Self-Supervised Marine Noise Learning with Sparse Autoencoder Network for Generative Target Magnetic Anomaly Detection

In certain scenarios, such as detection of unexploded ordnance (UXO), submarines, and intruders, magnetic anomaly detection (MAD) is an effective method due to the magnetic field advantages of small operating power, strong penetrability, and strong anti-interference ability. However, variability in the background geomagnetic field and low signal-to-noise ratio (SNR) in MAD cannot be avoided. In this article, we propose a MAD method employing an improved orthogonal basis function (OBF) for the magnetic tensor contraction decomposition. The orthogonal basis decomposition approach makes the best use of modeling the MAD output signal, whereas the magnetic tensor contraction aid in avoiding the background field problem. Furthermore, by increasing the baseline length of the magnetic gradiometer, the SNR can also be improved in some manner. A dual magnetic gradiometer framework is introduced to fulfill the demand of the magnetic anomaly localization. From our simulation results, the proposed method improves the SNR of the MAD raw signal and is capable of detecting and locating the magnetic anomaly. Moreover, field test has been carried out to evaluate the performance of the proposed method.

For searching and detecting near-field unknown ferromagnetic targets, four automatic search algorithms are proposed based on magnetic anomaly information from any position on planes or in space. Firstly, gradient search algorithms and enhanced gradient search algorithms are deduced using magnetic modulus anomaly information and magnetic vector anomaly information. In each algorithm, there are plane search forms and space search forms considering different practical search situations. Then the magnetic anomaly space data of typical magnetic source of oblique magnetization are forwardly simulated by ANSYS MAXWELL software. The plane distributions of some variables are numerically computed and the search destinations of different algorithms are predicted. Four automatic search algorithms are applied to simulate search paths on three characteristic orthogonal planes and in whole solution space. The factor affecting the performance of algorithms is analyzed. Features of each algorithm in different conditions are analyzed and suitable applications are discussed and verified by the experiment. The results show that proposed search algorithms require few prior information and have real-time performance for searching and tracking magnetic anomaly target.

Magnetic anomaly detection (MAD) has been widely used for detecting the magnetic targets in many areas. In low signal-to-noise ratio (SNR) situations, the magnetic anomaly is usually buried in the magnetic noise, which leads to a decrease in the detection performance of traditional MAD methods. In order to improve the detection probability of MAD in low SNR, we propose an adaptive method of MAD using support vector machine (SVM). First, we design a detection model using SVM with radial basis function (RBF) kernel function. Meanwhile, we analyze the magnetic anomaly signal and determined the orthonormal basis function (OBF) energy and magnetic entropy as the features of the magnetic anomaly, which are as the input of the model. Then, the model is trained based on the data set with the magnetic signals. Finally, the trained model is applied to the raw magnetic signal to detect the anomaly. Compared to the traditional methods, the proposed method has a better performance and adaptive ability. The proposed method is validated by the real magnetic noise. The results show that the proposed method can reduce the case of false alarm and improve the detection performance in low SNR.

Extraction of the magnetic anomaly signal is one of the difficulties in the magnetic anomaly detection as the weak features extracted are easily disturbed by strong background noise. To address this problem, a sparse feature extraction method based on the tunable Q-factor wavelet transform and overlapping group shrinkage is developed in this paper. Compared with the traditional wavelet transform, the proposed method has excellent characteristics, which can flexibly tune the Q-factor according to the oscillation characteristics of the useful features. In this way, the sparsity of extracted features can be induced more effectively. In addition, the non-convex overlapping group shrinkage can effectively extract weak features from signals with group property, enhancing the extraction accuracy of features. The practical experiment verifies the effectiveness of the proposed method in the magnetic anomaly detection.

