Recommendation system algorithms, having recently garnered significant attention in the field of digital Earth science, are expected to be widely applied in metallogenic prediction. Traditional metallogenic prediction studies fail to fully mine the various types of semantic information in massive geoscience data. The deep interest evolution network (DIEN), as a recommendation system algorithm, can fully mine semantic information to predict user preferences. Therefore, this study employed the DIEN model as the prediction model and the semantic information extracted from bedrock interpretation as the ore-controlling elements according to the database provided by the Western Australian government. The model was trained to perform metallogenic prediction for the study area. The prediction results indicate that 92.95% of uranium ore occurrences fell within the medium-high probability zone in the prediction map, with some unknown zones also showing high prediction probabilities. After removing known uranium ore occurrences in some zones, the retrained model still yielded medium-high prediction probabilities in these zones. The results suggest that the DIEN can effectively mine semantic information in metallogenic prediction studies, and the DIEN model exhibits strong predictive capacity for the study area, providing a novel approach for metallogenic prediction studies.

The Mianning-Dechang area in western Sichuan serves as the most important metallogenic belt of light rare earth elements (LREEs) in China. To make breakthroughs in the exploration of rare earth resources in the Dechang area, a 1:50 000 stream sediment survey was conducted in this study. The analysis of test data characteristics and the extraction of geochemical anomalies reveal that the total rare-earth oxides (REOs) in the area exhibit a pronounced enrichment pattern and that element anomalies largely overlap the spatial distribution of related geological bodies. Through comprehensive analysis using the cumulative frequency method and iterative method, this study determined the lower limits of anomalies and, accordingly, plotted element anomaly maps, with five main anomaly areas being delineated. The comprehensive evaluation of anomalies revealed four prospecting target areas: Huangjiaba, Jiaobacun, Yibasan, and Huajiaoyuan. The analysis of metallogenic geological conditions and the anomaly verification based on drilling in the Gannan area led to the discovery of three light rare earth deposits in the Huangjiaba prospecting target area, two heavy rare earth deposits in the Jiaobacun prospecting target area, two heavy rare earth deposits in the Yibasan prospecting target area, and two heavy rare earth deposits in the Huajiaoyuan prospecting target area. This study posits that the Triassic biotite moyite and biotite monzogranite in the study area are the metallogenic parent rocks of ion adsorption-type rare earth resources, with REEs primarily undergoing enrichment and mineralization in completely weathered layers. Therefore, a simple and effective prospecting pattern for "endogenous and exogenous" ion adsorption-type rare earth deposits in the Dechang area consists of analyzing the geochemical characteristics of the total REOs based on a 1:50 000 stream sediment survey, delineating metallogenic prospect areas that indicate parent rocks for REE enrichment, and selecting sections favorable for the formation and preservation of weathered crusts based on the hypergenic conditions of the enrichment areas, thus achieving the quick delineation of the prospecting target areas.

Over the past century, electromagnetic exploration technology has evolved from direct current resistivity and induced polarization methods to a comprehensive geophysical system. Yet, in China's new mineral exploration phase, challenges like deep-mining needs, cultural noise, and weak 3D interpretation limit traditional methods. High-power-ultrahigh-power electromagnetic technology, by boosting transmission current, combats these issues. It enhances detection depth, enables 3D exploration, and drives technological and application innovation. This paper reviews the development of high-power-ultrahigh-power electromagnetic instruments and current research. It emphasizes that technologies like true 3D full-waveform IP collection and inversion, tensor CSAMT collection and inversion, and multi-parameter joint inversion of time-and frequency-domain EM methods can strengthen deep-target detection. Future research should tackle anisotropic 3D inversion, full-domain inversion with a field source, and extracting polarization and magnetization rates under complex constraints. These advances will propel electromagnetic methods toward greater depth, precision, and intelligence, supporting China's renewed mineral exploration efforts.

China's complicated geological background and metallogenic environment have decided that the regional geochemical exploration can play an important role in the field of China's mineral resources prospecting. This is the strategic selection under special metallogenic conditions. The first round of regional geochemical exploration of China was carried out during mid-1950s. Since then, we carried out the second and third round of regional geochemical exploration. Until now, China's regional geochemical exploration has gone through a history of 70 years, and showed unique functions in fields of the exploration and development of nonferrous metals, precious metals as well as the rare minerals, rare earth and scattered minerals. According to incomplete statistics, during 36 years (1979~2015) of the second and third round of regional geochemical exploration, totally 182 different kinds of large to super-large scale deposits were discovered in China mainly by means of regional geochemical exploration. Of which, 61 deposits have nonferrous metals, 93 deposits have precious metals, 24 deposits have critical and scarce minerals, and 4 deposits have other kinds of minerals. On average, 5.06 large to super-large deposits were discovered each year, and one large to super-large deposit was discovered within every 37,000 square kilometers-Calculated on the basis that the second and third round of regional geochemical exploration covered 6.68 million square kilometers' area of China. Regional geochemical exploration has achieved outstanding results in the field of mineral prospecting, and made great contributions in terms of ensuring and promoting China's economic development. As our country's basic, public good and strategic geological work, regional geochemical exploration comes into the 21^st century marked by big data and earth system science. Based on high precision and good quality big data, guided by metallogenic geochemical theories, and supported by modern science and technology, the regional geochemical exploration will start a new period of development. In the future, we'll continue to carry out the high technical standard and high quality regional geochemical exploration, achieve our goal of complete covering the whole area of mainland China with regional geochemical exploration, and provide more strategic prospective areas for our country's mineral resources exploration. On the basis of high-precision big data, metallogenic theories and modern science and technology, we'll establish AI and IT application system, so as to push forward the exploration and utilization of deep minerals, buried deposits as well as the critical and scarce minerals which are difficult to identify, and achieve greater breakthroughs in the geological exploration of mineral resources. By using big data technics, China's regional geochemical exploration team will continuously solve the mineral resources exploration and evaluation problems, and in this process, our geochemical exploration system will be gradually integrated into our country's science and technology advancement.

With the application of high-precision,high-density 3D seismic technology and significant improvements in acquisition efficiency,the volume of data from nodal seismic data acquisition has expanded sharply.This leads to growing demands for the quality monitoring of nodal seismic acquisition.Especially,the lag in the synthesis of common shot gather data in node data has affected the quality monitoring and processing efficiency of seismic data.Therefore,the demand for quality control technology has become increasingly prominent in nodal seismic data acquisition.By delving into technologies including the visual monitoring of pre-acquisition node states,node data synthesis,and seismic data quality monitoring,this study independently developed a quality monitoring platform for nodal seismic data acquisition,enabling quality control for the whole nodal seismic data acquisition,involving equipment status,data synthesis,and data quality control.This platform has achieved satisfactory application results in a practical seismic data acquisition project.

