Prospectivity Mapping Research Articles

Geologically-constrained GANomaly network for mineral prospectivity mapping through frequency domain training data

Generative adversarial networks (GAN) and various deep autoencoders have been frequently executed to recognize multi-element geochemical anomalies linked to different ore resources in recent decade. Efficient recognition of multi-element geochemical anomaly patterns is a significant issue in mineral exploration targeting. Traditional procedures have not sufficient capability to perform efficient pattern recognition. While, deep learning algorithms as influential subset of machine learning algorithms can present magnificent conclusions in classification and pattern recognition. Because those have robust ability in extracting high-level features of complex inputs. Although, many deep learning algorithms were used to recognize geochemical anomalies but the GANs have demonstrated specific dignity in recognizing multi-element geochemical anomaly patterns. But, these frameworks should be constrained to learn geological knowledge and yield reasonable potential maps. In this regard, a novel geologically-constrained GANomaly was trained with frequency domain training data to recognize multi-element geochemical anomalies. Application of the geologically-constrained GANomaly network with considering mineral system parameters of the Au–Cu mineralization in the Feyzabad district, NE Iran was eventuated to suitable results. The success-rate curves demonstrated that produced map of frequency domain geochemical data has traced 86.68% Au–Cu occurrences via 30% corresponded area while produced map of spatial domain geochemical data has traced 80.13% Au–Cu occurrences via 30% corresponded area.

Integration of airborne magnetic and geological data for mineral prospectivity mapping in the Ambam-Amvom rainforest area, Archean Ntem Complex, southern Cameroon

The accurate mapping of mineral resources in rainforest areas is challenging due to dense vegetation and thick lateritic cover obscuring rock outcrops. This study utilized airborne magnetic survey data to map geological features and target mineral resources in the Ambam-Amvom area at the NW margin of the Congo craton. Linear and non-linear filtering algorithms were used, including reduction to the equator (RTE), upward continuation (UC) at 1000 m and 2000 m, first vertical derivative (FVD), and analytic signal (AS). Field surveys were carried out for sampling and petrographic investigations. The TMI map showed magnetic signals varying from -4489 nT to 2317 nT, while the RTE map displayed magnetic signals varying from -2984 nT to 3528 nT. The RTE-Reg map exhibited anomaly signal variations ranging between -2704 nT and +3637 nT and between -2829 and +3864 nT, respectively for UC at 1000 m and 2000 m. Geological structures were interpreted from the FVD, AS, and TMI-RTE images, and linear structures of regional extend mainly trend NW-SE, E-W, and NE-SW. Field investigations revealed that the study area comprises iron formations, amphibolite, gneiss, gabbro, biotite granite, and charnockite. The bulk chemical composition of iron formations showed a high content of TFe, averaging 39.80 wt.% for itabirite, 53.79 wt.% for hematite-geothite ore and 59.63 wt.% for quartz-hematite ore. The zones of peak magnetic anomalies coincided with the occurrence of the greenstone belt, and the integration of aeromagnetic and geological data led to the identification of three main target zones for iron ore minimization. High fracture density zones within the research were identified as high prospective targets for gold mineralization. The study's findings highlight the high potential of the Ntem Complex for mineral resources and provide valuable information for future research and development.

Workflow-Induced Uncertainty in Data-Driven Mineral Prospectivity Mapping

The primary goal of mineral prospectivity mapping (MPM) is to narrow the search for mineral resources by producing spatially selective maps. However, in the data-driven domain, MPM products vary depending on the workflow implemented. Although the data science framework is popular to guide the implementation of data-driven MPM tasks, and is intended to create objective and replicable workflows, this does not necessarily mean that maps derived from data science workflows are optimal in a spatial sense. In this study, we explore interactions between key components of a geodata science-based MPM workflow on the geospatial outcome, within the modeling stage by modulating: (1) feature space dimensionality, (2) the choice of machine learning algorithms, and (3) performance metrics that guide hyperparameter tuning. We specifically relate these variations in the data science workflow to the spatial selectivity of resulting maps using uncertainty propagation. Results demonstrate that typical geodata science-based MPM workflows contain substantial local minima, as it is highly probable for an arbitrary combination of workflow choices to produce highly discriminating models. In addition, variable domain metrics, which are key to guide the iterative implementation of the data science framework, exhibit inconsistent relationships with spatial selectivity. We refer to this class of uncertainty as workflow-induced uncertainty. Consequently, we propose that the canonical concept of scientific consensus from the greater experimental science framework should be adhered to, in order to quantify and mitigate against workflow-induced uncertainty as part of data-driven experimentation. Scientific consensus stipulates that the degree of consensus of experimental outcomes is the determinant in the reliability of findings. Indeed, we demonstrate that consensus through purposeful modulations of components of a data-driven MPM workflow is an effective method to understand and quantify workflow-induced uncertainty on MPM products. In other words, enlarging the search space for workflow design and experimenting with workflow components can result in more meaningful reductions in the physical search space for mineral resources.

