Search Results (506)

Deep foundation excavation in water-rich sand strata presents complex deformation characteristics driven by fluid–solid interaction, which distinguishes it from excavations in cohesive soft clay. This study investigates the spatiotemporal evolution and deformation mechanisms of retaining structures through a comparative analysis of field monitoring data and 3D numerical simulation, based on a subway station project in Xi’an. While the numerical simulation predicted a continuous “bulging” deformation mode, field monitoring revealed a distinct transition from a “bulging” profile to a “step-like” deformation pattern as the excavation deepened. Quantitatively, while the simulation captured the spatial trend, the measured maximum surface settlement (7.8 mm) exceeded the simulated value (1.2 mm), highlighting the dominant role of seepage consolidation. Detailed analysis indicates that this discrepancy—and the unique step-like evolution—is primarily driven by two mechanisms: the rapid stress relaxation of cohesionless sand during the time lag of support installation, and the superimposed seepage forces induced by continuous dewatering, which are often simplified in standard elastoplastic models. The study further identifies that the vertical displacement of the pile top is governed by the combined effects of basal heave and the “kick-out” deformation at the pile toe. These findings demonstrate that in high-permeability water-rich sand, deformation control depends critically on minimizing the unsupported exposure time of the excavation face. This research provides a theoretical basis for optimizing the spatiotemporal sequencing of excavation in similar geological conditions. Full article

Surface deformation induced by the extraction of natural resources constitutes a non-stationary spatiotemporal process. Modeling surface deformation time series obtained through Interferometric Synthetic Aperture Radar (InSAR) technology using deep learning methods is crucial for disaster prevention and mitigation. However, the complexity of model hyperparameter configuration and the lack of interpretability in the resulting predictions constrain its engineering applications. To enhance the reliability of model outputs and their decision-making value for engineering applications, this study presents a workflow that combines a Tree-structured Parzen Estimator (TPE)-based Bayesian optimization approach with ensemble inference. Using the Rhineland coalfield in Germany as a case study, we systematically evaluated six deep learning architectures in conjunction with various spatiotemporal coding strategies. Pairwise comparisons were conducted using a Welch t-test to evaluate the performance differences across each architecture under two parameter-tuning approaches. The Benjamini–Hochberg method was applied to control the false discovery rate (FDR) at 0.05 for multiple comparisons. The results indicate that TPE-optimized models demonstrate significantly improved performance compared to their manually tuned counterparts, with the ResNet+Transformer architecture yielding the most favorable outcomes. A comprehensive analysis of the spatial residuals further revealed that TPE optimization not only enhances average accuracy, but also mitigates the model’s prediction bias in fault zones and mineralize areas by improving the spatial distribution structure of errors. Based on this optimal architecture, we combined the ten highest-performing models from the optimization stage to generate a quantile-based susceptibility map, using the ensemble median as the central predictor. Uncertainty was quantified from three complementary perspectives: ensemble spread, class ambiguity, and classification confidence. Our analysis revealed spatial collinearity between physical uncertainty and absolute residuals. This suggests that uncertainty is more closely related to the physical complexity of geological discontinuities and human-disturbed zones, rather than statistical noise. In the analysis of super-threshold probability, the threshold sensitivity exhibited by the mining area reflects the widespread yet moderate impact of mining activities. By contrast, the fault zone continues to exhibit distinct high-probability zones, even under extreme thresholds. It suggests that fault-controlled deformation is more physically intense and poses a greater risk of disaster than mining activities. Finally, we propose an engineering decision strategy that combines uncertainty and residual spatial patterns. This approach transforms statistical diagnostics into actionable, tiered control measures, thereby increasing the practical value of susceptibility mapping in the planning of natural resource extraction. Full article

