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.

In recent years, the Yifeng-Fengxin area of Jiangxi Province has witnessed significant breakthroughs in the exploration of granite-hosted lithium deposits, making it one of China's significant lithium ore clusters. However, there have been few reports on the application and achievements of lithium-related techniques. Focusing on 2,343 stream sediment samples from a 1∶50,000 geochemical survey in the Yifeng-Fengxin area, this study statistically analyzed geochemical parameters of over 20 elements. Subsequently, beryllium (Be), lithium (Li), niobium (Nb), rubidium (Rb), and tin (Sn) were determined as elements related to the mineralization and association of lithium deposit. Among these, Li and Sn showed strong enrichment and significant differentiation. Using statistical methods such as cluster and factor analyses, the five elements were further grouped into three assemblages: Rb-Be, Li-Sn, and Nb. Besides, this study processed and analyzed the stream sediment data from the Yifeng-Fengxin area using the iterative method, determining the lower threshold of both geochemical anomalies. Then, single-element and multi-element geochemical anomalies were investigated, with lithium identified as the primary ore-forming element. These anomalies, featuring prominence and large scale, were primarily situated in the northern and northeastern parts of the survey area, with concentration centers located in muscovite granite outcrops. The Li-Sn assemblage, serving as an indicator for granite-hosted lithium deposits, coincides well with the distribution of ore bodies. This confirms that Li-Sn assemblage anomalies in stream sediments can effectively delineate granite-hosted lithium deposits. Based on the anomaly characteristics from the 1∶50,000 stream sediment survey, two prospect areas were delineated in the periphery, providing guidance for further lithium exploration.

The distribution of fractures influences the trajectory design of shale gas horizontal wells and the stimulation effectiveness of hydraulic fractures. Faults and joints serve as two types of fractures. Accordingly, this study proposed the seismic attribute-geomechanics integrated fracture modeling technology. Specifically, fault modeling is conducted using seismic attributes, and joint prediction is performed using geomechanical structural restoration and Mohr-Coulomb theory. The obtained results of faults and joints are then integrated into the discrete fracture network (DFN) modeling for spatial characterization of natural tectonic fractures. The proposed technology was applied to the exploration of Jurassic lacustrine shales in the Fuling area within the Sichuan Basin. Its fracture modeling results were consistent with the imaging log interpretation results, confirming the development zones and spatial distribution patterns of fractures. Moreover, the stimulation effectiveness of fracturing was evaluated by comparing the fracture modeling results of typical wells with the actual fracturing performance. The evaluation results indicate that tensile fractures that are vertical or oblique to well trajectories are favorable for fracturing. Overall, the results of this study hold positive implications for predicting fracture development zones and guiding horizontal well trajectory design in the exploration stage, serving as a reference for subsequent exploration deployment.

The Songshunangou gold mining area in Qinghai Province is situated in the central part of the North Qilian metallogenic/orogenic belt. Two deposits have been identified in its eastern and western portions, with cumulative proven Au resources exceeding 34 t, establishing them as large-scale deposits. Moreover, the area holds significant potential for ore prospecting, necessitating an urgent need for research into prospecting orientations to support mineral exploration efforts. Hence, based on the anomaly analysis of 1∶10 000 soil geochemical survey data and employing the two-window moving average method, this study identified the anomaly distributions and enrichment patterns of 10 elements (i.e., Au, As, Sb, Hg, Cu, Pb, Zn, Ag, W, and Mo) in the area. By analyzing elemental geochemical anomaly data and integrating the geological conditions for mineralization, this study delineated composite anomaly zones in the area using the average contrast method, followed by anomaly assessments. The results indicate that Au is the primary ore-forming element in the area, characterized by strong enrichment and significant variability, suggesting high mineralization potential. Seven composite soil geochemical anomaly zones were delineated in the area, with zones HP-1, HP-2, and HP-5 demonstrating the greatest potential for mineral exploration. Engineering validation confirms the presence of two ore/mineralized bodies, with one exhibiting an ore thickness of 1.5 m and a gold grade of 0.57 g/t in zone HP-2 and the other featuring an ore thickness of 1.42 m and a gold grade of 1.67 g/t in zone HP-5. Overall, this study provides geochemical evidence for determining prospecting orientations in the Songshunangou gold mining area while offering a reference for the deployment of subsequent mineral exploration work in the area.