The Triassic strata in the Lunnan area of the Tarim Basin represent a continental deltaic sedimentary system,with sedimentary subfacies including deltaic plain,deltaic front,and prodelta.The rapid phase transition of reservoirs and the high-frequency sand-mudstone alternating deposition create thin,highly heterogeneous reservoirs,rendering high-precision reservoir prediction extremely challenging.Under the guidance of sedimentary petrology and seismic sequence stratigraphy,this study conducted comprehensive seismic and geological research.Then,based on isochronous sequence interpretation,as well as seismic facies analytical results,this study established a lithofacies model.By constraining the pre-stack geostatistical inversion process,this study determined the spatial distribution of sand bodies.The practical application demonstrates that the proposed method enhances the vertical resolution of heterogeneous,thin reservoirs while significantly reducing lateral uncertainty.The predicted results align with actual drilling and production performance data and thus can provide valuable support for the efficient exploitation of hydrocarbon reservoirs.

To evaluate the resource potential of the Jinchanshan mining area, this study conducted analysis and tests of metal element content in soil samples from two geochemical profiles using the electro-adsorption method. By combining the geological conditions of the mining area, this study revealed pronounced positive geochemical anomalies of Au, Ag, Cu, Pb, and Zn above the known gold deposits. The anomalous areas corresponded well with deposits, confirming the effectiveness and feasibility of the electro-adsorption method in the study area. The study identified geochemical anomalies Ⅰ and Ⅱ to the west of the Guaibanggou-Yangpo-Xiaoxigou mineralization zone in the first geochemical profile. Along the second profile, located between the Guaibanggou-Yangpo-Xiaoxigou and Nandawa-Limazigou-Loufengmao mineralization zones, anomalies Ⅲ and Ⅳ were observed at the westernmost end. The comprehensive geological analysis of the mining area indicates that the Ⅰ and Ⅲ anomaly zones exhibit favorable geological conditions for mineralization, serving as play fairways. The play fairway prediction conducted in this study provides targets for future exploration in the mining area.

In practical detection operations using ground-penetrating radar (GPR), factors such as environmental noise and instrument errors frequently cause signals to be mixed with substantial noise, seriously reducing signal quality and the reliability of analytical results. To address this issue, this study proposed a time-frequency peak filtering method combined with minimum mean cross-entropy (TFPF-MMCE) for denoising GPR signals. This method combined time-frequency peak filtering with the cross-entropy function, enabling effective noise suppression and precise preservation of valid signals through precise optimization of the time-frequency representation, thereby significantly improving the quality of GPR signals. Numerical simulation and field GPR experiments validated that the TFPF-MMCE method exhibited a high noise removal capability and, thus, can effectively eliminate random noise while significantly improving signal clarity and reliability. Compared to traditional denoising methods, TFPF-MMCE shows significant advantages in denoising effectiveness and noise resistance stability, suggesting promising application potential and practical value in the field of GPR signal processing.

Due to dense buildings and structures and insufficient drilling surveys, the construction of shield tunnels passing through urban karst formations that host dense buildings faces significant risks of surface fracturing and subsidence caused by karst development. Hence, this study employed the microtremor survey technology with strong anti-interference capability in complex environments to address this challenge. Based on the technology, it analyzed the structural characteristics of wave velocities in underground rock formations through the inversion of the apparent shear-wave velocity profile. Combined with geological drilling data, it inferred the bedrock interface, highly weathered unconsolidated formations, and karst cave anomaly zones. Key findings are as follows: (1) The apparent shear-wave velocities in the study area gradually increased from the shallow to deep formations. Formations with wave velocities above and below 300 m/s were inferred to be limestone and Quaternary formations, respectively, with the rock-soil interface at depths approximately between 10~15 m; (2) Seven low-value anomaly zones of apparent shear-wave velocities ranging from 150~240 m/s were interpreted. They were presumed to be unconsolidated formations or karst caves at depths ranging from 8~30 m. Relying on strong anti-interference and high accuracy, the microtremor survey technology can accurately identify the shear-wave velocity structures of underground profiles, lithologic interfaces of formations, unconsolidated formations, and karst cave anomalies. Therefore, the technology is effective in the geological exploration of urban dense building areas with karst development.

This study investigated data from a 1:50,000 stream sediment survey in the Juxian-Wulian area, Shandong Province using factor analysis. By extracting eight representative factors, this study identified element assemblage types. Then, this study performed geochemical zoning based on the factor scores and discussed the relationships between various sub-zones and their geological background. Based on the geochemical zoning results, as well as geological conditions for mineralization, this study determined favorable geological bodies and prospecting targets and discovered new prospecting clues.The most significant mineralization combination in this area is Au-Cu-Ag-Mo-Bi, followed by Pb-Zn; The V-Ti-Co-Mn combination and Ni-Cr combination can serve as exploration elements, indicating that there may be favorable mineralization mechanisms in the center and deep parts of the combination anomaly zone. Finding volcanic craters and hidden explosive breccia tubes in the distribution areas of volcanic rocks in this region or area is an important direction for mineral exploration; In the area of Shichang Fangzi Village in the eastern part of the research area, there is a comprehensive zoning with a F₂-F₄-F₁ three factor combination model. The Mesozoic Buliu sequence and Weideshan sequence rock bodies are favorable geological bodies for mineralization, and the NE trending structural belt is a favorable place for mineralization, indicating the direction of prospecting.

The deep mining of geological mineral information from geochemical exploration data has been a hot research topic in mathematical geology and geological big data. Despite China's abundant regional geochemical exploration data, the deep mining of geological mineral information from these data remains limited, necessitating a scientific, efficient, simple, and feasible data processing workflow and analytical methodology. This study investigated the Gaoqiao area in Huixian County, Gansu Province, China. According to the geological background and the theories of element properties and geochemical behavior, this study processed relevant element data to mine the geological mineral information in geochemical exploration data. It established preliminary mathematical models for the boundary delineation and lithofacies classification of intermediate-acid rock masses, and the boundary delineation of mafic volcanic rocks in the Gaoqiao area to scientifically guide geological mapping. It preliminarily established the target delineation model for tectonic altered rock-hosted gold deposits. The model's effectiveness was substantiated by seven newly delineated target areas with promising prospecting potential. The mineral inspection identified five ore occurrences, including four reported for anomalies related to mineralization, with two currently under provincial geological exploration. The results of this study show that by calculating and processing data based on the geochemical properties of elements, the hidden information in regional geochemical exploration data can be further mined to effectively guide and modify geological mapping, thereby enhancing mapping efficiency and quality. High-quality anomaly information can be further extracted from geochemical exploration data to effectively guide mineral prospecting in new areas.

Seismic data collected in the field frequently suffer from missing values due to constraints of acquisition cost or terrain. Data reconstruction is a critical step in seismic data preprocessing. Based on the compressed sensing theoretical framework, this study subsampled synthesized data using the 2D random undersampling technique. Then, the 3D seismic data were divided into a series of time slices. By introducing the sparse Shearlet transform and using the convex set projection (POCS) algorithm, this study conducted sequential data reconstruction for various time slices. As a result, a Shearlet transform-based time-domain 3D seismic data reconstruction method was developed. Numerical experiments and measured results demonstrate that the proposed reconstruction method exhibits a higher signal-to-noise ratio, a higher computational speed, and better effects than a Curvelet transform-based approach.