There is still a lot of noise signal redundancy in the empirical frequency range of the vehicle magnetic anomaly detection (MAD) signal, and the extremely weak magnetic anomaly signal is still submerged in noise. This paper proposes a wavelet packet denoising algorithm combined with the denoising algorithm of the upper limit frequency of the magnetic anomaly feature. And because it is difficult to describe the magnetic anomaly signal in the characteristic layer in a single dimension, a multi-feature extraction method is used to process the magnetic anomaly signal. For the denoised magnetic anomaly signals, the dimensionless constants and singular values are extracted to construct feature vectors, and the support vector machine is used to classify and recognize the magnetic anomalies. Finally, through experiments, for the weak magnetic anomaly signal with an amplitude of the only 0.04nT, an accuracy of 80% can be achieved, and a good recognition effect has been achieved.

Magnetic anomaly detection is a passive and effective approach for detecting magnetic targets. During detection using one scalar magnetometer, temporal magnetic variations will affect the detection performance. In order to reduce the influence of magnetic variations on detection, we focus on the analysis of magnetic gradient signals and obtain four orthonormal basis functions (OBFs) of gradient signal. Based on the set of OBFs, we propose an improved method using the magnetic gradient signal. Meanwhile, we can also obtain the bearing of the target relative to the survey line in the method. The proposed method is validated by simulation and real-world magnetic noise. The results show that the proposed method has a good detection performance with low signal to noise ratio. Therefore, the proposed method can improve the detection probability of weak magnetic anomalies and also determine the bearing of targets relative to the survey line.

This study investigates magnetic field anomalies as precursors to the Mw7.7 Myanmar earthquake (March 28, 2025) using Swarm satellite data. We analyzed vector magnetic field (X, Y, Z) measurements over the Dobrovolsky region during the 10 days preceding the earthquake, applying cubic-spline filtering to identify residual anomalies. Our results revealed Y-component anomalies in 22 of 85 half-orbits, detectable up to 8 days before the event. Magnitude estimates were obtained using four empirical equations based on anomaly parameters such as duration, amplitude ("energy"), timing, and satellite-epicenter distance. Among these, the distance-based relation provided the most accurate estimate (M ≈ 7.2), showing reasonable agreement with the observed magnitude (M = 7.7). Interestingly, anomaly energy values consistently clustered within a narrow range (570-577), which may represent a characteristic signature of earthquake-related anomalies. These findings suggest that satellite-based magnetic anomalies could contribute to short-term earthquake forecasting systems, although further validation with larger datasets is needed to assess the robustness and general applicability of the approach.

Due to multi-collinearity of the traditional aircraft's magnetic inference model, it is difficult to estimate the parameters of model accurately. And the magnetic anomaly detection (MAD) in airborne is affected too. Because of the time-domain characteristics of magnetic target, the magnetic signal of target was considered as a kind of short-time signal. Using the statistical characteristics of aircraft's background noise, the method of adaptive detection removed magnetic noise adaptively. Based on the differences between statistical characteristics of target and noise, the detection factor proposed has strong magnetic anomaly detection capability with background of strong complex noise. The method could detect a magnetic anomaly less than 1 nT in 500m away. More than one target could be detected at the same time. The average of factor in different reference signal has better resolution than single factor in detection. Besides adaptive detection simplifies the compensation process and improves the detection efficiency of aeromagnetic survey.

