Natural Resources Research最新文献

Water–rock interactions affect mineral inclusions and the pore structure of rock, subsequently affecting its mechanical and seepage properties. A method for quantitative analysis of the pore and mineral inclusion evolution characteristics of coal samples based on CT scanning is proposed. Accordingly, numerical model construction and block division of mineral inclusions and pores in coal samples were realized. The effects of mineral inclusions and the pore structure on coal failure were simulated and analyzed. The results showed that the porosity and pore distribution in coal influence its strength. The development of plastic zones in coal affected by pores can be divided into three stages: (1) tensile failure initiation stage, (2) shear failure penetration stage, and (3) failure rapid expansion stage. The higher the fractal dimension of the pores is, the greater the strength of coal. Pores and mineral inclusions degrade the strength of coal and accelerate the development of plastic zones. In the loading process, plastic zones preferentially emerge around pores and mineral inclusions. The plastic zones around mineral inclusions connect gradually with those around pores, thus accelerating coal failure.

We present here the first experimental science (consensus)-based mineral prospectivity mapping (MPM) method and its validation results in the form of national prospectivity maps and datasets for PGE–Ni–Cu–Cr and Witwatersrand-type Au deposits in South Africa. The research objectives were: (1) to develop the method toward applicative uses; (2) to the extent possible, validate the effectiveness of the method; and (3) to provide national MPM products. The MPM method was validated by targeting mega-deposits within the world’s largest and best exploited geological systems and mining districts—the Bushveld Complex and the Witwatersrand Basin. Their incomparable knowledge and mega-deposit status make them the most useful for validating MPM methods, serving as “certified reference targets”. Our MPM method is built using scientific consensus via deep ensemble construction, using workflow experimentation that propagates uncertainty of subjective workflow choices by mimicking the outcome of an ensemble of data scientists. The consensus models are a data-driven equivalent to expert aggregation, increasing confidence in our MPM products. By capturing workflow-induced uncertainty, the study produced MPM products that not only highlight potential exploration targets but also offer a spatial consensus level for each, de-risking downstream exploration. Our MPM results agree qualitatively with exploration and geological knowledge. In particular, our method identified areas of high prospectivity in known exploration regions and geologically and geospatially corresponding to the known extents of both mineral systems. The convergence rate of the ensemble demonstrated a high level of statistical durability of our MPM products, suggesting that they can guide exploration at a national scale until significant new data emerge. Potential new exploration targets for PGE–Ni–Cu–Cr are located northwest of the Bushveld Complex; for Au, promising areas are west of the Witwatersrand Basin. The broader implications of this work for the mineral industry are profound. As exploration becomes more data-driven, the question of trust in MPM products must be addressed; it can be done using the proposed scientific method.

Graphical Abstract

Mineral resources are estimated to establish potential orebody with acceptable quality (grade) and quantity (tonnage) to validate investment. Estimating mineral resources is associated with uncertainty from sampling, geological heterogeneity, shortage of knowledge and application of mathematical models at sampled and unsampled locations. The uncertainty causes overestimation or underestimation of mineral deposit quality and/or quantity, affecting the anticipated value of a mining project. Therefore, uncertainty is assessed to avoid any likely risks, establish areas more prone to uncertainty and allocate resources to scale down potential consequences. Kriging, probabilistic, geostatistical simulation and machine learning methods are used to estimate mineral resources and assess uncertainty, and their applicability depends on deposit characteristics, amount of data available and expertise of technical personnel. These methods are scattered in the literature making them challenging to access when needed for uncertainty quantification. Therefore, this review aims to compile information about uncertainties in mineral resource estimation scatted in the literature and develop a knowledge base of methodologies for uncertainty quantification. In addition, mineral resource estimation comprises different interdependent steps, in and through which uncertainty accumulates and propagates toward the final estimate. Hence, this review demonstrates stepwise uncertainty propagation and assessment through various phases of the estimation process. This can broaden knowledge about mineral resource estimation and uncertainty assessment in each step and increase the accuracy of mineral resource estimates and mining project viability.