The exploration and exploitation practices in the Sichuan Basin in recent years indicate that breakthroughs have been achieved in the Jurassic continental tight sandstones.Nevertheless,due to the low porosity and permeability of tight sandstone,conventional post-stack inversion frequently exhibits limited resolution,failing to meet the accuracy requirements for the prediction of actual exploration reservoirs.This necessitates pre-stack inversion for detailed characterization of tight sandstones,while S-wave velocity is crucial to pre-stack inversion.Based on continental exploration wells drilled in the southeastern Sichuan Basin in recent years,this study developed a petrophysical modeling technique for dense sandstones in this region.Specifically,given the low permeability of tight sandstones and the uneven mixing of fluids in the pore space,the Domenico model was preferentially employed to calculate the pore fluid modulus.Although fluid modulus and density are inevitably variable under the actual subsurface conditions,previous studies typically use constant values to conduct petrophysical modeling for tight sandstones.In this study,depth-dependent values were applied.Tight sandstones in the southeastern Sichuan Basin generally exhibit a porosity of less than 10%.Therefore,calculations using the Nur and the Krief models will yield high errors.Given this,this study preferred using the Lee-Pride model to calculate the skeleton modulus and controlled the relationship between the rock matrix and the skeleton by introducing the value of the cementation parameter.The application of the established petrophysical model of tight sandstone to an actual survey area indicates high agreement with data from actual wells.Additionally,based on log statistics,Poisson's ratio,the most sensitive parameter is used for high-precision pre-stack inversion in the proposed technique,enabling detailed characterization and prediction of the internal structure of channel sandstones.

As an important part of gravity investigation, the performance tests of gravimeters are required throughout field surveys. In these tests, mean squared error (MSE), accuracy, and root mean squared error (RMSE) are commonly employedto quantitatively describe the test results. The comparison of the theoretical equations for dynamic and consistency tests in specifications on geological surveys, petroleum, and measurement reveals a pronounced confusion in the usage of MSE, accuracy, and RMSE. This issue is observed in the consistent equations forconsistency testsin these specifications. Through investigations into the equations used in the dynamic and consistency tests inthe threespecifications, this study analyzed the differences between mean MSE and RMSE, elucidated the normativity of RMSE relative to MSE, and determined the applicablerange of RMSE. To avoid confusion, it is recommended that accuracy be used for qualitative expression and RMSE for quantitative expression in these specifications.

Conventional splicing methods aim to serve structural interpretation and align the reflection energy and frequency of large strata based on seismic data themselves.They adopt seismic data only from one study area in processing data of overlap zones,failing to fully utilize the effective information in seismic data of all study areas.Consequently,the resulting low-fidelity seismic data in overlap zones severely affect subsequent discrimination of reservoir connectivity and hydrocarbon-bearing properties.This study proposed a weighted fusion-based data spicing method for overlap zones,where seismic data from different study areas are fused with corresponding weights to generate a trace of seismic data.The application of the proposed method to actual seismic data shows that the processing results based on the fusion and splicing of data in overlap zones can effectively improve the quality of seismic data,with high continuity of seismic events on the profile, contributing significantly to the discrimination of the connectivity of shallow sand bodies.

Understanding the distribution patterns of selenium in soils and crops is critical to developing selenium-rich industries. Enshi City in Hubei Province is known for its extensive selenium-rich soils, establishing this city as a promising area for selenium-rich agriculture. This study investigated Xintang Township in Enshi. Based on the organization and analysis of the geochemical data of 2 469 soil samples and 237 crop samples of maize, potatoes, rice, radish, cabbage, and tea, this study offered a systematic summary of the selenium distribution in soils and factors influencing selenium content in crops in the study area. The results indicate that the topsoils exhibit selenium content ranging from 0.14×10^-6 to 25.74×10^-6, with a background value of 0.81×10^-6, which is 3.7 times the national background of selenium content in soils. Selenium-rich soils cover 86.23% of the total area of the study area, and two NEE-directed selenium-rich belts are found. The spatial distribution of selenium in soils is closely related to soil-forming parent materials. Soils with Permian black rock series as parent materials exhibit notably higher selenium content, with an enrichment coefficient of 3.74. In high-selenium-content areas, rice, radish, and cabbage exhibit selenium enrichment rates exceeding 65%. Except for potatoes, crops display positive correlations between their selenium content and the selenium content in their root soils, with tea showing the highest correlation (P<0.01, R=0.84). This suggests a close relationship between the selenium content in crops and their root soils. The crops in cultivated areas with Permian black rock series and Triassic carbonate rock series as soil-forming parent materials exhibit high bioconcentration factors of selenium, with soils and crops with Permian black rock series as soil-forming parent materials presenting the highest average selenium content. This highlights the significant impacts of soil-forming parent materials on crop selenium content.

Small- and micro-scale faults fractures (fractures and faults with fault throw less than 10 m) that originally developed in shale strata have a significant impact on the probability of penetration, stimulation volume, and production capacity of high-quality reservoirs in horizontal well sections. Therefore, it is critical to conduct fine-scale fault and fracture prediction. However, any single method struggles to accurately identify and predict these faults and fractures. Based on the developmental conditions of small-and micro-scale faults and fractures in the shale gas reservoirs of the Longmaxi Formation in the southern Sichuan Basin, this study conducted forward modeling, response mechanism analysis, and characterization of fracture responses, developing a prediction method integrating predicting and modeling. Furthermore, this study preferentially investigated techniques including seismic data processing, small-scale fault and fracture prediction, multi-scale fracture modeling, and fusion characterization. The results of the proposed method were highly consistent with the geological anomalies including small and micro-scale faults, lost circulation, and inter-well pressure channeling observed during the drilling of horizontal wells in the shale gas reservoirs of the Longmaxi Formation. Furthermore, these results exhibit a strong positive correlation with the single-well production capacity. All these corroborate that it is feasible to use this method to predict small- and micro-scale faults and fractures. This study can serve as a reference for predicting small-scale faults and fractures in other strata of the same type.

Despite bearing valuable recyclable elements, mine tailings pose environmental risks. However, there is a lack of studies on the geochemical characteristics of tailings in China and abroad, hindering their appropriate treatment and reuse. This study collected tailing samples from the Han-Xing Iron deposit in Hebei Province, China and conducted the analysis and tests of these samples. This study determined the concentrations and chemical speciation (i.e., exchangeable, carbonate-bound, Fe-Mn oxide-bound, organic-bound, and residual forms) of elements including Fe, Co, S, Cu, and Zn in tailing particles with varying grain sizes. Accordingly, it explored the implications for the exploitation and utilization of tailing resources, along with the assessment of environmental risks. This study provides deeper insights into the geochemical characteristics of tailings, producing positive impacts on the exploitation and utilization of tailing resources, as well as the prevention of environmental risks.