The plate tectonic history of the Easter microplate has been reconstructed by “closing” the microplate in a series of steps using the Pacific‐Nazca magnetic anomalies north and south of the microplate and the NUVEL 1 global plate motion model. After each step, the Easter microplate was rotated rigidly to match the Nazca and Pacific anomalies. Gaps and overlaps formed by this kinematic treatment indicate compressional and tensional deformation, respectively, and show that rigid plate motions are insufficient to explain the complete tectonic evolution. Analysis of the magnetic anomaly data was guided by contoured SeaMARC II, Sea Beam, and 3.5‐kHz bathymetry data and a lineament map derived from SeaMARC II side scan and Sea Beam bathymetry data. The patterns of lineaments and bathymetric structures suggest that rotational deformation of the Nazca plate is the general mechanism that accommodates the space problems arising from transfer of the Nazca plate to the microplate and rapid rotation of the microplate against the Nazca plate. Similar but smaller amounts of deformation are predicted along the southern boundary of the microplate. Prior to the origin of the microplate, the East Pacific Rise (EPR) was offset in at least two places according to the older magnetic anomalies, yet there is no evidence of linear fracture zones within the sparse data set except for occasional small consistent changes in regional depth across these age offsets. The magnetic, bathymetry, and satellite altimetry data indicate that the microplate initially formed at (or perhaps southeast of) Easter Island near a left‐lateral offset of the EPR sometime between anomaly 3 and 3'. The East Rift started propagating north from the present location of Easter Island at ∼4.5 Ma, which is ∼1.5 m.y. earlier than previously proposed. However, the magnetic data that support this interpretation are sparse and complicated by recent volcanic flows and associated rough bathymetry west of Easter Island. The geometry of the microplate changes very rapidly during its evolution. At the initial stages of development, the microplate resembles a large propagating rift system, suggesting that deformation may have been occurring throughout most of its interior up to about 2.47 Ma. At this time, the length to width of overlap ratio of the two rifts reaches a value of 3, the northward propagation slows down, the curved opening of the Southwest Rift becomes well established, and rigid rotation of the previously deformed transferred lithosphere probably starts to predominate. At this time, the offset distance between the two overlapped rifts starts to increase. Some time after 2.47 Ma and before 1 Ma, the East Rift starts propagating northwestward, probably in response to the microplate rotation, and continues up until present. Also during this time period, the East Rift breaks into a series of northward propagating rifts, each propagating into the microplate interior, thereby transferring lithosphere from the microplate to the Nazca plate and reducing the total growth rate of the microplate.

Magnetic modulation methods especially Micro-Electro-Mechanical System (MEMS) modulation can improve the sensitivity of magnetoresistive (MR) sensors dramatically, and pT level detection of Direct Current (DC) magnetic field can be realized. While in a Low Frequency Alternate Current (LFAC) magnetic field measurement situation, frequency measurement is limited by a serious spectrum aliasing problem caused by the remanence in sensors and geomagnetic field, leading to target information loss because frequency indicates the magnetic target characteristics. In this paper, a compensation field produced with integrated coils is applied to the MR sensor to remove DC magnetic field distortion, and a LFAC magnetic field frequency estimation algorithm is proposed based on a search of the database, which is derived from the numerical model revealing the relationship of the LFAC frequency and determination factor [defined by the ratio of Discrete Fourier Transform (DFT) coefficients]. In this algorithm, an inverse modulation of sensor signals is performed to detect jumping-off point of LFAC in the time domain; this step is exploited to determine sampling points to be processed. A determination factor is calculated and taken into database to figure out frequency with a binary search algorithm. Experimental results demonstrate that the frequency measurement resolution of the LFAC magnetic field is improved from 12.2 Hz to 0.8 Hz by the presented method, which, within the signal band of a magnetic anomaly (0.04-2 Hz), indicates that the proposed method may expand the applications of magnetoresistive (MR) sensors to human healthcare and magnetic anomaly detection (MAD).

Magnetic anomaly detection (MAD) is widely applied in the fields of resource exploration, hidden target detection, and explosive ordnance disposal. Traditional methods, such as orthonormal basis functions (OBFs), are proposed to extract anomaly signals from ambient noises and device noises. Due to the weakness of the signal, the detection probability has always been limited by a low signal-to-noise ratio (SNR). To surmount the limitation, a full connected neural network (FCN) with OBF features is trained to do the detection. Nonetheless, its effect is not reliable enough under a low SNR, and it is sensitive to the orientations. This letter introduces a multifeature fusion-based neural network with three subclassifiers to conduct MAD. The first subclassifier uses the time–frequency feature, the second uses the statistical feature, and the last concentrates on the magnetic moment feature. The outputs of the subclassifiers are analyzed synthetically by weighted voting, and the optimized weights are picked on the basis of individual performance. The real noise is recorded by experiments to test the performance of our network. The result indicates that a multifeature-based neural network shows a higher detection probability than the ordinary FCN by 5% medially. At very low SNR, the multifeature-based neural network can achieve a detection probability 13% higher than FCN. Sensitivity to orientations is also improved by the multifeature-based neural network.