This paper describes a comprehensive approach to predict sand production in the Karazhanbas oilfield using machine learning (ML) techniques. By analyzing data from 2000 wells, the research uncovered the complex dynamics of sand production and emphasized the critical need for accurately predicting the peak sand mass and its occurrence time. ML techniques can have a significant impact on prediction of sand production and on the optimization of oilfield operation, which can be improved with the combined use of enriched training data and domain-specific knowledge. The research underscored the influence of geological factors, especially fault proximity, on prediction accuracy. Domain and field knowledge is needed to formulate different production scenarios for prediction purposes such that the relevant data can be selected for the training of ML models. Moreover, new metrics are needed to evaluate model performance as the applied method is tailored for different operational strategies. As the peak sand mass is considered a pivotal event in field operation, new metrics in terms of peak prediction accuracy and peak time prediction accuracy were introduced to evaluate the performance of ML models. A suite of ML algorithms was employed in the study, which demonstrated notable accuracy in the classification of sand-producing wells.

High-resolution mapping of deep-sea polymetallic nodules is needed (a) to understand the reasons behind their patchy distribution, (b) to associate nodule coverage with benthic fauna occurrences, and (c) to enable an accurate resource estimation and mining path planning. This study used an autonomous underwater vehicle to map 37 km² of a geomorphologically complex site in the Eastern Clarion–Clipperton Fracture Zone. A multibeam echosounder system (MBES) at 400 kHz and a side scan sonar at 230 kHz were used to investigate the nodule backscatter response. More than 30,000 seafloor images were analyzed to obtain the nodule coverage and train five machine learning (ML) algorithms: generalized linear models, generalized additive models, support vector machines, random forests (RFs) and neural networks (NNs). All models ML yielded similar maps of nodule coverage with differences occurring in the range of predicted values, particularly at parts with irregular topography. RFs had the best fit and NNs had the worst spatial transferability. Attention was given to the interpretability of model outputs using variable importance ranking across all models, partial dependence plots and domain knowledge. The nodule coverage is higher on relatively flat seafloor ( < 3°) with eastward-facing slopes. The most important predictor was the MBES backscatter, particularly from incident angles between 25 and 55°. Bathymetry, slope, and slope orientation were important geomorphological predictors. For the first time, at a water depth of 4500 m, orthophoto-mosaics and image-derived digital elevation models with 2-mm and 5-mm spatial resolutions supported the geomorphological analysis, interpretation of polymetallic nodules occurrences, and backscatter response.

Among the various hazards induced by underground coal mining, surface subsidence tends to cause structural damage to the ground. Therefore, accurate prediction and evaluation of surface subsidence are significant for ensuring mining security and sustainable development. Traditional methods like the probability integral method provide effective predictions. However, these methods do not take into account the consolidation behavior of thick soil layers. In this study, based on the principle of superposition, an improved probability integral method that includes surface subsidence caused by rock layer movement and the consolidation behavior of thick soil layers is developed. The proposed method was applied in the Zhaogu No. 2 coal mine, located in the Jiaozuo mining area. Utilizing unmanned surface vehicle measurement technology, it was found that the maximum subsidence values of the two survey lines were 5.441 m and 4.842 m, with maximum subsidence rate of 62.9 mm/day at observation points. Experimental tests have shown that surface subsidence in deep mining areas with thin bedrock and thick soil layers exhibited a large subsidence coefficient and a wide range of subsidence, closely related to the consolidation behavior of thick soil layers. After verification, compared to the probability integral method, the improved probability integral method incorporating soil consolidation showed a 14.7% reduction in average error and a 22% reduction in maximum error. Therefore, the improved probability integral method proposed can be a very promising tool for forecasting and evaluating potential geohazards in coal mining areas.