To improve the inversion accuracy of reservoirs in the Paleogene strata with limited wells and sedimentary and structural complexity, two key technologies were used in seismic data processing: sparse pulse inversion for primary wave estimation and anisotropic Q-pre-stack depth migration (PSDM). This contributed to improved quality of seismic gathers and imaging. Then, the pre-stack simultaneous inversion method was applied as follows: (1) Stacking velocity and layer-constrained Dix inversion were employed to obtain a low-frequency model of P-wave impedance; (2) Elastic impedance inversion was performed using angle-stacked data and well-calibrated wavelets, yielding far, medium, and near elastic impedance; (3) Initial P- and S-wave impedance, as well as initial density, were obtained through Fatti inversion; (4) Pre-stack simultaneous inversion was performed to obtain the final P- and S-wave impedance and density; (5) Lithology and physical property inversion results were used to predict the reservoir distribution range. This method, driven by three-dimensional seismic data and exhibiting low dependence on logs, can serve as a reference for reservoir prediction under similar geological settings.

Slope tomography is a method to estimate subsurface velocity macromodels from the slopes and traveltimes of local coherent reflection events. In geologically complex areas, the macromodels obtained from slope tomography tend to yield larger errors. To address this issue, this study proposed a method for error correction of the models using deep learning. Specifically, with macromodels determined using slope tomography-based inversion serving as input and corresponding theoretical models as labels, a neural network was trained, yielding a nonlinear mapping from the slope tomography-derived macromodel to the corresponding theoretical model. To ensure that the trained neural network was applicable to measured seismic data, the training samples were generated from the inversion model and migration profiles of measured seismic data. Tests based on the data synthesized using the theoratical model validated the accuracy and effectiveness of the proposed method. The proposed method was then applied to the 2D measured seismic data from beaches and shallow seas, yielding velocity models with elevated precision and depth migration imaging profiles with high quality.

Land emerges as the most fundamental resource for agricultural production, and addressing land quality-related issues is the key to ensuring high food quality. A comprehensive understanding of land quality is crucial to the scientific and sustainable utilization and development of land resources. This study conducted a 1:50,000 geochemical survey of land quality in the contiguous farmland in Dongchuan Town, Yao'an County, Yunnan Province. Accordingly, this study provided a comprehensive assessment of soil nutrients, environment, dry and wet atmospheric deposits, and agricultural irrigation water. Based on the assessment results, this study determined the production areas of green food and pollution-free agricultural products, aiming to provide a reference for the development and utilization of sustainable, green cultivated land. The results indicate that soils in the study area are nutrient-rich and enjoy a favorable environment, high-quality irrigation water, and an excellent atmospheric environment. The comprehensive grade assessment results of soil quality align with those of land quality, revealing that the first and second-grade high-quality soils account for 68.98% of the total area. A batch of production areas of pollution-free and green food were determined, and those that can be directly used to develop green, zinc- and copper-rich crops were delineated. These findings will provide geochemical data support for land use planning, food security, building agricultural products with plateau characteristics, and serving rural revitalization efforts.

In the era of rapid digitalization development, artificial intelligence (AI) technology has brought revolutionary changes to traditional work patterns. Based on Baidu Comate, this study proposed an automatic numbering method for sampling points in irregular geochemical networks. Automatic numbering tests, conducted on 12 000 geochemical sampling points, demonstrate that the method improved the efficiency by 99.8% and achieved 100% accuracy compared to the traditional manual method. This indicates that the proposed method is more efficient and accurate than traditional approaches, effectively avoiding human errors and improving work efficiency. This study also discussed the challenges AI faces in processing complex instructions, the importance of instruction clarity, the identification of complex logic, and the necessity of developing knowledge reserves. Although AI technology has significantly improved the efficiency and accuracy of the automatic numbering of sampling points in irregular geochemical networks, the early development of packaging tools requires personnel who can read codes to modify and verify the codes. Additionally, AI-assisted demand processing should be in phases, and ultimately, it is necessary to encapsulate verified codes into a tool for reuse.

Based on high-precision, high-quality data on SiO₂, Al₂O₃, Fe₂O₃, and organic matter in topsoils obtained from the 1:250,000-scale multi-purpose regional geochemical survey in Fujian Province, this study assessed topsoils and, for the first time, established a geochemical method-based geochemical assessment system for desertified land in the province. Specifically, this system involves methods for calculating the silicon-aluminum-iron ratio (Saf), carbon-silicon ratio (K_Si), and their comprehensive index value (Szh_ f), which serves as the indicator for assessing the degree of land desertification. Specifically, a higher Szh_ f value indicates a higher degree of desertification, and vice versa. Furthermore, this study determined the statistics of geochemical classification parameters for desertified land in Fujian Province and delineated the distribution ranges of geochemical grades. This aims to assess the current status of desertified land in the province from the microscopic perspective of soil elements and to further ascertain the distribution characteristics of the land. The results indicate that the land with strong, moderate, slight, and very slight desertification in Fujian Province exhibits areas of 39 531 hectares (0.326%), 65,790 hectares (0.542%), 103 601 hectares (0.853%), and 360 329 hectares (2.968%), respectively, primarily distributed along the coastal zone to the south of the Minjiang River and in Changting County of Longyan City. Field verification demonstrates that this evaluation method is scientific and reliable and yields accurate classification results of desertification grades, thus objectively reflecting the distribution status of desertified land in Fujian Province.

In actual exploration, the surface-to-borehole transient electromagnetic (TEM) method is prone to be affected by irregular transmitter loops and inclined boreholes, resulting in increased measurement errors of the three-component induced magnetic fields and decreased interpretation accuracy. By establishing surface-to-borehole TEM method-based three-component measurement models under the excitement of transmitter loops of various shapes, this study derived the calculation formulas for surface-to-borehole TEM responses under inclined boreholes through coordinate transformation. Then, it achieved one-dimensional forward modeling of the full-space TEM field using a numerical filtering algorithm. The calculation results of multiple typical models indicate that the three components of the transient magnetic fields were significantly influenced by the shapes of the transmitter loops, with the impacts on horizontal components x and y far more significant than those on vertical component z. The transmitter loops of regular polygons with even edges as the sources exhibited uniform and symmetric distribution of the TEM fields of the three components. Under the condition of the same perimeter, a greater number of edges of the transmitter loops associated with greater primary field energy excited by the loops. Therefore, rectangular transmitter loops as the sources prove the most cost-effective. The inclination and offset primarily affected the amplitude of the three-component responses. In contrast, the borehole azimuth mainly influenced the sign of the horizontal components, bearing rich information on location. Therefore, in the exploration using the surface-to-borehole TEM method, it is necessary to accurately determine source paths and the geometric morphologies of boreholes to make essential corrections, thus improving the accuracy and reliability of interpretations.

The spectral parameters of the Cole-Cole model can improve the resolution of comprehensive interpretation of time-domain induced polarization (IP) data, contributing somewhat to the exploration of metal deposits. Applying the backpropagation neural network (BPNN) model to the prediction and inversion of spectral parameters can avoid high computational complexity to improve the inversion speed. Moreover, the BPNN model can fully explore the utilization efficiency of time-domain IP data to enrich the characteristic information of subsurface ore bodies. Based on this, this study derived the mathematical expression of the time-domain apparent polarizability attenuation curve using the digital filtering algorithm. With the mathematical expression as the forward/inverse model, this study comparatively analyzed the impacts of four factors-the sample size of the training set, the number of neurons in the input layer, the node number of hidden layers, and the number of hidden layers-on the training and inversion effects of the BPNN model, determining the optimal model. Furthermore, this study trained the BPNN model using time-domain IP data from eight time windows. Finally, this study applied the trained BPNN model for prediction and inversion based on the measured time-domain IP data. The results indicate that the BPNN model is feasible in inverting spectral parameters based on both theoretical and measured datasets, manifesting high inversion accuracy and minor errors. Overall, the results of this study can assist in distinguishing paragenetic and associated minerals and reducing misinterpretation.