Deep coalbed methane (CBM) development in the Eastern Ordos Basin indicates that strong vertical heterogeneity within the Benxi Formation No. 8 thick coal seam can severely constrain well productivity. Here, twelve coal samples from two typical wells (W1: upper coal seams; W2: lower coal seams) were analyzed to quantify vertical variability in pore structure and its controls. Proximate and maceral analyses were combined with low-temperature N₂ adsorption (2–100 nm) and CO₂ adsorption (<2 nm) to characterize mesopores and micropores, respectively; mono-fractal and multifractal approaches were further applied to quantify pore-system heterogeneity. The results indicate that upper coal seams (W1) exhibit more developed micropores and stronger adsorption capacity, while the lower coal seams (W2) display more significant heterogeneity in pore structure, particularly at the micropore scale. Ash content is identified as the dominant control factor for vertical variations in pore characteristics, showing a negative correlation with both micropore and mesopore volumes, while coal rank and maceral composition exert secondary influences. A vertical zoning model has been established based on multiple parameters: the upper section is classified as a high-quality sweet-spot interval, whereas only localized layers in the lower section retain development potential. These findings can serve as a geological basis for optimizing target layer selection and fracturing design in deep coalbed methane wells. Full article

The Datangpo-type Mn ore deposits in northeastern Guizhou (southern China) are a relatively newly discovered type of sedimentary exhalative manganese ore deposit. Previous three-dimensional geological modeling has revealed an NW-trending trough-like depression that obliquely intersects the ENE-trending Nanhua Rift within the Nanhua System in this area. This depression likely represents a paleorift that was present before the metallogenetic period; its intersection with the Nanhua Rift corresponds precisely with the area in which a series of super-large and large new-type Mn ore deposits are located. Here, we used remote sensing image processing techniques, along with hierarchical spatial data fusion and mining methods adopted for exploration, to investigate this paleorift. Specifically, Bouguer gravity data were used to obtain middle–lower-crust structural information; aeromagnetic ΔT data were used to obtain middle–upper-crust structural information; and remote sensing and outcrop data coupled with regional geological survey, mineral exploration, and geochemical exploration data were used to obtain near-surface structural information. Combining these data, we determined the control that different deep tectonic frameworks exert on the formation and distribution of Mn ore deposits within the study area. This study proposes a new conceptual method and technical protocol permitting an improved understanding of the material source and mineralization pattern of Mn ore deposits within the study area, while verifying the existence of the NW-trending Tongren Paleorift. Full article

The Loess Plateau is highly susceptible to landslides due to its fragile geological structure and frequent human activities, particularly old landslides with historical structural damage. The features of these landslides in remote sensing images become blurred over time, leading to huge challenges in detection. Considering that old landslides exhibit obvious shape characteristics, we propose ResU-SPMNet, a deep learning model that integrates shape characteristics into the baseline ResU-Net. The proposed model consists of three components: ResU-Net, shape prior module (SPM), and the atrous spatial pyramid pooling (ASPP) module, which jointly enhance segmentation performance from the perspectives of shape constraints and multi-scale feature representation. To validate the effectiveness of the proposed approach, old landslides in representative regions of the Loess Plateau were selected as the study targets. Results show that the proposed model outperforms ResU-Net, SegNet, MultiResUnet, and DeepLabv3+ in old landslide segmentation, achieving an F1-score of 0.6669 and an MCC of 0.6167. Moreover, generalization tests conducted in independent regions indicate that the model exhibits strong robustness across different seasons. The best performance is achieved in summer, whereas performance declines in winter due to adverse factors such as reduced illumination and snow or ice cover. Full article

Accurately delineating soil elements from satellite imagery is fundamental for regional geological mapping and survey. However, vegetation cover and complex geomorphological conditions often obscure diagnostic surface information, weakening the visibility of key geological features. Additionally, long-term tectonic deformation and weathering processes reshape the spatial organization of soil elements, resulting in substantial within-class variability, inter-class spectral overlap, and fragmented structural patterns—all of which hinder reliable segmentation performance for conventional deep learning approaches. To mitigate these challenges, this study introduces a Reinforced Feature and Multiscale Feature Fusion Network (RFMFFNet) tailored for semantic interpretation of soil elements. The model incorporates a rectangular calibration attention (RCA) module into a ResNet101 backbone to recalibrate feature responses in critical regions, thereby improving scale adaptability and the preservation of fine geological structures. A complementary multiscale feature fusion (MFF) component is further designed by combining sparse self-attention with pyramid pooling, enabling richer context aggregation while reducing computational redundancy. Comprehensive experiments on the Landsat-8 and Sentinel-2 datasets verify the effectiveness of the proposed framework. RFMFFNet consistently achieves superior segmentation performance compared with several mainstream deep learning models. On the Landsat-8 dataset, the oPA and mIoU increase by 2.4% and 2.6%, respectively; on the Sentinel-2 dataset, the corresponding improvements reach 4.3% and 4.1%. Full article