The Xiongcun ore concentration area in Xietongmen County, Tibet, is situated in the central segment of the Gangdise metallogenic belt. This study aims to investigate the distribution characteristics and geochemical anomalies of ore-forming elements in the study area and its periphery. This will guide the deployment of mineral exploration work in the study area and its periphery and promote the construction of the Xiongcun large-scale copper-gold resource base. Through the assessment of 1∶50000 stream-sediment geochemical anomalies in the study area and its periphery, this study determined the geochemical anomaly distributions and enrichment patterns of five elements (i.e., Cu, Au, Pb, Zn, and Ag) in the study area. Based on the analytical results of element anomalies and the geological characteristics of the study area, this study delineated the composite anomaly zones in the study area. Furthermore, this study assessed the prospecting potential of the study area through follow-up geochemical surveys. The results indicate that Cu and Au serve as the principal ore-forming elements in the study area. Both elements are characterized by strong enrichment and strong variability, showing high mineralization and prospecting potential. Four composite geochemical anomaly zones of stream sediments were identified. Among them, zones HS-1 and HS-2 exhibit highly consistent composite anomalies. Both zones show the distribution of known ore deposits (occurrences) or significant mineralization shows, suggesting considerable potential for ore prospecting. Overall, this study provides geochemical evidence for geological prospecting in the Xiongcun ore concentration area and its periphery while also offering ideas and a reference for subsequent mineral exploration targets in the study area.

Differing from onshore oilfields,deep-water oilfields center their exploitation on economic efficiency, employing a strategy of achieving higher production via fewer wells while maintaining formation energy balance.Their exploitation plans are flexible and emphasize dynamic oilfield monitoring,allowing for adjustment and optimization during the exploitation,thereby achieving fast and efficient exploitation.Therefore,the placement of infill wells serves as a crucial step for stable production of deep-water oilfields in the middle and late exploitation stages.Considering oil reservoir characteristics and production well waterflooding,this study investigated the deep-water turbidite reservoirs in the Bata oilfield,West Africa.A rapid infill-well optimization plan was proposed based on a tectono-sedimentary study of the oil reservoirs.The proposed plan centers on predicting high-quality turbidite sand bodies using the prestack amplitude versus offset(AVO) attributes,identifying waterflooding fronts through time-lapse seismic surveys,performing a fine-scale description of turbidite sandstone reservoirs and predicting the distributions of residual oil in the reservoirs.The implementation of the proposed plan demonstrated satisfactory production performance.Specifically,the infill wells achieved daily crude oil production of 12,000 barrels,establishing them as a primary contributor to oilfield production.This result validates the effectiveness of rapid infill-well optimization technology.Overall,this study provides a significant reference for enhancing the oil recovery of deep-water reservoirs through the placement of infill wells for deep-water oilfields in the middle and late exploitation stages.

During the reorganization of historical magnetic survey data for the South China Sea, it was found that the processed magnetic survey data of a specific area in the southern South China Sea exhibited low accuracy, affecting the effectiveness of data use. Data analysis reveals that the low accuracy was primarily caused by the significant errors in diurnal correction. Therefore, it is necessary to modify the diurnal correction method and reprocess the data. Based on the recollected data from various available geomagnetic observation stations and the variations of the geomagnetic field, this study simulated the magnetic diurnal variations in the study area during the survey period using three algorithms: the CM4 model, regression analysis, and regression analysis based on the data sample control of the CM4 model. The historical magnetic survey data were recorrected for diurnal variations, with data affected by magnetic disturbances removed based on geomagnetic indices, significantly enhancing data accuracy. This study addresses the problem of low accuracy in historical magnetic survey data in the study area, providing a novel approach for the fine-scale reprocessing of magnetic survey data related to the South China Sea.