The nanometer-sized pores within coal are the primary sites for CH₄ adsorption and competitive adsorption with CO₂. Reasonable modification of the nanopore structure to enhance CH₄ desorption, diffusion rates, and CO₂ competitive adsorption effects can enhance significantly coalbed methane (CBM) production. However, ultrasonic synchronous modification of multiple features of nanopores leads to complex and variable gas adsorption behaviors in coal. To reveal the effect of ultrasonic modification of coal nanopores on gas adsorption, pore measurement experiments and molecular simulation studies were conducted. The results showed that the volume ratio of diffusion pores to adsorption pores (V₂/V₁) decreased significantly after ultrasonic excitation. In the original coal sample, V₂/V₁ was 3.05, while in the coal sample after ultrasonic treatment, V₂/V₁ ranged from 0 to 2.54. With decrease in the proportion of the volume of diffusion pores, the proportion of CH₄ migration from the pore walls of the adsorption pores increased continuously. The proportion of CH₄ migration from the pore walls of the diffusion pores to the pore space of the diffusion pores decreased continuously. The results of gas–solid interaction energy calculation showed that ultrasonic treatment of coal decreases the V₂/V₁ ratio, leading to 7.1–23.3% increase in CO₂ competitive adsorption effect. It also resulted in 4–49% improvement in competitive adsorption efficiency. Additionally, based on gas–solid interaction energy data, an adsorption capacity evaluation model for coal under different gas compositions and pore volume ratios was constructed. The findings can guide ultrasonic-enhanced CBM.

The evaluation of oil- and gas-bearing areas (OGBAs) during the pre-exploration stage has always posed challenges due to the lack of an effective geological evaluation model and validation data. This paper introduces a novel quantitative evaluation method based on the vectorization of key geological factors related to hydrocarbon accumulation. In this study, we focused on the Lianglitage Formation in the Central Uplift Belt and aimed to evaluate the application of the proposed method to the OGBA in the Tarim Basin. First, the reservoir-forming parameters were quantified based on geological analysis and expert experience. Second, the weights of the main parameters were determined using a combination of the gray correlation method and expert knowledge. Finally, the OGBA was evaluated using a multifactor fusion method. The comprehensive evaluation results indicate that the platform margin in the northeastern part of the Katake Uplift shows promise for exploration, while the southern region has a good potential for future exploration. This study emphasizes the significance of selecting key factors and vectorizing evaluation parameter mapping for accurate and quantitative evaluation of an OGBA. The results of this study provide a valuable foundation for evaluating the OGBAs in the Lianglitage Formation within the Tarim Basin and offer a valuable reference for OGBAs in similar regions during the pre-exploration stage.

Prospectivity mapping based on deep learning typically requires substantial amounts of geological feature information from known mineral deposits. Due to the limited spatial distribution of ore deposits, the training of predictive models is often hampered by insufficient positive samples. Meanwhile, data-driven mineral prospectivity mapping often overlooks domain knowledge and expert experience, leading to poor interpretability of predictive results. To address this problem, we employed the Gaussian mixture model (GMM) for spatial feature classification to expand the number of positive samples. The approach integrated the embedding of geological map knowledge graphs with geological exploration data to enhance the knowledge constraints of the prospecting model, which enabled the integration of knowledge with data. Considering the complex spatial structure of geological elements, a bi-branch utilizing the 1-dimensional convolutional neural network (CNN1D) and graph convolutional network (GCN) was used to extract geological spatial features for model training and prediction. To validate the effectiveness of the method, a gold mineralization prediction study was conducted in the Wulonggou area (Qinghai province, western China). The results indicate that, when the number of GMM spatial feature classifications was 17, the positive-to-negative sample ratio was optimal, and the embedding of the knowledge graph controlled the prediction area distribution effectively, which demonstrated strong consistency between the prospecting area and the known mineral deposits. Compared with the predictions by CNN1D, the fused prediction model of CNN1D and GCN yielded higher accuracy. Our model identified 11 classes of mineralization potential areas and provides geological interpretations for different prediction categories.

Mineral exploration combined with prospectivity mapping has become the standard process for utilising mineral exploration data. Nowadays, most techniques integrate multiple layers of information and use machine learning for both data-driven and knowledge-driven approaches. This study introduces a novel and generalised methodology for comparing different layers of information by using superpixels instead of pixels to identify similarities. This methodology provides an enhanced statistical representation of regions, facilitating and enabling effective comparisons. Three different statistical distance measures were considered: Kullback–Leibler divergence, Wasserstein distance and total variation distance. We apply the proposed process to data from the Antofagasta region of northern Chile, a well-known area for metallogenic belts, that contain notable copper reserves. Each metric was used and compared, resulting in different similarity maps highlighting interesting mineral exploration areas. The study results lead to the conclusion that the proposed methodology can be applied at different scales and helps in the identification of areas with similar characteristics.