The rise of artificial intelligence has led to the emergence of geoscience knowledge graphs (GeoKG) as effective tools for organizing and representing complex knowledge. The growing complexity of geoscience data calls for innovative strategies for structuring and interpreting extensive information. Conventional knowledge extraction methods often rely on manual annotation and deep learning techniques, which can be costly and inefficient. Herein, we leverage a large language model (LLM) to address the challenges of knowledge extraction and fusion in creating a knowledge graph focused on gold deposits. First, we developed an ontology explicitly designed for gold deposits, drawing on insights from geological experts. Next, we formulate a prompt to guide the LLM to accurately extract geological entities and their semantic relationships in accordance with the knowledge graph schema. Subsequently, we conducted geological entity alignment and integration to construct the gold deposit knowledge graph, which encompasses over 3738 entities and 3900 semantic relationships. Finally, we identified an optimal configuration balancing F1-score and computational cost through comparative experiments on locally deployed models with varying parameters. Our findings demonstrate that an LLM can effectively capture long-range contextual relationships to identify geological entities and their semantic connections, demonstrating strong performance in handling diverse expressions. Full article

Landslides are geological hazards that endanger socioeconomic development and ecological security, with landslide susceptibility mapping (LSM) playing a critical role in risk management and spatial planning. Recently, ensemble learning (EL) models have gained attention for effectively addressing the limitations of individual deep learning (DL) models in LSM. However, EL models always built on single-pixel, multi-factor inputs struggle to capture the spatial structure features of terrain units, limiting their ability to depict complex disaster patterns. Moreover, the scarcity of landslide samples and high annotation costs constrain model performance in LSM. To overcome these challenges, we propose a Stacking model based on multidimensional feature collaboration and pseudo-labeling techniques, referred to as MFP_Stacking. A stacking EL model is first employed in MFP_Stacking to integrate global statistical attribute features extracted from one-dimensional vectors with multi-scale spatial topological features derived from three-dimensional vectors. This strategy of multidimensional feature collaborative modeling enhances the model’s ability to learn complex environmental patterns associated with landslides. Subsequently, pseudo-labeling techniques are adopted to incorporate unlabeled data into auxiliary training, thereby addressing the problem of sample scarcity. MFP_Stacking was applied to LSM in the Zigui–Badong section of the Yangtze River Basin and in Ya’an City, Sichuan Province. Experimental results demonstrate that the proposed model performs well in overcoming limitations in feature representation, alleviating sample scarcity, and enhancing the quality of LSM outcomes. It achieved an average improvement of 2.4% for the Zigui–Badong section and 2% for Ya’an City across various evaluation metrics compared to other models. Full article

Against the background of global energy transformation and low-carbon development, numerous difficult-to-mine coal resources (e.g., deep, thin coal seams and low-quality coal) remain underdeveloped, leading to potential resource waste. This study systematically summarizes the feasibility of developing these resources via underground coal gasification (UCG) technology, clarifies its basic chemical/physical processes and typical gas supply/gas withdrawal arrangements, and establishes an analytical framework covering resource utilization, gas production quality control, environmental impact, and cost efficiency. Comparative evaluations are conducted among UCG, surface coal gasification (SCG), natural gas conversion, and electrolysis-based hydrogen production. Results show that UCG exhibits significant advantages: wide resource adaptability (recovering over 60% of difficult-to-mine coal resources), better environmental performance than traditional coal mining and SCG (e.g., less surface disturbance, 50% solid waste reduction), and obvious economic benefits (total capital investment without CCS is 65–82% of SCG, and hydrogen production cost ranges from 0.1 to 0.14 USD/m³, significantly lower than SCG’s 0.23–0.27 USD/m³). However, UCG faces challenges, including environmental risks (groundwater pollution by heavy metals, syngas leakage), geological risks (ground subsidence, rock mass strength reduction), and technical bottlenecks (difficult ignition control, unstable large-scale production). Combined with carbon capture and storage (CCS) technology, UCG can reduce carbon emissions, but CCS only mitigates carbon impact rather than reversing it. UCG provides a large-scale, stable, and economical path for the efficient clean development of difficult-to-mine coal resources, contributing to global energy structure transformation and low-carbon development. Full article