Wavefield separation serves as a key step in processing the data of vertical seismic profiles (VSPs). Its accuracy directly influences seismic imaging, inversion of elastic parameters, lithology identification, and interpretation of hydrocarbon-bearing properties. Traditional methods face challenges in wavefield separation. For example, the median filtering requires manual intervention, often introducing errors and thus compromising separation accuracy; the FK filtering yields high accuracy but low efficiency. In contrast, deep learning techniques offer high automation, enabling both high accuracy and efficiency in wavefield separation. Hence, this study proposed a deep learning-based method for separating up- and down-going waves in zero-offset VSPs. First, the up- and down-going waves were separated through FK transform, generating a dataset. Second, a deep learning-based model, Unet++, was constructed for separating these waves in VSPs. Third, the relative down-going wavefield (obtained by subtracting the predicted up-going wavefield from the full wavefield) was incorporated into the loss function to mitigate the impacts of amplitude differences between up- and down-going waves on network updates. Moreover, the structural similarity index measure (SSIM) was employed as a regularization constraint to assist the network in learning the structural characteristics of the wavefield. The test results of actual VSP data demonstrate that the trained network can effectively learn the characteristics of the up- and down-going waves, achieving high accuracy and efficiency in wavefield separation.

The construction of three-dimensional visualization models in key urban areas is important for multidisciplinary data integration and intuitive presentation of spatial stratigraphic distribution.However,existing models face limitations in integrated geological-geophysical interpretation and concrete visualization of results.In response to this,this study established a 3D visualization model based on comprehensive geophysical data,specifically including a 3D structural model and a 3D velocity model for the Xiong'an core area down to 5 km depth.It revealed the correspondence between the 3D S-wave velocity distribution and the spatial morphology of strata,and validated the structural framework of alternating uplifts and depressions.To be specific,within the depth range of 0~1 km,the lateral homogeneity of S-wave velocity reflects the stable sedimentary characteristics of Quaternary and Neogene strata.Between 1 km and 2.2 km,the velocity zoning is unclear,primarily due to fracture development in the Jixianian geothermal reservoir,which leads to a reduction in S-wave velocity.Below 2.2 km,the significant velocity contrast between the Xushui Depression and the Rongcheng Uplift is mainly attributed to lithological differences and unconformable contacts.The construction of the 3D velocity model helps overcome the limitations of sparse geological data and reveals the subsurface 3D structures in the Xiong'an core area from a physical property perspective.

Stream sediment survey is the most widely applied method in regional geological surveys due to its simplicity, efficiency, low cost, and proven effectiveness in mineral exploration. This method shows promising application potential in Southwest China, characterized by well-developed hydrographic nets. Sampling quality directly determines the representativeness and accuracy of geochemical exploration data. However, sampling point arrangements in alpine gorge areas remain challenged by insufficient coverage of lower-order streams, omission of coarse-grained clastics in high-energy zones, and interference from human-induced contamination. To address these challenges, this study innovatively proposed an optimization strategy combining synergistic optimization and the on-site fine-tuning method for the 1∶50 000 stream sediment survey in the Fanshen Village area, Huize County, Yunnan Province. This strategy integrates critical technologies, including two-level dynamic grids (a 1 km×1 km basic grid and a 500 m infill grid), dynamic channel alignment offset (50 m to 100 m), and pre-set contamination buffer zones (200 m), for fieldwork. The results indicate that compared to traditional fixed-grid methods, the optimization strategy achieved a significantly increased coverage rate of 72% for tertiary tributaries, a capture rate of 82 % for coarse-grained clastics (>2 mm), and a reduced occurrence rate of 5% for human-induced pseudo-anomalies, with the overall cost increase controlled within 12%. Overall, the optimization strategy can effectively enhance the reliability of sampling data and the accuracy of anomaly delineation in complex topographic areas, providing an optimized solution for geochemical surveys in alpine gorge areas, Southwest China.