Assessment of potential cobalt zones by modeling radiometric and ASTER data using the AHP method: A case of the Bou Azzer El Grara inlier, Central Anti-Atlas, Morocco

The Bou Azzer El Grara inlier belongs to the Central Anti-Atlas range, renowned for its mineral deposits. The study carried out on the inlier aims to identify potential areas of cobalt mineralization based on a multi-source methodology of radiometric and remote sensing data processing. The data used includes hydrothermal alteration minerals (argillic, phyllic, propylitic, iron oxides) mapped with Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER) imagery using band ratio and principal component analysis (PCA) techniques, potassium enrichment zones through K/eTh and K/eU ratios calculated from radiometric data, as well as geological data (serpentinite outcrops and geological faults). The analytic hierarchy process (AHP) was used to combine the various thematic layers. Validation was carried out through the field studies in several anomalous zones. The mineral prospectivity map generated highlights nine major zones that appear the most favorable for mineralization, some of which coincide with zones in production and others requiring further exploration.

Denoising of Geochemical Data using Deep Learning–Implications for Regional Surveys

Regional geochemical surveys generate large amounts of data that can be used for a number of purposes such as to guide mineral exploration. Modern surveys are typically designed to permit quantification of data uncertainty through data quality metrics by using quality assurance and quality control (QA/QC) methods. However, these metrics, such as data accuracy and precision, are obtained through the data generation phase. Consequently, it is unclear how residual uncertainty in geochemical data can be minimized (denoised). This is a limitation to propagating uncertainty through downstream activities, particularly through complex models, which can result from the usage of artificial intelligence-based methods. This study aims to develop a deep learning-based method to examine and quantify uncertainty contained in geochemical survey data. Specifically, we demonstrate that: (1) autoencoders can reduce or modulate geochemical data uncertainty; (2) a reduction in uncertainty is observable in the spatial domain as a decrease of the nugget; and (3) a clear data reconstruction regime of the autoencoder can be identified that is strongly associated with data denoising, as opposed to the removal of useful events in data, such as meaningful geochemical anomalies. Our method to post-hoc denoising of geochemical data using deep learning is simple, clear and consistent, with the amount of denoising guided by highly interpretable metrics and existing frameworks of scientific data quality. Consequently, variably denoised data, as well as the original data, could be fed into a single downstream workflow (e.g., mapping, general data analysis or mineral prospectivity mapping), and the differences in the outcome can be subsequently quantified to propagate data uncertainty.

Quantifying uncertainty and improving prospectivity mapping in mineral belts using transfer learning and Random Forest: A case study of copper mineralization in the Superior Craton Province, Quebec, Canada

Mineral prospectivity mapping (MPM) involves identifying locations with a higher potential for mineral exploration based on a set of explanatory variables. In cases where there is a scarcity or absence of unfavorable sites that adequately represent the geological context for deposit discovery, generating synthetic negative data sets becomes necessary to employ a machine learning algorithm optimally. Moreover, when favorable sites are insufficient for deposit discovery within a geological zone, machine learning methods can potentially result in large and highly uncertain prospecting areas. This article proposed a concept based on transfer learning by applying the knowledge gained from mineral belt signatures in different geological zones to a related area. The positive training data were taken from five mineral belts distanced from each other, while the negative data were sampled using geological constraints based on the distance to occurrences and spatial associativity. The results demonstrate that transfer learning, combined with geological constraints applied to the creation of negative datasets, improves model performance and prediction of known deposits while significantly reducing uncertainties. Mineral prospectivity models for predicting potential copper formations were generated using data from the Quebec Government's spatial reference geomining information system, SIGEOM. The case study for this work focused on the geological province of the Superior Craton, which encompasses the vast majority of northeastern Quebec.