<p>Quantifying past motions of tectonic plates in the southwest Pacific is important because the Pacific-Antarctic ridge is the only non-destructive boundary between the Pacific plate and other major plates. However, formation of sea-ice near Antarctica impairs the collection of magnetic anomaly data that are necessary to calculate plate rotations. A detailed analysis of all ship-track magnetic data available in the southwest Pacific (61 cruises, 153 profiles, including several cruises collected after 1995) is presented here. Four different sources of uncertainty are quantified: (1) confidence of magnetic anomaly identification, (2) magnetic reversal location picking precision, (3) ship navigation precision, and (4) magnetic data quality. Finite plate rotations are calculated for the southwest Pacific (42.5 to 79 Ma) using the resulting magnetic anomaly database (1528 magnetic reversal data). Finite rotations were calculated using the Hellinger criterion, or by a new method presented here that assumes orthogonality between fracture zones and ridge segments. The new method requires less parameters and is hence able better estimate rotations in cases with an uneven distribution of sparse magnetic data. Rotations and formal uncertainties are calculated for thirty-one chrons (c20y to c33o). They confirm the existence of a three plate system (Pacific, Marie Byrd Land, Bellingshausen) in the southwest Pacific from before c31o (68.7 Ma) until c28y (62.5 Ma). After c28y, the Bellingshausen and Marie Byrd Land plates moved as a single plate.</p>

Geothermal zones and hot mineral springs are primarily associated with magmatism along the Eurasian Continental Margin. In our study, we examine the correlation between geothermal anomalies observed in Bulgarian territory and the magnetic anomalies produced by magmatic and metamorphic bodies. Specifically, if these magnetic sources are of relatively recent origin, they could contribute to heating the surrounding geological environment, thereby influencing geothermal patterns. We use geological and hydrological information, along with temperature distribution at depth, to clarify the geological environment. Next, we apply geophysical data processing and direct inversion techniques to compute the total field modulus and Euler solutions, refining the interpretation of magnetic sources and enabling a more precise evaluation of subsurface magnetic structures and their potential impact on geothermal activity. A thorough analysis of the geomagnetic field, including over forty described anomalies, is presented for the examined territory. The correlation between geothermal and magnetic anomalies is confirmed using the Chi-square test. The null hypothesis is rejected, and the zones that significantly contributed to this result are outlined. Our study identifies seven regions with the strongest correlations in terms of temperature and magnetic anomalies: (1) A broad area in the western part of the Moesian platform near Kozloduy, characterized by increased temperatures and moderate magnetic anomalies; (2) A well-defined area east of Sofia, the capital city, which features high temperatures and intense magnetic anomalies, as well as numerous hot springs; (3) The region north of Blagoevgrad, located at the foothills of the Rila Mountains, where the hottest spring, Sapareva Banya, can be found; (4) The lower course of the Struma River, near the town of Petrich, where higher temperatures coincide with a distinct group of magnetic anomalies caused by granite bodies locally enriched in ferromagnetic iron minerals; (5) A local area near Velingrad exhibiting high temperatures alongside magmatic bodies, set in a complex tectonic environment; (6) A confined zone to the north of Dospat, where rhyolites are exposed along a significant fault line; (7) A broad zone with the highest temperatures (exceeding 100 °C at a depth of 1,000 m) that aligns with magnetic anomalies from extensive outcrops of Precambrian metamorphic rocks. We analyze these relationships in the context of identified hydrogeological zones, providing a detailed examination of the spatial distribution of geothermal and magnetic features. Furthermore, we explore the correlation between these anomalies and the depth to the Curie point, as inferred from magnetic data, to better understand the thermal and magnetic structure of the region. Graphical Abstract The present study aims to identify the spatial distribution of geothermal zones in Bulgaria that coincide with the appearance of magnetic anomalies. To delineate geothermal anomalies, we use temperatures at a depth of 1000 m below the surface, obtained from temperature logs of wells. Magnetic anomalies are calculated from the vertical component of the anomalous geomagnetic field across Bulgaria. Additionally, we support our interpretation with geological information and hydrogeological zoning of mineral waters, including the distribution of hot springs. Two types of analysis are performed: (1) geophysical data processing in terms of magnitude calculation and the direct inverse method of Euler deconvolution, which outlines the magnetic sources, and (2) statistical examination using a Chi-square test. A uniform grid with over 200,000 cells is generated, with each cell containing geothermal and magnetic data organized by classes (levels). A contingency table is compiled that records the frequency of each combination of geothermal and magnetic classes. The Chi-square test results indicate a highly significant statistical correlation between geothermal and magnetic anomalies. Consequently, we reject the null hypothesis and conclude that geothermal activity and magnetic variations are closely linked, possibly due to shared geological structures, heat-altered magnetism, or fault systems. The cells that contribute most to this correlation are identified and discussed.