Aerial gamma spectroscopy measurement has important application value in mineral geological exploration, environmental radiation monitoring, and nuclear emergency response due to its advantages of high efficiency, flexibility, and avoidance of personnel radiation exposure risks. With the rapid development of drone technology, drones equipped with gamma-ray spectrometers have become a more flexible and cost-effective low altitude measurement method. However, drones typically fly at low altitudes below 40 meters, and complex terrain can significantly affect the solid angles of detection and the attenuation of gamma rays in the air, thereby reducing the accuracy of measurement results. This article proposes a terrain correction method for unmanned aerial vehicle (UAV) gamma spectroscopy measurement based on digital elevation model (DEM) data, targeting typical complex terrains such as mining pits, stepped mining faces, ore piles, and waste rock piles in open-pit rare earth mines. By establishing a micro element detection factor model and combining it with finite element discretization algorithm, quantitative correction of terrain undulations in the detection area can be achieved. The Monte Carlo simulation and field measurement results show that this method can effectively control the gamma ray intensity response error of terrain such as ridges, valleys, gentle slopes, and slopes within 10%, significantly improving the data quality of low altitude drone gamma spectrum measurement. The unmanned aerial vehicle (UAV) airborne gamma spectroscopy measurement in the open-pit mining area of rare earth mines shows that the relative error between the element content measured by UAV airborne gamma spectroscopy after terrain correction in the measurement area and the weighted average element content measured by surface gamma spectroscopy within 90% correction range is within 30% of the number of points, and the uranium content has increased from 53.2% without terrain correction to 74.3%; The thorium content has increased from 80.3% without terrain correction to 93.3%; The potassium content has increased from 94.7% without terrain correction to 97.2%. The terrain correction method has been verified to have strong practicality and reliability.

Accurate estimation of the quality factor(Q) is essential for enhancing seismic data resolution and reservoir characterization.Conventional Q estimation methods generally utilize post-stack data, which neglect the impacts of raypaths.Moreover,the average effect of stacking alters the attenuation of seismic data,reducing the accuracy of Q estimation.Compared to post-stack data,the pre-stack data more faithfully preserve the attenuation properties of subsurface media,enabling more accurate Q estimation.Therefore,this study converted pre-stack data into the apparent velocity and travel time(R-T) domain,using the radial trace(RT) transform.Combined with the logarithmic spectral area double difference(LSADD) method,a pre-stack Q estimation method named QVAV_LSADD was proposed.This method accounted for the impacts of raypaths under imprecise interval velocities.Its high accuracy and strong noise resistance were validated through the processing of both synthetic and real data.

Conventional migration imaging,which only considers primary reflected waves,fails to effectively image subsurface high-steep geobodies.Compared to primary reflected waves,prismatic waves travel an additional path, enabling the imaging of high-steep structures.However,the additional travel path increases the computational load.Seismic interferometry can shift the surface observation system downward to an artificially selected subsurface calibration plane.Consequently,subsequent calculations only need to be performed on the model below the subsurface calibration plane,there by improving computational efficiency.Hence,this study proposed a novel imaging method integrating prismatic waves and seismic interferometry.In the proposed method,the acoustic wave equation was replaced by the prismatic wave equation.Virtual records were generated through cross-correlations between data from the surface and subsurface reference planes.These records were combined with reverse time migration(RTM) for imaging.The proposed method was verified using the horizontal layered, L-shaped, and salt dome models.Specifically,the horizontal layered model exhibited consistent imaging results with those below the global reference plane;the L-shaped model outperformed conventional methods in imaging steep structures;the salt dome model displayed enhanced pre-salt imaging resolution.Due to the downward shift of the surface observation system,the following imaging process entailed a reduced vertical depth and a shorter shot record time.The proposed method reduced the computational cost to 26.6% of that using the conventional RTM imaging.Overall,through the downward shift of the observation system and the utilization of multiples,the proposed method achieved both satisfactory accuracy and efficiency,overcoming the limitations of traditional interferometry and providing a novel solution for exploring high-steep structures.Notably,there still exist discrepancies between the simplified models and the actual heterogeneity,requiring further optimization of the computational cost.