Machine Learning-Based Uranium Prospectivity Mapping and Model Explainability Research

Sandstone-hosted uranium deposits are indeed significant sources of uranium resources globally. They are typically found in sedimentary basins and have been extensively explored and exploited in various countries. They play a significant role in meeting global uranium demand and are considered important resources for nuclear energy production. Erlian Basin, as one of the sedimentary basins in northern China, is known for its uranium mineralization hosted within sandstone formations. In this research, machine learning (ML) methodology was applied to mineral prospectivity mapping (MPM) of the metallogenic zone in the Manite depression of the Erlian Basin. An ML model of 92% accuracy was implemented with the random forest algorithm. Additionally, the confusion matrix and receiver operating characteristic curve were used as model evaluation indicators. Furthermore, the model explainability research with post hoc interpretability algorithms bridged the gap between complex opaque (black-box) models and geological cognition, enabling the effective and responsible use of AI technologies. The MPM results shown in QGIS provided vivid geological insights for ML-based metallogenic prediction. With the favorable prospective targets delineated, geologists can make decisions for further uranium exploration.

Three-dimensional mineral prospectivity mapping Considering structural restoration in the Dayingezhuang gold Deposit, eastern china

In the era of big data and the rapid advancements in machine learning, particularly deep learning, three-dimensional mineral prospectivity mapping (3D MPM) has emerged as a powerful tool for accurately deciphering the complex nonlinear relationships between ore-controlling factors hidden within geoscientific datasets and known mineralization. However, post-mineralization structural deformations can significantly distort the structures and mineralization distribution. This distortion introduces a substantial amount of error into 3D MPM, thereby diminishing its effectiveness in mineral resource exploration. In response to this challenge, our study introduces an innovative 3D MPM approach incorporating structural restoration. This approach has been applied to the Dayingezhuang gold deposit, where the Zhaoping fault, experienced disruption due to the post-mineralization NW-trending Dayingezhuang fault. The existing 3D model of the Zhaoping fault was first constructed using the observation data. A 3D geometric transformation technique was then employed to restore the original structures and mineralization distribution during the mineralization period. Subsequently, morphological analysis and prospectivity modeling were performed using deep learning algorithms including the Bayesian decomposition model for regression and multilayer perceptron model for classification. The results indicated several crucial findings: (1) structural restoration effectively eliminated the spatial distortions (250 m to 500 m) within the Zhaoping fault and rectified the discontinuities in mineralization zones; (2) the post-mineralization ore-controlling features exhibited smaller dips and undulations, with the main mineralization concentrated in areas transitioning from steep to gentle dips (-10°/100 m to 2°/100 m); (3) the post-restoration regression and classification models showed higher accuracy and reliability compared to the pre-restoration models, with an improvement of approximately 27 % and 10 %, respectively. As per our research findings, structural restoration effectively eliminates distortions in both fault and mineralization induced by post-ore structural deformations. Consequently, this method significantly enhances the reliability of 3D MPM. In the context of structural deformations, it underscores the necessity and effectiveness of structural restoration in conducting 3D MPM on a deposit scale.

ArcMPM: An ArcEngine-Based Software for Mineral Prospectivity Mapping via Artificial Intelligence Algorithms

ArcMPM: An ArcEngine-Based Software for Mineral Prospectivity Mapping via Artificial Intelligence Algorithms

Combining numerical modeling and machine learning to predict mineral prospectivity: A case study from the Fankou Pb–Zn deposit, southern China

Predictive modeling of mineral prospectivity, which integrates diverse evidence of ore-forming processes based on an ore deposit genetic model constitutes a significant task in mineral exploration targeting. Descriptive deposit models suffer from overwhelming dependence on the explicitly mappable characteristics of a given deposit used to simplify the complex mineral system, but this hinders mineral prospectivity modeling. This study proposes a hybrid approach that combines numerical modeling with machine learning to predict mineral prospectivity. This approach first uses the numerical modeling method to provide more insights and better characterize a complex mineral system. These comprise implicit footprints such as singular stress deformation, heat transmission and fluid flux inherent in the transient geodynamics of ore genesis. The approach then applies a machine learning algorithm as a data-fusion tool to integrate the simulated evidence to minimize uncertainty in the predictive modeling of mineral prospectivity, which fundamentally results from an incomplete understanding of the targeted mineral system. Finally, we use the Fankou Mississippi Valley type Pb–Zn deposit as an example to illustrate the application of the proposed hybrid method. The prospectivity modeling result is critically compared with that of the spatial modeling based on descriptive deposit modeling by means of measures of accuracy, area under the receiver operating characteristic curve, and normalized density of prediction–area plot. It is shown that the overall performance of the prospectivity map predicted using a hybrid method is significantly improved. As such, this study demonstrates that a hybrid framework combining numerical modeling with machine learning provides a useful and novel way of predicting mineral prospectivity. The proposed hybrid method embodies a pioneering paradigm that integrates physics and data, assimilating ore deposit genesis models and reconnaissance data for the prediction of mineral prospectivity in support of exploration targeting. This approach can be implemented to enhance exploration information systems through the predictive modeling of mineral prospectivity in a unified computing and decision-making environment.

