A Monte Carlo-based framework for risk-return analysis in mineral prospectivity mapping

Quantification of a mineral prospectivity mapping(MPM)heavily relies on geological,geophysical and geochemical analysis,which combines various evidence layers into a single map.However,MPM is subject to considerable uncertainty due to lack of understanding of the metallogenesis and limited spatial data samples.In this paper,we provide a framework that addresses how uncertainty in the evidence layers can be quantified and how such uncertainty is propagated to the prediction of mineral potential.More specifically,we use Monte Carlo simulation to jointly quantify uncertainties on all uncertain evidence variables,categorized into geological,geochemical and geophysical.On stochastically simulated sets of the multiple input layers,logistic regression is employed to produce different quantifications of the mineral potential in terms of probability.Uncertainties we address lie in the downscaling of magnetic data to a scale that makes such data comparable with known mineral deposits.Additionally,we deal with the limited spatial sampling of geochemistry that leads to spatial uncertainty.Next,we deal with the conceptual geological uncertainty related to how the spatial extent of the influence of evidential geological features such as faults,granite intrusions and sedimentary formations.Finally,we provide a novel way to interpret the established uncertainty in a risk-return analysis to decide areas with high potential but at the same time low uncertainty on that potential.Our methods are illustrated and compared with traditional deterministic MPM on a real case study of prospecting skarn Fe deposition in southwestern Fujian,China.

Leucogranite mapping via convolutional recurrent neural networks and geochemical survey data in the Himalayan orogen

Geochemical survey data analysis is recognized as an implemented and feasible way for lithological mapping to assist mineral exploration.With respect to available approaches,recent methodological advances have focused on deep learning algorithms which provide access to learn and extract information directly from geochemical survey data through multi-level networks and outputting end-to-end classification.Accordingly,this study developed a lithological mapping framework with the joint application of a convolutional neural network(CNN)and a long short-term memory(LSTM).The CNN-LSTM model is dominant in correlation extraction from CNN layers and coupling interaction learning from LSTM layers.This hybrid approach was demonstrated by mapping leucogranites in the Himalayan orogen based on stream sediment geochemical survey data,where the targeted leucogranite was expected to be potential resources of rare metals such as Li,Be,and W mineralization.Three comparative case studies were carried out from both visual and quantitative perspectives to illustrate the superiority of the proposed model.A guided spatial distribution map of leucogranites in the Himalayan orogen,divided into high-,moderate-,and low-potential areas,was delineated by the success rate curve,which further improves the efficiency for identifying unmapped leucogranites through geological mapping.In light of these results,this study provides an alternative solution for lithologic mapping using geochemical survey data at a regional scale and reduces the risk for decision making associated with mineral exploration.

Explainable artificial intelligence models for mineral prospectivity mapping

Mineral prospectivity mapping(MPM)is designed to reduce the exploration search space by combining and analyzing geological prospecting big data.Such geological big data are too large and complex for humans to effectively handle and interpret.Artificial intelligence(AI)algorithms,which are powerful tools for mining nonlinear mineralization patterns in big data obtained from mineral exploration,have demonstrated excellent performance in MPM.However,AI-driven MPM faces several challenges,including difficult interpretability,poor generalizability,and physical inconsistencies.In this study,based on previous studies,we devised a novel workflow that aims to constructing more transparent and explainable artificial intelligence(XAI)models for MPM by embedding domain knowledge throughout the AI-driven MPM,from input data to model design and model output.This newly proposed approach provides strong geological and conceptual leads that guide the entire AI-driven MPM model training process,thereby improving model interpretability and performance.Overall,the development of XAI models for MPM is capable of embedding prior and expert knowledge throughout the modeling process,presenting a valuable and promising area for future research designed to improve MPM.

Mineral Exploration Using Subtle or Negative Geochemical Anomalies

Mineral resources prediction and assessment is one of the most important tasks in geosciences.Geochemical anomalies,as direct indicators of the presence of mineralization,have played a significant role in the search of mineral deposits in the past several decades.In the near future,it may be possible to recognize subtle geochemical anomalies through the use of processing of geochemical exploration data using advanced approaches such as the spectrum-area multifractal model.In addition,negative geochemical anomalies can be used to locate mineralization.However,compared to positive geochemical anomalies,there has been limited research on negative geochemical anomalies in geochemical prospecting.In this study,two case studies are presented to demonstrate the identification of subtle geochemical anomalies and the significance of negative geochemical anomalies.Meanwhile,the opportunities and challenges in evaluating subtle geochemical anomalies associated with mineralization,and benefits of mapping of negative anomalies are discussed.