Manganese ores are significant raw materials in the iron and steel industry. Accurately determining their major components is critical for the analysis of manganese ore composition. Based on three acid dissolution systems, i.e., aqua regia (HNO3+HC), tetracid (HNO₃+HF+HClO₄+HCl), and pentaacid (HNO₃+HF+H₂SO₄+HClO₄+HCl) solutions, and three extraction conditions, i.e., hydrochloric acid, nitric acid, and aqua regia, this study determined five oxides in manganese ores using inductively coupled plasma-optical emission spectroscopy (ICP-OES). The results indicate that the pentaacid or pentaacid solution and the extraction with hydrochloric acid achieved encouraging determination results. Spectral lines with wavelengths of 766.490 nm (K), 184.006 nm (Ca), 589.592 nm (Na), 279.553 nm (Mg), and 396.152 nm (Al) were analyzed. The possible interference in the determination process was eliminated based on the standard solution matrix matching principle. The detection limits of Na, Mg, K, Ca, and Al were 0.000 27%, 0.000 21%, 0.000 15%, 0.000 17%, and 0.000 23%, respectively. The determination results of all five oxides in two national primary reference materials for manganese ores showed relative standard deviations (RSD) not exceeding 5.0%, suggesting fair precision. The verification results of all five oxides in five reference materials showed relative errors (RE) below 10%, demonstrating high accuracy, with the measured values roughly consistent with the recommended values. The spiked determination of five oxides in five manganese ores with unknown content yielded recovery rates ranging from 90% to 110%, suggesting that the accuracy met the analytical requirements. Therefore, the ICP-OES method simplifies sample pretreatment, improves efficiency, and reduces costs, thereby applying to batch sample analysis. The verification using reference materials demonstrates that its accuracy and precision meet industrial standards, establishing the ICP-OES method as an effective approach for determining the five oxides in manganese ore samples.

Overpressured sedimentary basins are ubiquitous and extensively distributed in marine areas.They typically possess favorable conditions for hydrocarbon accumulation,thus holding considerable significance for guiding hydrocarbon resource evaluation and exploration deployment.Velocity serves as a key parameter for research on overpressure distributions.The increasing demands for higher accuracy in overpressure distribution prediction impose higher requirements on the accuracy and quality of seismic velocities.Compared to well patterns in land areas,the more sparse well patterns in marine areas lead to insufficient well constraints on seismic velocities,requiring a comprehensive reliability evaluation.Additionally,seismic velocities obtained from conventional velocity analysis or modeling often show insufficient resolution.Therefore,this study proposed a high-density bispectral velocity picking method under the theoretical framework of high-order normal-moveout(NMO) based on uncorrelated parameters.The proposed method can obtain higher-resolution seismic velocity volumes for favorable target areas,providing an effective approach to exploring overpressure distributions in marine basins.This study first presented the theoretical principles of high-order NMO correction.Subsequently,it demonstrated the application of the proposed method in a specific exploration area in the East China Sea.The practical application suggests that seismic velocity volumes subjected to high-order NMO correction can effectively reflect subsurface overpressure distributions,showing promising prospects in the research on overpressure distributions in marine areas.