Mineral prospecting mapping with conditional generative adversarial network augmented data

Mineral Prospectivity Mapping (MPM) plays a pivotal role in identifying geo-anomalies that are indicative of potential mineralization, drawing upon various geological, geophysical, geochemical, and remote sensing data. With the increasing availability of such data in recent years, data-driven MPM methods have proven effective in discovering new mineral deposits, especially when utilizing machine learning techniques to unveil complex relationships between exploration data and mineral occurrences. Deep learning, specifically Convolutional Neural Networks (CNN), has demonstrated its superiority in this regard. However, these models encounter challenges due to the limited and imbalanced nature of geological exploration data. In this study, we address these challenges by proposing the adoption of a Conditional Generative Adversarial Network (cGAN) strategy for data augmentation. Additionally, we employ the sliding window algorithm for comparative analysis, and CNNs are utilized to assess the effectiveness of various data augmentation strategies. The results indicate that the cGAN-based data augmentation strategy exhibits higher resistance to overfitting, a critical concern in MPM applications. Furthermore, this research successfully delineated a prospectivity map for gold deposits in the Hezuo-Meiwu district, Gansu, China, providing valuable insights for future exploration efforts.

Geochemical prospectivity mapping using compositional balance analysis and multifractal modeling: A case study in the Jinshuikou area, Qinghai, China

To effectively map the prospecting targets related to deposit-type sought, local indicators of spatial association (LISA), compositional balance analysis (CoBA), and multifractal modeling were applied to process stream sediment geochemical data that are collected from the Jinshuikou area, Qinghai province, China. In this study, 14 elements, including Zn, Cu, Ni, Mo, Co, V, Ti, Sb, Ag, Pb, Sn, Hg, Cd, and W, were analyzed in 493 samples. The result of LISA shows that the ore-forming elements Zn and Cu have a close spatial autocorrelation, which supports the construction of balance b1 related to the Zn-Cu mineralization assemblage. A traditional contrast additive method was applied to compare with balance b1 for geochemical anomaly identification. The results clearly indicate that balance b1 significantly assisted with not only reduction of geochemical data internal imbalance caused by extreme value, but also intensification of Zn-Cu assemblage geochemical anomalies associated with Zn polymetallic mineralization. The high anomalies of the two balances, b5 (Co-V-Ti) and b8 (Sn-Pb-Ag-Sb), reflect well the distribution of basic and intermediate-acid rocks in the area, respectively. The geochemical distribution pattern of the balance b1 can be depicted by multifractal parameters such as singular index α, as well as asymmetry index R, which are useful for describing the spatial heterogeneity of the balance b1. The singularity indices of balance b1 were classified by the concentration-area (C-A) fractal model to produce the high potential areas of the Zn-Cu assemblage. The high potential areas were obviously correlated with all Zn polymetallic occurrences. So, combined with the metallogenic geological characteristics of the study area, five prospective prediction targets were mapped. The anomaly verification demonstrates that the Zn concentration has reached the lowest industrial grade and cut-off grade corresponding to the target area I-1 and I-2, respectively, and Cu also has certain metallogenic potential. There is good potential for Zn-Cu mineralization as represented by this integrated method.