The stability of underground excavations is a critical factor in the safety and efficiency of mining operations, particularly in structurally complex and geomechanically variable rock mass. This study presents a comparative evaluation of empirical and numerical methods for the design of tunnel support systems in the Gilar underground mine, located in the Gedabek Contract Area of Azerbaijan. To validate and optimize the empirical Q-system-based support designs, Finite Element Method (FEM) simulations were conducted using RS2 software. These simulations enabled the modeling of stress distribution, deformation, and support–rock interaction under in situ conditions. Critical sections along the main ramp were analyzed in detail to determine safety factors during excavation and post-support installation. The study reveals that, although the Q-system provides a practical and time-efficient method for support selection, it may underestimate the reinforcement required in highly fractured or low-strength zones. Numerical modeling proved to be essential in identifying zones with low strength factors and in optimizing support configurations by adjusting rockbolt spacing and shotcrete thickness. The hybrid approach adopted in this study—empirical classification followed by numerical verification and optimization—demonstrated significant improvements in long-term tunnel stability. This research highlights the importance of integrating empirical and numerical approaches for robust ground support design in underground mining. The proposed methodology not only enhances the accuracy of support recommendations but also provides a more reliable basis for decision-making in complex geological settings. The results are particularly relevant for deep and geologically active mines requiring long-term stability of access tunnels. Full article

High-precision gravity surveys are effective in detecting concealed geological structures and mineral deposits with density contrasts. In this study, 754 dense gravity measurements (average accuracy: 0.0043 mGal, or 4.3 × 10⁻⁸ m/s²) were deployed in Dingzhai Township, northeastern Zhejiang, China, to investigate concealed ore bodies and structural controls on mineralization. Using the mean-field method for source-field separation of Bouguer anomalies, combined with density inversion and edge detection, we delineated subsurface density distributions and fault systems. A newly identified “tongue-shaped” high-density anomaly near Xiashadi is interpreted as resulting from local upward intrusion of intermediate-acid porphyry from the Chencai Group basement, indicating significant exploration potential. Beneath Quaternary cover, a previously unrecognized east–west-trending concealed fault was detected, which may have controlled the structural evolution of mineralization at the Daqi’ao Ag deposit and Miaowan Cu deposit. Gravity profile inversion reveals a deep high-density anomaly beneath Xie’ao–Xi’ao’an, possibly representing the deep extension of the Hengtang Cu–Mo deposit. Low-density anomalies near Chenxi and Dongli villages are attributed to Early Cretaceous low-density intrusions (e.g., monzogranite) and multi-phase volcanism in the Shangshawan caldera. This work provides robust geophysical constraints for deep mineral exploration and advance understanding of the metallogenic tectonic evolution in northeastern Zhejiang. Full article

A central challenge in tight fault-depression reservoirs is understanding how three-dimensional fracture structures control gas storage and flow. This study introduces a data-driven, geologically informed framework that integrates structural-mechanical coupling to decipher fracture networks within the Shahezi Formation. Our model, based on rock failure criteria, achieves quantitative fracture prediction across one-dimensional to three-dimensional scales. This capability overcomes the limitations inherent in single-method approaches for tight, fracture-dominated reservoirs. By synthesizing sedimentary facies-controlled reservoir modeling, sweet-spot inversion, and geo-engineering integration, we establish a predictive system for accurate reservoir assessment. The continental clastic Shahezi Formation is typified by secondary fractures. This study utilizes leverage small-scale data (core, thin section, log) to quantify key parameters (fracture density, aperture), enabling a systematic analysis of fracture typology, heterogeneity, and controls. Building on this foundation, and spatially constrained by large-scale datasets (seismic interpretation, stress-field simulations), we developed a robust fracture development model for deep tight reservoirs. Stress-field modeling delineated fracture-prone zones, where a discrete fracture network (DFN) model was built to characterize 3D fracture geometry and connectivity. Integrating simulated fracture size and aperture-derived permeability allowed us to quantify fracture contribution to total permeability, ultimately mapping favorable targets. The results identify favorable zones primarily in the western sector of the study area, forming an NS-trending, belt-like distribution. They are mainly concentrated around the wells Changshen-4, Changshen-40, and Changshen-41. This distribution is clearly controlled by the Qianshenzijing Fault. Full article