With the gradual depletion of shallow mineral resources, deep mineral exploration has emerged as an essential development trend in the mining industry. Increasing the distance between receiver electrodes of induced polarization (IP) devices is the most direct and effective approach to enhance the exploration depth. However, a long distance can cause strong electromagnetic (EM) coupling effects, severely interfering with IP signals. Addressing this challenge, this study calculated the EM coupling effects of various measuring devices in homogeneous half-space and layered media using analytical methods. Furthermore, this study comparatively analyzed the impacts of various factors, including measuring device type, wiring layout, distance between receiver electrodes, earth resistivity, and frequency, on the EM coupling intensity. Based on the phase differences between IP and EM coupling effects in the frequency domain, this study derived the calculation equation of the relative phase spectrum, followed by a theoretical analysis of the decoupling effects in application scenarios. The results indicate that increasing distance between receiver electrodes, decreasing earth resistivity, and raising working frequency all significantly intensified the EM coupling interference. Under consistent conditions and detection depths, the Schlumberger array suffered from higher EM coupling interference compared to the pole-dipole array. Compared to the traditional IP phase spectrum, the relative phase spectrum enhanced the maximal working frequency of the pole-dipole and Schlumberger arrays by four and 10.6 times, respectively, suggesting the decoupling capability of the relative phase method in large-depth IP exploration. Overall, this study provides significant guidance for the field implementation of large-depth IP exploration.

Artificial intelligence algorithms have been developed to automatically identify the spatial structures of formation lithologies from multivariate log data. They represent a promising approach to reducing lithology logging costs and mitigating the subjectivity inherent in lithology identification. Considering the imbalanced distribution of lithology sample data and the spatialtemporal variability in the relationships between log attributes and lithologies, this study constructed a synthetic minority oversampling technique (SMOTE)-long short-term memory (LSTM) hybrid model. The SMOTE algorithm effectively balances the sample distributions of different lithologies, while the LSTM algorithm, using its deep learning architecture, extracts lithological characteristics from the log sequence data. With the borehole log data and lithology records from a sandstone uranium deposit as training data, the SMOTE-LSTM hybrid model achieved a prediction accuracy exceeding 85% in lithology classification. Compared to several other machine learning methods, the SMOTE-LSTM hybrid model demonstrated significantly improved accuracy and reliability in lithology identification.

Due to the dynamic and multidimensional properties of mineral resource data, the information on the utilization status of mineral resources fails to accurately reflect the actual utilization status of mineral resources, leading to inaccurate predictions of mineral resources. Therefore, this study proposed a reserve prediction method based on the dynamic information on the utilization status of mineral resources. The exploration data of mineral resources, subjected to format conversion and encoding, were input into the geographic information system (GIS). They were categorized by the GIS into spatial and attribute data, thereby establishing a resource utilization status management system. A fully relational database was introduced to manage the utilization status of mineral resources. A 3D geological model was employed to calculate the ore-forming favorability and delineate the mineralization target area. Based on this, a grade-tonnage model was constructed. Finally, the statistical sampling theory was applied to predict the resource reserves in the mining area. The experimental results demonstrate that the mineral reserves predicted using the proposed method aligned with the actual reserves, indicating a relatively high prediction accuracy.

Based on the geochemical data of farmland soils in Jiangsu Province, obtained from regional eco-geochemical surveys and the geochemical assessment of land quality in past years, this study delved into the geochemical characteristics of Zn in soils, aiming to explore the prospects of producing natural zinc-rich food in Zn-rich soils. The results indicate that Zn in farmland soils in Jiangsu Province exhibited an uneven distribution, with an average Zn content of approximately 70 mg/kg. The available Zn accounted for about 20% of the total Zn in soils. The total Zn in soils dictated the distribution of available Zn, with a significant positive correlation between both. Significant factors influencing the enrichment and distribution of Zn in soils included soil texture, genetic type, total organic carbon (TOC) content, and Fe content. Among various soils, limestone soils were the most enriched in Zn in Jiangsu Province, while the Holocene marine silty soils served as soil parent materials most enriched in Zn. Zn in soils manifested (relatively) significant positive correlations with Se, Cu, Fe, Al, Mo, and TOC contents. Rice seeds showed an average Zn content of 18 mg/kg, with an average bio-concentration factor (BCF) value of 0.2. Zn in rice seeds was significantly positively correlated with Zn, Se, and TOC content in soils but significantly negatively correlated with soil pH. In contrast, wheat seeds showed an average Zn content of 28 mg/kg, with an average BCF value of 0.36. Zn in wheat seeds was significantly positively correlated with Zn, Se, B, and TOC content in soils. Additionally, a significant positive correlation between Zn and Se was observed in both rice and wheat seeds. According to the industrial standards, the proportions of zinc-rich soils, zinc-rich rice seeds, and zinc-rich wheat seeds in Jiangsu Province were 11.39%, 29%, and 13.69% respectively, suggesting promising prospects for the development and utilization of zinc-rich soil resources in Jiangsu Province. The development and utilization efficiency of zinc-rich soil resources can be significantly enhanced by combining the production of natural zinc-rich food and the amelioration of farmland soils (e.g., improving TOC content and pH in soils). Overall, the results of this study provide a basis for scientifically utilizing beneficial trace elements such as Zn in soils in Jiangsu Province.