Predictive Mapping of Prospectivity for Gold in the Central Portion of the Tapajós Mineral Province, Brazil

This work aims to model mineral prospectivity for intrusion–related gold deposits in the central portion of the Tapajós Mineral Province (TMP), southwestern Pará state. The scope includes experimentation and evaluation of knowledge and data-driven methods applied to multisource data to predict potential targets for gold mineralization. The radiometric data processing allowed to identify a hydrothermal alteration footprint of known gold deposits, providing information in regions with little or no field data available. The aeromagnetic data analysis prompted the identification of high magnetic zones, which are probably related to hydrothermal fluid transport. Linear features extracted from digital elevation data revealed an NNW–SSE general trend, which is consistent with the main structural control of deposits. The data were integrated through three modeling techniques—fuzzy logic (knowledge-driven), weights of evidence (WofE, data-driven), and a machine learning algorithm (SVM, data-driven)—resulting in three prospective models. In all models, the majority of indicated prospective regions coincide with the known deposits. The results obtained in the models were combined to generate an agreement map, which mapped the overlapping of their highest prospective scores, indicating new areas of prospective interest in the central portion of the TMP.

3D Mineral Prospectivity Mapping from 3D Geological Models Using Return–Risk Analysis and Machine Learning on Imbalance Data

Three-dimensional Mineral Prospectivity Mapping (3DMPM) is an innovative approach to mineral exploration that combines multiple geological data sources to create a three-dimensional (3D) model of a mineral deposit. It provides an accurate representation of the subsurface that can be used to identify areas with mineral potential. These 3D geological models are the typical data source for 3D prospective modeling. Geological data sets from multiple sources are used to construct 3D geological models. Since in practice there is a significant imbalance in the ratio of mineralized to non-mineralized classes, the classification results will be biased in favor of the more observed classes. Borderline-SMOTE (BLSMOTE) is an oversampling technique used to solve the problem of unbalanced datasets and works by generating synthetic data points along the boundary line between the minority and majority classes. This helps to create a more balanced dataset without introducing too much noise. Non-mineralized samples can be generated by randomly selecting non-mineralized locations, which means that uncertainties are generated. In this paper, we take the shallow-forming low-temperature hydrothermal deposit Guizhou Lannigou gold deposit as an example to extract the ore-controlling elements and establish a 3D geological model. A total of 50 training samples are generated using the sampling method described above, and 50 mineralization prospects are generated using Random Forests. A return–risk analysis was used to explore the uncertainties associated with synthetic positive samples and randomly selected negative samples, and to determine the final mineral potential values. Based on the evaluation metrics G-mean and F-value, the model using BLSMOTE outperforms the model without the synthetic algorithm and the models using SMOTE and KMeansSMOTE. The optimal model BLSMOTE18 has an AUC of 0.9288. The methodology also performs superiorly with different levels of class imbalance datasets. Excluding the predictions where the results highly overlap with known deposits, five target zones were circled for the targets using a P-A plot, all of which have obvious metallogenic geological features. Among them, Target1 and Target2 have good potential for mineralization, which is of great significance for future mineral exploration work.

A Framework for Data-Driven Mineral Prospectivity Mapping with Interpretable Machine Learning and Modulated Predictive Modeling

A Framework for Data-Driven Mineral Prospectivity Mapping with Interpretable Machine Learning and Modulated Predictive Modeling

Integration of ASTER and Soil Survey Data by Principal Components Analysis and One-Class Support Vector Machine for Mineral Prospectivity Mapping in Kerkasha, Southwestern Eritrea

Integration of ASTER and Soil Survey Data by Principal Components Analysis and One-Class Support Vector Machine for Mineral Prospectivity Mapping in Kerkasha, Southwestern Eritrea

A Review on Global Cooperation Network in the Interdisciplinary Research of Geochemistry Combined with Artificial Intelligence

With the rapid development of modern geochemical analysis techniques, massive volumes of data are being generated from various sources and forms, and geochemical data acquisition and analysis have become important tools for studying geochemical processes and environmental changes. However, geochemical data have high-dimensional, nonlinear characteristics, and traditional geochemical data analysis methods have struggled to meet the demands of modern science. Nowadays, the development of big data and artificial intelligence technologies has provided new ideas and methods for geochemical data analysis. However, geochemical research involves numerous fields such as petrology, ore deposit, mineralogy, and others, each with its specific research methods and objectives, making it difficult to strike a balance between depth and breadth of investigation. Additionally, due to limitations in data sources and collection methods, existing studies often focus on a specific discipline or issue, lacking a comprehensive understanding of the bigger picture and foresight for the future. To assist geochemists in identifying research hotspots in the field and exploring solutions to the aforementioned issues, this article comprehensively reviews related studies in recent years, elaborates on the necessity and challenges of combining geochemistry and artificial intelligence, and analyzes the characteristics and research hotspots of the global collaboration network in this field. The study reveals that the investigation into artificial intelligence techniques to address geochemical issues is progressing swiftly. Joint research papers serve as the primary means of contact within a worldwide collaborative network. The primary areas of focus in the ongoing research on the integration of geochemistry and artificial intelligence include methodologies for analyzing geochemical data, environmental modifications, and mineral prospectivity mapping. Geochemical data analysis is currently a significant focus of research, encompassing a range of methods including machine learning and deep learning. Predicting mineral resources for deep space, deep Earth, and deep sea is also a pressing topic in contemporary research. This paper explores the factors driving research interest and future trends, identifies current research challenges, and considers opportunities for future research.