To determine the optimal prospecting target in the Bishan area, Xianghuang Banner, Inner Mongolia, this study conducted a geochemical survey at 1∶10,000 scale. Using multivariable statistical analysis, this study analyzed 14 elements (i.e., Au, Ag, Cu, Pb, Zn, Mn, Mo, Li, W, Sb, Bi, B, Hg, and As) in soil samples from the study area. The results indicate that six elements, including Au, Ag, Pb, Zn, Li, and As, exhibited high enrichment degrees, suggesting their significant metallogenic potential. The anomalies of high-content elements are principally dictated by the outer contact zone between tuffs/tuffaceous volcanic breccias and granites in the Lower Permian Sanmianjing Formation. The ore-forming elements are primarily controlled by the NW-trending fault zone. Alterations like silicification, binarite, pyrite, galena, and sphalerite alterations serve as principal prospecting markers. Based on the geological conditions for mineralization in the Bishan area, this study delineated three composite anomalies and two prospecting targets, with Ag ore bodies discovered through engineering validation. Soil geochemistry proves crucial for the successful identification of polymetallic ore bodies, thereby providing a significant foundation for subsequent mineral exploration efforts.

Ray tracing-based first-arrival traveltime tomography is widely used to construct near-surface velocity models to achieve the static correction of seismic data from complex near surface.However,this method necessitates the calculation of ray paths for first-arrival traveltimes and the iterative updating of initial velocity models.As a result,significant computational time is required when applying this method to measured 3D high-density seismic data.To address this issue,this study proposed a method for quickly building 3D near-surface velocity models utilizing diving wave traveltimes under rugged surface.Specifically,based on the ray and traveltime equations of diving waves corresponding to velocities subjected to lateral and vertical changes under rugged surface,the velocity distribution from the observation surface downward was determined using common offset gathers.The proposed method eliminates the need for ray tracing and iterative updates of initial velocity models,offering high modeling efficiency.Tests based on data from theoretical models verified the effectiveness of the proposed method.When applied to measured 3D seismic data,the proposed method yielded static correction results comparable to those obtained using the Fresnel-volume first-arrival traveltime tomography while significantly improving computational efficiency.

The construction and presentation of a geological modelprove to be ahot topic and challenge in research on 3D geological modelling. Given the large scale, involvement of complex surfaces, and insufficient geological constraints of geological body data, this study achieved the rapid construction of a large-scale geological surface model using the domain decomposition-based implicit generation method. Initially, implicit functionswere constructed by taking radial basis functions as the kernel functions.Then, the distribution functions of various domainswere solved in parallel usinganoverlapping domain decompositionmethod, reducing the spatiotemporalcost and accelerating the solving process.Subsequently, normal vectors were extracted to generate control pointsand formconstraints on surface fluctuation, thereby effectively controlling the model boundaries. The experimental results indicate that the method proposed in this study can significantly improve the efficiency associated with the solving of distribution functionswhile ensuring the high quality of the model. This study effectively solves the problem of balance between efficiency and precision in geological modeling and provides methodological support for the refinement of geological surfaces.

Due to numerous thin interbeds in weakly elastic media,seismic excitation typically yields rapidly attenuated seismic wave energy and a narrow dominant frequency band,resulting in low-resolution seismic data.Therefore,selecting a favorable lithology plays a crucial role in improving the seismic excitation effect.This study explored the dominant factors influencing the quality of seismic data obtained from the northern Jiangsu exploration area,a region with a dense river system.Specifically,this study determined the top boundary of the high-velocity layer based on microlog surveys and the dominant lithologic member using the cone penetration test and lithologic coring.It quantitatively analyzed seismic wavelet attributes,including octave band,resolution,main-to-side lobe energy ratio,and wavelet clarity,establishing their matching relationship with the lithology for seismic excitation.By selecting a lithologic surface featuring a high seismic wave propagation velocity,a favorable elastic property,and a wide frequency band in the study area,it plotted a surface lithology map for pointwise well depth design,ensuring wide-frequency excitation.The above techniques were applied to well depth design for seismic excitation in the YA and SDX areas,achieving well-normalized single-shot frequencies and widening the dominant frequency band of the target layer in the seismic profile by over 10 Hz,with an increase of 1.5 octave bands.The results show that the excitation strategy of "selecting the dominant lithology from weakly elastic media" in regions with dense river systems can effectively enhance the seismic excitation effect in weakly elastic media,thereby improving the imaging accuracy and resolution of seismic data.

Noise detection and suppression of oil well pumping units pose challenges in data processing for mature exploration areas.The conventional method in the industry is to identify pumping unit noise through manual interactions and then suppress it as high-amplitude interference.However,manual identification wastes manpower and yields low detection accuracy,often resulting in missed detections.Hence,based on the noise characteristics of pumping units,this study conducted noise detection on seismic data containing pumping unit noise using deep learning methods.It then estimated the bandwidth of the detected noise using mathematical morphology techniques to determine the final position and distribution pattern of the noise.This allows for adaptive parameter support for the anomalous amplitude attenuation(AAA) method to achieve automatic detection and efficient suppression of pumping unit noise.The processing results of actual seismic data reveal that the methodology used in this study enables intelligent detection of pumping unit noise,significantly reducing the manual effort required for noise identification,improving the detection accuracy,and enhancing the fidelity and robustness of the AAA method.

Seismic waves, particularly their high-frequency components, will undergo energy attenuation during subsurface propagation. This results in a low signal-to-noise ratio (SNR) within the high-frequency band, significantly reducing the accuracy of seismic exploration for deep oil and gas. To enhance the SNR of deep seismic data, this study employed the time-frequency analysis technique to describe the time variations of signal frequency based on the non-stationary characteristics of seismic signals. Accordingly, this study proposed a method for suppressing deep high-frequency noise based on time-frequency analysis. Considering the stable and similar time-frequency characteristics of effective signals and the uncertain and random high-frequency interference, this study proposed an adaptive threshold selection strategy based on correlation analysis. This strategy involves extracting the time-frequency characteristics of effective signals from an advantageous frequency band and comparing them with the time-frequency characteristics within the high-frequency band. Subsequently, feature-constrained attenuation was applied to time-frequency spectra in the high-frequency band, thereby suppressing high-frequency noise. Theoretical models and actual data processing results demonstrate that the proposed method can effectively suppress high-frequency noise and significantly enhance the SNR of deep seismic data.