This study investigated the shallow surface substrate in the Taonan area, western Songnen Plain. Accordingly, it revealed the elemental differentiation between surface (0~20 cm) and deeper (150~200 cm) soils in the shallow surface substrate layer, as well as the genetic mechanisms of salinization in this layer. The results show that compared to deeper soils, surface soils in the Taonan area are strongly enriched in organic carbon (Corg) and N and slightly enriched in Br, P, S, Se, and total carbon (TC). In contrast, no significant differences are identified in heavy metals, rare earth elements (REEs), and other trace elements. These findings suggest the primary causes of the enrichment of various element indicators in surface soils include agricultural activities, biogeochemical cycles, and water-salt migration. The factor analysis indicates that for surface soils, factor F1 is dominated by the heavy metal-REE combination (variance contribution rate: 26.66%), with its spatial distribution associated with fluvial deposition and agricultural activities. Furthermore, factor F2 for these soils is the salt-related element combination (including CaO and MgO; variance contribution rate: 11.24%), indicating the risk of salinization in low-lying zones. In contrast, for deeper soils, factor F1 is the combination of elements such as Al₂O₃, B, La, and Sc (variance contribution rate: 27.34%), reflecting the compositional characteristics of bedrocks or soil parent materials. Factor F2 for these soils is the combination of elements related to geological settings and salinity (variance contribution rate: 13.09%), indicating geological settings and salinization. The weathering and leaching coefficient, represented by the Ba value, shows significant spatial differentiation. Compared to deeper soils, surface soils manifest a larger range of high Ba values, primarily distributed in the zone south of Jubao Township and west of Datong Township, as well as the southern part of Erlong Township. This distribution, coinciding with the high-value zones of salt-related factor F2, is principally affected by topography and deep parent material types. In the low-lying plain area and the front of alluvial fans, the low-lying terrains, poor drainage, and intense evaporation lead to salt accumulation, causing a high risk of salinization. In the hilly area, the high values of factor F2 are associated with the bedrock lithology, with salts originating from weathered bedrocks. The results of this study will provide a geochemical basis for land resource optimization and ecological restoration in the Taonan area.