Auto encoder generative adversarial networks - based mineral prospectivity mapping in Lhasa area, Tibet

When conducting Mineral Potential Mapping (MPM) using multiple sources of data such as geology, geochemistry, and geophysics, it often encounters the challenges of complex and diverse data distributions within these datasets. Enhancing the capability to extract nonlinear data features and further uncover metallogenic information is a crucial research objective. This study utilizes the C-A multifractal approach to extract anomalous information related to metallogenic elements, employs compositional data analysis methods for quantitatively extracting geochemical associations of mineralization, and utilizes GIS spatial analysis techniques to quantitatively extract predictive indicators from various data sources, including geology, geophysics, and remote sensing, to construct an MPM prediction dataset. Building upon the foundation of AutoEncoder (AE), this study introduces a discriminator and employs the AEGAN (Auto Encoder Generative Adversarial Network) algorithm, which combines AutoEncoder and Generative Adversarial Network (GAN), for metallogenic prospectivity prediction in the Lhasa region. Compared to AE algorithms, AEGAN combines the strengths of AE and GAN, significantly improving the model's ability to reconstruct input data through the interaction between the generator and discriminator. Additionally, this study designs comparative experiments with AE, and the results demonstrate that the AEGAN model can more accurately identify the correlation between high anomaly areas and polymetallic deposits, providing a more precise delineation of anomalous extents. The Area Under the Receiver Operating Characteristic Curve (AUC) further validates the superior performance of the AEGAN model. These findings indicate that the AEGAN model exhibits outstanding capabilities in learning the internal connections and features among multiple data sources, holding significant potential for practical applications in mineral exploration.

Mineral prospectivity modelling to delineate potential areas for gold deposits: a case study of Lupa Goldfield, South West Tanzania

ABSTRACT This study used the weight of evidence model to delineate prospective areas for gold deposits in the Lupa Goldfield. The method establishes a spatial relationship between known mineral deposits and evidential maps to create a mineral prospectivity map. Four evidential maps derived from geochemical, geological, geophysical, and topographical Digital Elevation Model (DEM) datasets were analysed and integrated to create a mineral prospectivity map. Results have revealed five classes ranked from the highest, high, medium, low, and lowest favourability patterns with probability values of 0.051, 0.0426, 0.0225, 0.0172, and 0.0091, respectively. The highest favourable areas have the best gold potentials based on the presence of predictor patterns from all four evidential maps. The posterior probability map revealed good prediction, existing gold deposits, such as Saza Mine and Shanta Gold Mine, were mapped. Results have shown that the weight of evidence model was successful and can be applied in mineral exploration targeting.

Mineral Prospectivity Mapping for Orogenic Gold Mineralization in the Rainy River Area, Wabigoon Subprovince

Random Forest classification was applied to create mineral prospectivity maps (MPM) for orogenic gold in the Rainy River area of Ontario, Canada. Geological and geophysical data were used to create 36 predictive maps as RF algorithm input. Eighty-three (83) orogenic gold prospects/occurrences were used to train the classifier, and 33 occurrences were used to validate the model. The non-Au (negative) points were randomly selected with or without spatial restriction. The prospectivity mapping results show high performance for the training and test data in area-frequency curves. The F1 accuracy is high and moderate when assessed with the training and test data, respectively. The mean decrease accuracy was applied to calculate the variable importance. Density, proximity to lithological contacts, mafic to intermediate volcanics, analytic signal, and proximity to the Cameron-Pipestone deformation zone exhibit the highest variable importance in both models. The main difference between the models is in the uncertainty maps, in which the high-potential areas show lower uncertainty in the maps created with spatial restriction when selecting the negative points.