To understand the characteristics and sources of heavy metal contamination in the sediments of the Jinsha River basin, this study investigated the Qingling River basin-a primary tributary of Longchuan River on the south bank of the Jinsha River. Samples were collected from the sediments of 22 representative sections, and the concentrations of eight heavy metal elements, i.e., As, Cd, Cr, Cu, Hg, Ni, Pb, and Zn, were tested and analyzed. Statistical analysis was conducted on the distribution characteristics of these heavy metal elements in the sediments along the basin. The sources of these heavy metals were investigated using correlation analysis and principal component analysis, and the degree of heavy metal contamination in the sediments was assessed using the geo-accumulation index and the Nemero index. The results indicate that As, Cd, Cu, Hg, Ni, Pb, and Zn are primarily influenced by mining, agricultural, and industrial activities. Cr and Ni originate primarily from soil-forming parent materials. Besides natural sources, Ni is also affected by anthropogenic sources. The assessment results derived using the geo-accumulation and Nemero indices reveal that the eight heavy metal elements exhibit moderate or low contamination on average. However, partial elements, represented by Cd, Hg, Pb, and Zn, exhibit localized enrichment within the basin, primarily concentrated in the Laojiezi Au-Pb-Ag polymetallic mining area and the urban-rural junction in the lower reaches of the county.

The Jinchanshan gold deposit, identified as a magmatic-hydrothermal vein-type gold deposit associated with Yanshanian magmatism, is located in the Chifeng-Chaoyang gold ore concentration area within Harqin Banner, Inner Mongolia. The ore bodies in the deposit primarily occur in the fault structural zone of the Anjiayingzi pluton. This study conducted a systematic analysis of the structural superimposed halo anomalies of the Dawa and Limazigou ore sections from south to north in the eastern mineralized zone in the Jinchanshan gold deposit. Results indicate that the axial zoning and characteristic parameters of primary haloes in Jinchanshan gold deposit differ from the typical high-, medium-, and low-temperature element axial zoning sequence in hydrothermal deposits. The overlapping of high-temperature and low-temperature elements, suggests multi-stage and multi-phase mineralization processes. For the first time, this study analyzed the correlation between elements using element correlation curves at varying elevations. F and Ba were identified as front halo elements, Au, Cu, and Ag as near-ore halo elements, and Co, Ti, and V as tail halo elements. By establishing a structural superimposed halo model, this study determined prospecting target areas. Mining validation indicates that the identified ore-discovery middle section aligns with the predicted target area.

Traditional gridding methods struggle to balance computational accuracy and efficiency when processing irregularly distributed magnetic anomaly data. To address this issue, this study applied the classic least-squares collocation method from geodesy to the gridding of ground-based magnetic anomaly data. This application was verified through the test and analysis of the simulation data and the actual coalfield data. The results indicate that the computational accuracy of gridding based on least-squares collocation is dictated by the error estimation of discrete observational data and the selection and fitting of the covariance function. More accurate error estimation contributes to higher-accuracy interpolation. A polynomial function is a simple and effective empirical covariance function for processing magnetic anomaly data. The least-squares collocation method demonstrates more effective noise suppression compared to the Kriging, minimum curvature, and radial basis function methods. Overall, applying the least-squares collocation to the gridding of magnetic anomaly data can enhance the accuracy and efficiency of data processing.

To investigate the contamination, source, and ecological risk of soil heavy metals in a typical industrial town in Shandong Province, China, this study collected 499 topsoil samples from the study area from August to October 2022. The contents and spatial distributions of heavy metals like Hg, Cd, As, Pb, Cu, Cr, Zn, and Ni in the samples were analyzed using classical statistics and spatial interpolation methods. The source apportionment of heavy metals in the study area was explored through the principal component analysis (PCA). The contamination levels of heavy metals in the study area were assessed using the contamination index method. The results indicate that the average contents of Hg, Cd, As, and Pb in soils all exceeded their background values in Yantai City, and high-value zones were observed for all eight elements, indicating various degrees of enrichment. The analysis of coefficients of variation reveals that except for Ni, other heavy metals were significantly influenced by human activities. The PCA suggests that Cd, Pb, Cu, Zn, and Cr originated primarily from industrial and traffic sources. As and Hg were predominantly derived from industrial, agricultural, and domestic sources, while Ni was primarily from natural soil parent materials. The analyses based on the single-factor contamination index, geoaccumulation index, and Nemerow contamination index show that apart from Hg and Cd, other soil heavy metals in the study area exhibited no or slight contamination overall, demonstrating that the study area was principally contaminated by Hg and Cd. The potential ecological risk assessment suggests that the overall heavy metal contamination posed a minor risk level. A few sites with relatively severe contamination were primarily located around the industrial area. The waste gas, wastewater, and industrial residue generated by industrial activities constituted the dominant factor influencing the enrichment of heavy metals in surrounding soils. Overall, soil heavy metal contamination in the study area was at a moderate to low level, with some metals, particularly Hg and Cd, severely exceeding standard levels, warranting attention. It is recommended to strengthen the monitoring of heavy metals in soils around the industrial area, and adopt scientific and reasonable measures to ensure sustainable soil utilization.

The contamination of soils and groundwater caused by landfill leachate is increasingly prominent. Given the distinctive electrical properties of leachate, such as low resistivity, high polarization, high time constant, and low-frequency correlation coefficient, this study performed three-dimensional finite element numerical simulations of landfills using the spectral induced polarization (SIP) method. The aim is to explore the theoretical effects of this method in detecting landfill leachate. First, this study conducted forward modeling for typical geoelectric models using the algorithm developed in this study. The result revealed a maximum relative error between the numerical and analytical solutions of less than 2.5%, demonstrating the algorithm's accuracy and effectiveness. Second, the parameters of the SIP model were set based on the actual electrical characteristics of landfills, ensuring that the numerical simulations closely reflected actual conditions. Finally, SIP anomaly responses were compared and analyzed for landfill leachate leakage under different scenarios. The results indicate that the leachate locations can be accurately delineated using the apparent complex resistivity's phase and apparent dispersion, regardless of whether landfills are equipped with high-resistivity impermeable walls. For leachate plumes of the same scale, shallower locations corresponded to more pronounced anomaly morphologies of apparent complex resistivity's phase and apparent dispersion. Compared to the direct current resistivity method, the SIP method can detect richer underground electrical parameters, achieving better application effects in detecting landfill leachate.

The exposed crust is the critical interface where the lithosphere interacts with the pedosphere, hydrosphere, and atmosphere. The petrogeochemical map of the exposed crust can provide essential fundamental geochemical data for investigating the distributions and cycles of elements among different spheres. However, the plotting of the large-scale petrogeochemical map of the exposed crust has been constrained by the limited data volume and mapping technology. Consequently, no such a map covering the Chinese continent is available. This study proposed an innovative mapping technology roadmap based on over 16,000 petrogeochemical data and fundamental geological information. First, the databases for petrogeochemical information, stratigraphic structure information, and spatial information of geological units were constructed. Second, the geospatial information of strata and rock masses was extracted from the basic databases. Third, the petrogeochemical information was assigned to strata and rock masses to obtain their spatial distributions and element contents. Fourth, the spatial distribution patterns of 76 elements in the exposed crust were visualized using geographical information system (GIS) technology. Additionally, this study analyzed the challenges in the mapping process, including geological information accuracy, the lack of samples for special lithologies, reliability assessment, and scope of application, finally proposing corresponding solutions. The petrogeochemical map of the exposed crust demonstrates significant application potential, providing foundational data for investigating geochemical background and rock-sediment element cycling.

