also shows possibly two layers deposits in which the deep layer would be located at a depth of ∼1000 m. Although the deeper deposit results in a very smooth and low amplitude magnetic anomaly, the forward modeling shows that the anomalies from both the shallow magnetite and the inferred deep one fit the observed anomaly better. One deep drillhole reveals the fact that the magnetite is existed at a depth of ∼1000 m which highlights the interpretation of the deep deposit. The results show the contribution of the fine processing techniques which would be used to distinguish zones of potentially concealed mineralization.

The municipality of Castrocielo (Frosinone, Italy) is a historically significant center which includes several centers of great archaeological importance, including part of the archaeological site of the ancient Roman city of Aquinum. In this work, we show the results of geophysical surveys performed in two different areas: the first area is close to the Monacato of Santa Maria al Palazzolo, built on the foundation slab of a Roman villa dating back to the 1st-2nd century BCE; the second area is close to the charging station Casilina Est, where several burials, dating back to different periods from 4th century BCE to 4th century CE, were found. The aims of geophysical investigations is to identify structures linked to the ancient Roman villa (Villa Eucheria) in Area 1, and to identify the tombs of the necropolis in Area 2. The two areas were investigated in two different days, on 27th and 28 ^th March 2023 respectively, through a multi-channel georadar system (GPR). In the second area, an electro-magnetometric survey was also performed. This choise is to address the heavy rain developed during the night before the acquisition. Infact the GPR survey performed during the second day of the geophysical campaign did not provide good results.Based on the geophysical results, the archaeological excavation in Area 1 confirmed the detected anomalies, documenting a section of wall and other structures and elements brought to light over a length of approximately 9 m. The results obtained in Area 2 confirmed the cropmarks visible in the aerial photo, highlighting the traces of buried structures.

In the present work, we present a 2D pixel-based Finite Element strategy to simulate the elastic wave propagation in heterogeneous media. An assembly-free approach is employed for the stiffness matrix, leveraging a pixel-based structured mesh to reduce the memory required to store computations. Additionally, a diagonal Lumped-Mass matrix technique is utilized to address challenges associated with the inversion and storage of the mass matrix. The Leap-frog integration method, known for its amalgamation of stability, precision, and efficiency, is adopted. The combination of these features is aimed at facilitating massive parallel computations for very large systems with 10⁸ to 10⁹ degrees of freedom. In that sense, the present work can be understood as a first step toward a very efficient massive parallel GPU-based voxel-based Finite Element implementation to treat very large digital images with personal computers. The implementation presented here has been validated against theoretical predictions and analytical results derived from classical wave propagation theory. Finally, the transmission test is simulated in two digital models, one representing a layered medium and another representing a medium with complex microstructue obtained via micro-tomography. For the first model, the results are compared with the so called Bakus average, while, for the second model, the results are compared with the corresponding outcomes acquired through an in-house developed static finite element homogenization implementation.

Opposing coils transient electromagnetic method (OCTEM) adopts small and weak-coupling transmitting-receiving coils configuration, which helps to reduce side effect and improve the detection resolution. With such advantages, it has been widely used for shallow sub-surface target detection. However, when the method is used in urban area, measured data may be distorted by electromagnetic (EM) interference from nearby metal objects. In practical application, it is necessary to perform modelling to provide guidance for measuring data analysis. Two OCTEM application cases in detecting shallow sub-surface karst caves in urban area with metal objects nearby are presented in this paper. Corresponding modelling are carried out to study the interference effect of the nearby metal objects. The first application case is about the EM interference of a vertical steel tower, of which the influence distance reaches up to 9 m by modelling. The second application case is about the EM interference of a thin aluminum fence, of which the influence distance reaches up to 6 m by modelling. Only when the observation is outside the influence zone, the metal influence can be ignored. When the measurement is inside the influence zone, the metal influence cannot be ignored. However, as the nearby metal objects mainly affects the early data, the subsurface target may also be detected in condition that the target response is stronger than the metal interference, or the target response time window is wider than that of the metal interference.

The self-potential (SP) method is a classical geophysical exploration method which has a wide application prospect in underground pollutant monitoring and other fields. However, due to the complexity of the formation mechanism and the lack of prior information, there are still quite a few difficulties in the precise quantitative inversion of the SP sources, and qualitative interpretation is frequently adopted in practical applications. In this work, we carry out inversion research on the SP data of geomicrobes to accurately invert and locate the spatial distribution of the SP sources which is closely relevant to microbial activities. The resistivity-based depth weighting matrix is added to the inversion algorithm to promote the migration of the SP sources from the earth surface to their original depth. And to conform to the actual distribution of the SP sources, the minimum support stabilizing function is introduced to impose additional compact constraint. Two synthetic models are firstly designed to verify the effectiveness and accuracy of the proposed algorithm. On this basis, the sandbox experiment that continuously observes and records the SP signals generated by the typical organism: Shiwanella Oneida MR-1 breaking down the organic matter is carried out. Then the observed data is inverted to locate the SP sources. The inversion results demonstrate that with the addition of Shiwanella Oneida MR-1 into the humus, the negative SP source immediately appear on the top of the humus, which increase sharply, then remain stable and then slowly decay over time. The negative SP sources are concentrated on the top of the humus, which is consistent with the theoretical analysis of the biogeobattery model.

Deep learning is one of the best machine learning algorithms for modeling complex mapping relationships between independent and dependent variables, and thus it can be viewed as an ideal approach to predict porosity. In this study, to overcome the deficiencies in current porosity prediction based on deep learning and improve the prediction accuracy, we proposed a deep learning model based on bidirectional temporal convolutional network (BTCN) and bidirectional long short-term memory (BLSTM) network, called bidirectional spatio-temporal neural network (BSTNN), to establish a porosity prediction model. First, the maximum information coefficient is used to analyze the correlation between well logs and porosity, which provides a basis for determining the inputs of the prediction model. Then, a hybrid network structure is constructed by using BTCN and BLSTM, in which BTCN goes to learn the bidirectional long sequence features and BLSTM goes to learn the variation trend and context information with depth, so the hybrid network structure can learn richer logging signal features. Finally, the extracted features are passed through the fully connected layer to output the porosity prediction results. Porosity prediction experiment are conducted by using the actual field data set. The results show that the proposed method has the lower prediction errors for the porosity modeling (RMSE = 0.368 and MAE = 0.260) compared to the benchmark models convolutional neural network (RMSE = 0.404 and MAE = 0.292) and long short-term memory network (RMSE = 0.418 and MAE = 0.298), which verifies the effectiveness of this prediction method.