The Chongzuo area, located in southwestern Guangxi, encompasses Jiangzhou District, Daxin County, and Longzhou County, with the majority featuring karst topography. This study focused on 242 samples of rice grains and their corresponding rhizosphere soils from contiguous farmland in the region. These samples were analyzed to measure the contents of 26 elements in the soils, including arsenic (As), cadmium (Cd), and chromium (Cr), as well as the contents of As, Cd, mercury (Hg), and lead (Pb) in rice grains, using inductively coupled plasma mass spectrometry (ICP-MS), inductively coupled plasma optical emission spectrometry (ICP-OES), and atomic fluorescence spectrometry (AFS). By analyzing the characteristics of heavy metals in soils and rice grains, the influencing factors and fitting models of heavy metals from soils to rice grains were investigated. The results indicate that the content of oxides in soil was generally lower than the national average, while the content of heavy metals was relatively high, especially Cd and Hg. As and Cd in soils exhibited relatively high pollution risks. The contents of As, Cd, Hg, and Pb in rice grains from non-karst areas were generally higher than those from karst areas. The contents of As, Cd, Hg, and Pb in rice grains generally complied with food safety standards. As and Pb in rice grains showed significant correlations (dominated by negative correlations) with metal elements, non-metal elements, and oxides in rhizosphere soils, while Cd and Hg exhibited significant correlations with oxides in rhizosphere soils. Various fitting models of As in rice grains generally presented a coefficient of determination (R²) above 0.5, indicating better model performance than those for Cd, Hg, and Pb. After distinguishing between karst and non-karst areas, the R² values of the fitting models were further improved. Among the factors influencing the contents of As, Cd, Hg, and Pb in rice grains, parent material played a more significant role than rice variety. This study preliminarily clarifies the key driving factors of heavy metal transfer in the soil-rice system in the karst area of Chongzuo, providing a theoretical and practical basis for the safe production of agricultural products, classification-based management of contaminated farmland, and policy formulation in similar karst areas of Southwest China.

Advance geological prediction in tunnels faces technical challenges,including strong non-stationarity of ground-penetrating radar(GPR) signals and insufficient resolution of conventional time-frequency analyses.Hence,this study proposed an improved method based on adaptive local maximum synchrosqueezing transform(LMSST).The proposed method significantly enhanced the time-frequency resolution and noise robustness of traditional LMSST through a dynamic bandwidth optimization algorithm and local extremum search strategies.Theoretical analysis and synthetic signal testing demonstrated the superior time-frequency energy concentration characteristics of the proposed method in analyzing cross-frequency modulation components.Furthermore,the proposed method was applied to the karst tunnel section of a high-speed railway in Southwest China.Combined with the GprMax forward modeling and GPR measurements,the proposed method successfully identified geological anomalies such as karst caves.Subsequent excavation verification confirmed the identification accuracy,with positional errors of anomaly boundaries below 0.3 m.Overall,the results of this study suggest the proposed method's efficiency in enhancing time-frequency resolution and substantial engineering applicability,offering reliable technical support for tunnel construction safety in karst areas.

In karst subgrade exploration engineering, the application of the surface multi-electrode resistivity method faces challenges: on the one hand, the limitation of electrode array length constrains the effective exploration depth of this method; on the other hand, as the exploration depth increases, the available current distribution information from deeper layers decreases, resulting in a gradual weakening of the method's resolution capability, particularly in accurately predicting small-scale karst development areas at depth. To address this issue, the article proposes a solution: introducing wellbore electrodes on the basis of traditional surface multi-electrode resistivity observations to achieve “surface-to-wellbore” resistivity data acquisition. This strategy aims to increase and obtain deep current distribution information by adding wellbore electrodes, thereby enhancing the local effective exploration depth and improving the resolution of inversion results. To evaluate the effectiveness of the “surface-to-wellbore” resistivity observation method in obtaining deep karst information, the article first conducts an in-depth analysis using numerical simulation methods. Subsequently, the practical application value and effectiveness of this method are further demonstrated through inversion results based on measured data from karst subgrades. The research results show that the multi-electrode resistivity measurement technique with wellbore electrodes can significantly improve the resolution for identifying deep anomalies, providing an effective path to overcome the limitations of multi-electrode resistivity methods in deep karst exploration.

To reduce the economic losses caused by pipeline leakage, this paper applies a comprehensive geophysical exploration technology integrating ground-penetrating radar, multi-channel transient surface waves, and resistivity imaging to detect pipeline leakage at a power plant in Huainan, Anhui Province. The results show that the ground-penetrating radar oscillatory signals can reveal the leakage zone, the multi-channel transient surface wave can reflect the leakage severity within the detected area, and the electrical resistivity tomography can present the low-resistance morphology of the leakage zone. Demonstrated by the satisfactory outcomes, this integrated geophysical prospecting method proves to be an effective means to accurately locate the leakage positions for similar pipelines.