    Magnetic anomalies commonly contain anomalies generated by crustal rocks that have variable mineral compositions and natural remanent magnetizations. Understanding the magnetic susceptibility, remanent magnetization, magnetization direction, and distribution is important for studying the spatial location, formation, and evolution of underground rocks. However, superposition of magnetic anomalies leads to nonnegligible errors of inversion and interpretation. To overcome the interpretation problems caused by source interference, it is necessary for the target magnetic anomaly to be extracted from the observed magnetic anomaly data. Different methods have been developed to separate the magnetic anomalies of different sources using the spectral differences of regional and residual anomalies. Such methods, which include matched filtering, Wiener filtering, and wavelet analysis, have been successfully applied to solve many geological problems. However, these methods cannot extract the anomalies caused by the interference of sources at similar depths because the spectra of the target and residual magnetic anomalies are similar. Effective techniques to obtain additional magnetic information regarding the distribution of rocks at different layers and with different magnetization directions remain lacking.    Unlike existing regional-residual separation methods used for separating superimposed magnetic anomalies caused by sources with a large depth separation, this study focuses on magnetic anomalies generated by variability of the magnetic parameters and source interference with and without depth differences. We propose a new and useful method for extracting a target magnetic anomaly from an observed magnetic anomaly field. An optimization scheme is proposed for approximating the low-rank component of an observed magnetic anomaly field on the basis of low-rank theory. The magnetic dipole layout is added as a constraint based on the assumed source location. The optimal magnetizations of the magnetic dipoles are then obtained to minimize the objective function. The sum of the magnetic anomalies generated by the magnetic dipoles is calculated as the target magnetic anomaly. The synthetic and field data experiments indicate that the proposed method can accurately and robustly recover target magnetic anomalies. In the field data experiments, the magnetization information of the various isolated sources is derived via 3D fuzzy C-means inversion using the extracted magnetic anomalies. The results show that the proposed method can extract the geometric and physical information of each target magnetic source, even when the observed magnetic anomaly field is generated by various superimposed sources with target source interference at similar depths. The proposed method has the potential for dealing with the separation problems of potential field data with different scales, including the separation of the geomagnetic core field and the lithospheric magnetic field as well as the extraction of target magnetic anomalies from satellite magnetic measurements. Therefore, this approach could be of great importance for geological investigations and mineral exploration.

Gülbahçe Fault Zone (GFZ) is a significant tectonic structure and seismic source for İzmir city and its surroundings. The major earthquakes occurred at the surroundings of GFZ are 2005 Seferihisar, 2020 Samos Island-Aegean Sea and 2021 Seferihisar earthquakes. In this study, the crustal thinning and tectonic stress distribution of GFZ and its surroundings were analyzed by using the new gravity and magnetic data. The correspondence analysis was applied by appraising together the magnetic and free-air anomaly data. Moreover, the average density of the subsurface structure was calculated by applying the Parasnis method. The deviatoric stress calculation was executed and the change of the physical properties that controls the gravity and magnetic anomaly of the crustal structure of the GFZ and its surroundings from north to south was investigated. Therefore, the average density values, which were obtained from gravity anomalies, were computed as 2.59 gr / cm3 in the north of the study region and it decreased to 2.06, 1.8 and 1.49 gr / cm3 towards to the south. The freeair gravity anomaly values were between 0-70 mGal and the magnetic anomalies were between -450-150 nT. The deviatoric stress values were between -0.2-0.1 MPa.