Spatial Analysis of Fe Deposits in Fujian Province, China: Implications for Mineral Exploration

Spatial point pattern statistics, fractal analysis and Fry analysis in support of GIS were applied to explore the spatial distribution characteristics of mineral deposits and the spatial relationships between mineralization and geological features in Fujian Province（China）. The results of Ripley＇s K（r） revealed a clustered distribution of Fe deposits in space with a fractal dimension of 1.38. Fry analysis showed that Fe deposits distributed mainly along a NNE-NE trend. Buffer analysis showed that most of the known Fe deposits developed within 4 km buffer zones of the NNE-NE-trending faults, Yanshanian intrusions, and Late Paleozoic marine sedimentary rocks and the carbonate formations（C–P Formation）, indicating that they possibly control the spatial distribution of Fe mineralization. This is possibly because the NNE-NE-trending faults, Yanshanian intrusions, and C–P Formation provided pathways of fluids, energy and a part of metal, and zones of deposition for the Fe mineralization, respectively. The fractal relation of the number of Fe deposits occurring within the buffer zones of geological features was observed. The fractal dimension suggested that the significance of Yanshanian intrusions and C–P Formation are greater than that of NNE-NE-trending faults in controlling the formation of Fe mineralization. These findings are useful for better understanding the formation of the mineralization and provide significant information for further mineral exploration.

A Fractal Measure of Spatial Association between Landslides and Conditioning Factors

Measuring the relative importance and assigning weights to conditioning factors of land- slides occurrence are significant for landslide prevention and/or mitigation. In this contribution, a fractal method is introduced for measuring the spatial relationships between landslides and condition- ing factors （such as faults, rivers, geological boundaries, and roads）, and for assigning weights to condi- tioning factors for mapping of landslide susceptibility. This method can be expressed as p=C~, where d is the fractal dimension, and C is a constant. This relationship indicates a fractal relation between land- slide density （p） and distances to conditioning factors （6）. The case of d〉0 suggests a significant spatial correlation between landslides and conditioning factors. The larger the d （〉0） value, the stronger the spatial correlation is between landslides and a specific conditioning factor. Two case studies in South China were examined to demonstrate the usefulness of this novel method.

Mapping Spatial Distribution Characteristics of Lineaments Extracted from Remote Sensing Image Using Fractal and Multifractal Models

Mapping mineral prospectivity in vegetated areas is a challenge. For this reason, we aimed to map spatial distribution characteristics of linear structures detected in remote sensing images using fractal and multifractal models. The selected study area was the Pinghe District of the Fujian Province（China）, located in the Shanghang-Yunxiao polymetallic and alunite ore belt（within the Wuyishan polymetallic belt）, where mineral resources such as copper, molybdenum, gold, silver, iron, lead, zinc, alunite and pyrophyllite have been discovered. The results of our study showed that：（1） the values of fractal dimension for all lineaments, NW-trending lineaments, and NE-trending lineaments, are 1.36, 1.32, and 1.23, respectively, indicating that these lineaments are statistically self-similar;（2） the fractal dimensions of the spatial distribution of the linear structures in the four selected hydrothermal-type ore deposits of the Pinghe District, named Zhongteng, Panchi, Xiaofanshan and Fanshan, are 1.43, 1.52, 1.37 and 1.37, respectively, which are higher than the mean value in South China;（3） the spatial distribution of the linear structures extracted from the remote sensing image and displayed by the contour map of fractal dimensions, correlates well with the known hydrothermal ore deposits; and（4） the results of the anomaly map decomposed by the spectrum-area（S-A） multifractal model is much better than the original fractal dimension contour map, which showed most of the known hydrothermal-type deposits occur in the high anomalous area. It is suggested that a high step tendency possibly matches with the boundary of the volcanic edifice and the deep fault controlling the development of the rock mass and the volcanic edifice. The complexity of the spatial distribution of mapped lineations（faults） in the Pinghe District, characterized by high values in the anomaly map, may be associated with the hydrothermal polymetallic ore mineralization in the study area.

Preface to the Special Issue on Digital Geosciences and Quantitative Exploration of Mineral Resources

Prof.Pengda Zhao is one of the pioneers in the fields of mathematical geology and mineral exploration in China and an honorary member of the International Association for Mathematical Geosciences(IAMG).During the 29 th International Geological Congress,held in Kyoto,Japan,in August 1992.