This study analyzed the distributions, enrichment characteristics, and health risks of heavy metals (Cd, Cr, Hg, As, and Pb) in soil-crop (maize, rice, and wheat) systems in the Shizuishan area, Ningxia Hui Autonomous Region. The results indicate that the contents of heavy metals in the soils were below national background values, with the geoaccumulation index indicating an overall clean soil environment. Different crops showed significant differences in their ability to enrich heavy metals, with Cd and Pb being more enriched in wheat, and As and Hg being more enriched in rice. Moreover, significant correlations were observed between soil pH and heavy metal enrichment coefficients. With an increase in pH value, maize exhibited markedly increased enrichment coefficients for Cr and As, whereas rice manifested a notably decreased enrichment coefficient for As. Soil organic matter inhibited the ability of crops to absorb and enrich heavy metals by reducing the bioavailability of heavy metals. The health risk assessment reveals a high non-carcinogenic risk index for children via the hand-to-mouth route. The As content in some samples exceeded the limit specified in food safety standards, requiring targeted monitoring. The results of this study provide a scientific basis for soil contamination control and safe crop production.

The Xinfang gold deposit resides in the gold polymetallic ore concentration area in eastern Liaoning Province. Based on regional metallogenic patterns and existing exploration results, the southern part of the Yalu River fault zone, where the Xinfang gold deposit is located, shows significant potential for gold resources. As exploration progresses, more gold ore bodies are expected to be discovered in the gold deposit area. However, the lack of effective deep exploration techniques limits further exploration and exploitation in the area, failing to effectively reveal the extensions of lithological boundaries and the distributions of fault structures in the area. This study presented the first application of the wide-field electromagnetic method in the area. Based on the obtained results and geological engineering information, this study delineated the electrical layers within a depth of 3 km, determining the thickness of the Neoproterozoic cap rocks and inferring several fault zones. Combined with existing geological data, this study identified the low-resistivity anomaly zone at a depth of about 1 km as an extensional structural plane caused by ductile shearing. Overall, this study provides valuable geophysical data for establishing a geological prospecting model for eastern Liaoning Province, holding significant research value for geological and mineral exploration in the area.

The Weihe Basin boasts abundant helium resources.Vigorously exploiting these resources is significant for national development and social progress.However,the complex geographical and geological conditions in the basin pose great challenges in seismic image processing and interpretation,severely hindering the identification of helium-rich natural gas traps and reservoir prediction.This study delved into the difficulties and challenges in processing the seismic data from the Huazhou-Huayin area.Based on the seismic exploration objectives,an effective seismic data processing workflow was developed,along with targeted technical solutions,to address the difficulties in seismic data processing.The results of this study demonstrate that the quality of seismic data processing can be effectively improved using techniques,such as pseudo-3D tomographic static correction,global optimization residual static correction,multi-domain and multi-method serial denoising,multiple amplitude compensation and wavelet correction,and prestack time migration.These improvements provide critical technical support for the exploration and production of helium-rich natural gas in the exploration area,and valuable insights for seismic data processing for other regions within the basin.

Offset and source weight constitute critical factors influencing the data acquisition of transient Rayleigh surface waves. Hence, it is necessary to examine the characteristics of dispersion spectra derived from surface wave data acquired under different offsets and source weights. This study investigated the influence of the two critical factors on dispersion spectra through theoretical analysis and numerical simulation. Based on field experiments, this study delved into the characteristics of dispersion spectra from surface wave data collected at various excitation locations of seismic sources with different weights. To address the discrepancies in dispersion spectra caused by varying source locations during the data acquisition of transient surface waves, this study proposed a superposition and fusion method for multi-dispersion spectra, significantly enhancing the effectiveness of surface wave detection results.

Rapid,convenient,and reliable acquisition of shallow urban underground structures in densely populated areas with intense anthropogenic noise is significant for promoting the digital transparency and safe development of urban underground spaces.With the advancement of nodal seismometers,passive-source seismic imaging methods have been widely applied to image underground structures at various scales,successfully demonstrating the detection of shallow underground structures in urban underground spaces.Under the constraints imposed by urban roads and narrow spaces,linear dense arrays show high adaptability among various passive-source array deployment patterns.In a test area with known underground pipeline anomalies,this study designed three linear array arrangement patterns with spacings of 1 m,3 m,and 5 m for 1 h continuous observation of noise data.This study employed the extended spatial autocorrelation(ESPAC) method to extract surface-wave frequency dispersion data for shear-wave velocity inversion.Moreover,by comprehensively analyzing the raw data,frequency dispersion curves,and the shear-wave velocity profile obtained through inversion,this study provided a scientific understanding and basis for the parameter selection of the closely spaced linear dense array observation system for passive-source seismic detection of urban underground spaces.Finally,based on the experimental results,this study selected a scientifically reasonable observation system for detection in a real-world urban underground space construction and exploration project,revealing the complete stratigraphic structure and water-conducting structures like fracture zones at the construction area.Therefore,closely spaced linear dense arrays can yield higher resolution and accuracy in detecting urban underground spaces,showing higher adaptability in areas with severe anthropogenic interference.

The Laowan gold belt is located in the eastern part of the Qinling orogenic belt, between the Nanyang and Wucheng basins. It boasts super-large gold resources hosted by moderate- to low-temperature magmatic-hydrothermal gold deposits formed during the Late Yanshanian. To explore the second exploration space and achieve new prospecting breakthroughs, this study investigated the primary halo of the No.59 ore body in the Laowan gold belt. Multiple statistical methods were employed to analyze the correlations among ore-forming elements. Accordingly, reliable geochemical indicators were identified to predict deep prospecting orientations. The results show that the primary halo of the No. 59 ore body exhibits an axial zoning sequence of Sn, Ba, As, Au, Pb, Cu, Sb, Ag, Mo, Bi, W, Hg, and Zn. The front-halo element Hg appears late in sequence, positioned after ore-forming elements Au, Ag, Pb, and Cu, suggesting the presence of blind ore bodies at depth or significant extension of the known ones to a burial depth of 650 m and above. In contrast, the rear-halo element Sn appears early in the sequence, indicating that the ore body was somewhat denuded. Correlation, cluster, and factor analyses were used to investigate the affinities between ore-forming elements and other metallic elements. Strong correlations were observed among Cu, Ag, and Au, establishing them as effective prospecting indicators within the deposit area. The cumulative index ratio of front- to rear-halo elements showed an increasing trend at a burial depth of 650 m, where a well-developed and non-closed anomaly was revealed by the F2 factor score contour. This suggests promising prospecting potential at this depth and deeper parts. Based on the comprehensive analysis, this study posits that the area at and below the burial depth of 650 m in borehole ZK3041 along the exploration line 304 is an exploration target. A prediction borehole drilled north of borehole ZK3041 saw the extension of No.59 ore body at a burial depth of about 720 m. Moreover, the assessment indicators show that the ore body still holds significant prospecting potential in the deeper part.