Xinfeng County, Shaoguan City, Guangdong Province, China, is a typical vegetation-covered area that suffers from severe attenuation of rock and mineral spectral information in remote sensing images owing to dense vegetation. This situation limits the accuracy of traditional lithological mapping methods, making them unable to meet geological mapping demands under complex conditions, and thus necessitating a tailored lithological identification model. To address this issue, in this study, the penetration capability of microwave remote sensing (for extracting indirect textural features of lithology) was combined with the spectral superiority of hyperspectral remote sensing (for capturing lithological spectral features), resulting in a dual-branch deep-learning framework for lithological classification based on multisource remote sensing data. The framework independently extracts features from Sentinel-1 imagery and Gaofen-5 data, integrating three key modules: texture feature extraction, spatial–spectral feature extraction, and attention-based adaptive feature fusion, to realize deep and efficient fusion of heterogeneous remote sensing information. Ablation and comparative experiments were conducted to evaluate each module’s contribution. The results show that the dual-branch architecture effectively captures the complementary and discriminative characteristics of multimodal data, and that the encoder–decoder structure demonstrates strong robustness under complex conditions such as dense vegetation. The final model achieved 97.24% overall accuracy and 90.43% mean intersection-over-union score, verifying its effectiveness and generalizability in complex geological environments. The proposed multi-source remote sensing–based lithological classification model overcomes the limitations of single-source data by integrating indirect lithological texture features containing vegetation structural information with spectral features, thereby providing a viable approach for lithological mapping in vegetated regions. Full article

During preliminary reconnaissance at a hydropower station site in Southwestern China, a unique phenomenon of deep-seated fractures was identified within the slopes, which were symmetrically developed on both banks. These features occur within unloading zones and manifest as tensile fractures with deep-seated fractures exhibiting unloading characteristics. This study systematically analyzes the spatial distribution, developed patterns, and structural attributes of these deep fractures. Through numerical model of stress field dynamics during valley evolution, we investigate the relationship between stress states and deep fracture formation. Research demonstrates that these fractures result from energy release through unloading at stress-concentration zones in slope interiors, driven by rapid valley incision under high in situ stress conditions. This process is further conditioned by specific slope geometries, rock mass structures, and geomorphic settings. Crucially, river incision rate governs fracture depth, while the number of incision cycles significantly controls fracture aperture. These findings provide a theory for understanding deep-seated slope failure mechanisms and engineering mitigation in analogous geological environments. Full article

Vertical shafts are the lifelines of coal mines, serving as critical conduits for resources and personnel. However, the long-term exposure of shaft walls to groundwater erosion significantly reduces their service life and increases the risk of structural failures. This issue is particularly pressing in Inner Mongolia and Henan Provinces, two of China’s major coal-producing regions, where the challenge of sulfate attack on shafts in deep stratigraphic environments has become a growing concern. This study focused on the corrosion damage observed in these two typical auxiliary shafts: the net diameters and depths of the auxiliary shafts in Shunhe Coal Mine and Mataihao Coal Mine are 6 m and 768.5 m and 9.2 m and 457 m, respectively. The rock section shaft walls in the study range from 5 to 10 m in thickness and are constructed using C40 to C60 grade concrete. To assess the extent of this damage, we conducted a comprehensive analysis of shaft wall samples using water analysis, XRD (X-ray diffraction) analysis, FT-IR (Fourier transform infrared) spectroscopy, and XRF (X-ray fluorescence) analysis. The findings reveal that the identified secondary sulfate reaction products within the shaft wall concrete include calcium sulfate, gypsum, ettringite, and thaumasite. The CaO loss rates in the auxiliary shaft walls of Shunhe Coal Mine and Mataihao Coal Mine are as high as 66% and 47%, respectively. Additionally, the concentrations of SO₃ and MgO in both mines exceed normal levels by up to 5 and 11 times, and 13 and 3 times, respectively. Despite this, severe corrosion is primarily confined to the inner surface of the auxiliary shaft walls, without significant penetration into the deeper shaft structure. The corrosion damage is predominantly concentrated in the shaft sections where the geological environment is characterized by bedrock. This study provides field evidence and laboratory analyses to inform the mitigation of sulfate attack in auxiliary shafts. Full article