The Audio-frequency Magnetotelluric (AMT) sounding method is an electrical exploration technique widely used for shallow to medium-depth geological investigations. Its data quality is influenced by various factors, including cultural electromagnetic interference, electrode polarization, and grounding resistance. This study focuses on the impact of electrode polarization and grounding resistance on AMT data quality. First, the mechanisms of electrode polarization and the influence of grounding resistance on apparent resistivity and impedance phase are theoretically analyzed. Then, based on measured data, the effects of electrode polarization and grounding resistance on apparent resistivity and phase curves are investigated. Finally, a comparative analysis of 2D inversion results from field data is conducted to reveal how grounding resistance affects geological interpretation. The results indicate that electrode polarization significantly degrades data quality in the low-frequency band, though this can be mitigated through data processing techniques, while its effect on high-frequency data is negligible. Unstable grounding resistance mainly influences the high-frequency range, causing notable reductions in apparent resistivity and phase, which gradually stabilize toward the low-frequency band. Moreover, unstable grounding resistance may lead to inversion results that deviate from the true geoelectrical structure, thereby resulting in erroneous geological interpretations. The findings of this study provide important guidance for field data acquisition and processing.

To address the challenges in predicting superimposed thin sandstones in lacustrine basins,this study proposed a technical workflow for their prediction using the seismic meme inversion(SMI) method.First,an initial inversion model was constructed using log curves from sample wells selected based on seismic waveform similarity.Second,the initial inversion model was iteratively optimized in a Bayesian framework to yield high-resolution SMI results.Third,the SMI results were integrated with low-frequency inversion results to yield high vertical resolution while effectively characterizing the lateral superimposed patterns and boundaries of sand bodies.Case studies demonstrate that the SMI method achieved a prediction accuracy of 2 m to 3 m for superimposed thin sandstones.The coincidence rates for calibration and validation wells reached 91.5 % and 85.2 %,respectively,confirming the effectiveness of the SMI method and the high precision of the inversion results.Overall,this study provides an effective technical approach to predicting superimposed thin sandstones in lacustrine basins.

Seismic velocity model accuracy significantly influences the quality of depth-migrated images.However,a quantitative analysis and clear understanding of this relationship remain lacking.Based on the seismic images of ultradeep fault-controlled fractured-vuggy reservoirs in the Tarim Basin,this study constructed a depth-domain numerical velocity model that approximates the conditions of subsurface media.The 3D seismic forward modeling was performed using a seismic acquisition and observation system for practical production.Different interval velocities within the numerical model were smoothed and used to quantitatively analyze their effects on reverse time migration(RTM) images based on identical parameters.Using seismic simulation data as input,the isotropic and anisotropic depth-domain velocity models were constructed following a standard industrial workflow.A quantitative analysis was subsequently conducted to analyze the differences of the constructed models and corresponding migration images from the numerical velocity model and its RTM image across different intervals and for various geological features.Overall,this study quantitatively analyzes the velocity accuracy of different velocity types in various intervals and its influence on migration imaging, obtaining several clear insights.

Accurately and quickly determining the geological properties related to magnetic anomaly sources is a key technical challenge in magnetic prospecting, directly influencing the accuracy of geological interpretation using magnetic survey data. The induced magnetization varies with time, whereas the remanent magnetization typically remains constant over time. Therefore, there exists a theoretical basis for investigating the properties of magnetic anomaly source bodies by detecting the time variations of magnetic anomaly intensity. Despite geophysicists' relevant research in this field, practical technical methods have not been established. Hence, this study proposed a novel method for determining magnetite ore-induced magnetic anomalies. Based on the observational data of geomagnetic diurnal variations, the proposed method constructed parameters, including the variations (A), variation rate (η), and normalized variations (F) of magnetic anomaly intensity, to evaluate the properties of strong magnetic anomaly source bodies. Accordingly, the proposed method determined the possibility of magnetite ore-induced magnetic anomalies, showing critical significance for magnetite exploration.

To address the key issues in marine geomagnetic field modeling, this paper systematically explored the theoretical basis, modeling method, and optimization strategy of the equivalent source technology. By analyzing the geometric parameters and spatial configuration strategies of equivalent sources, a terrain-following vertical hexahedral equivalent source configuration scheme was proposed, significantly enhancing the accuracy of magnetic field models. In terms of algorithm implementation, a sliding window-based coverage calculation scheme was employed, effectively overcoming the bottleneck in the high-precision processing of massive magnetic survey data. Experimental results show that maintaining an overlap rate of 15%~20% in the sliding window ensures both boundary continuity and optimal computational performance. This method provides a reliable technical support for high-precision marine geomagnetic field modeling, with its effectiveness having been verified across various geological models (with the errors less than 5%).

Reflection waveform inversion(RWI) provides an effective initial model for full waveform inversion(FWI) by alternately updating the low-to-intermediate and high wavenumber components in the model.However,time-domain RWI requires storing the cross-correlation between the source background wavefield and the receiver scattered wavefield over the entire time series to compute gradients,demanding substantial computational storage.Although frequency-domain RWI exhibits multi-scale properties,it also imposes high demands on computational memory.Based on the discrete Fourier transform,this study proposed a RWI method that extracts frequency-domain wavefields from corresponding time-domain wavefields for multi-scale inversion in the time-frequency domain.The proposed method requires storing only a few single-frequency wavefield snapshots,showing significantly lower storage demands compared to conventional time-domain RWI and reduced computational costs relative to frequency-domain RWI.Therefore,the proposed method effectively combines the computational efficiency of time-domain RWI with the multi-scale properties of frequency-domain RWI.Considering the frequently missing low-frequency data in actual seismic data,this study further integrated envelope-based RWI to reconstruct low-frequency information for medium-deep subsurface structures.This enables stage-wise optimization of the inversion process,reducing the dependency on low-frequency long-offset data.Finally,numerical examples validate the effectiveness of the time-frequency domain RWI method based on envelope data in recovering low-frequency information.

The Dafengding tunnel is an extra-long deep tunnel in the seismically active mountainous region characterized by exceptionally complex geologic and hydrogeologic conditions in Southwest China. The primary challenges in the tunnel investigation arise from complex geological conditions and unfavorable exploration environments. Moreover, developing effective seismic mitigation measures necessitates precisely determining the fragmentation and water-bearing characteristics of rock masses at the tunnel site. Hence, based on the initial classification of regional lithologies using geologic data, this study conducted integrated geophysical exploration at the tunnel site by combining the audio-frequency magnetotelluric method with high-resolution electrical resistivity tomography. The results indicate that the integrated geophysical exploration effectively revealed the overall variation in apparent resistivity at the tunnel site and the shallow apparent resistivity distribution, particularly the characteristics of the electric field near the tunnel portal. These findings enabled the inference of the fragmentation and water-bearing characteristics of rock masses at the tunnel site. Moreover, the borehole data effectively validated the effectiveness of the integrated geophysical exploration, demonstrating that the integrated geophysical exploration can provide a valuable reference for the investigation and construction of extra-long deep tunnels in seismically active regions.