Ahmad Reza Mokhtari, Pouran Behnia, Bruno Lafrance, Mostafa Naghizadeh, Jack M. Simmons, Jeff R. Harris
{"title":"Mineral prospectivity mapping of orogenic gold mineralization in the Malartic-Val-d’Or Transect area, Metal Earth project, Canada","authors":"Ahmad Reza Mokhtari, Pouran Behnia, Bruno Lafrance, Mostafa Naghizadeh, Jack M. Simmons, Jeff R. Harris","doi":"10.1016/j.oregeorev.2025.106498","DOIUrl":null,"url":null,"abstract":"<div><div>Mineral Prospectivity Mapping has been applied to define exploration targets for orogenic gold mineralization in the world-class Malartic-Val-d’Or area (Quebec) of the Abitibi greenstone belt, a region that contributes significantly to Canada’s annual gold production. This research utilizes lithological, geophysical, and structural data from the Malartic transect, collected as part of the Metal Earth project, to investigate factors controlling gold distribution in the region. Stepwise logistic regression and random forest algorithms were employed to map mineral prospectivity for gold potential. The overall accuracy indicates that the random forest method has outperformed logistic regression, although the latter produced an acceptable model. This conclusion is supported by the classification accuracy validated using an independent Au occurrence database and the performance metrics generated by the random forest and logistic regression models. Crustal density, shear zones, and faults are strong predictors for distinguishing mineralized and non-mineralized locations, as identified by both Random Forest and Logistic Regression models. In the Malartic-Val-d’Or region, subvertical conductive anomalies revealed by magnetotelluric data align closely with major shear zones and crustal-scale structures, such as the Cadillac-Larder Lake deformation zone, which hosts significant gold deposits. These anomalies suggest a paleo-hydrothermal footprint of mineralizing fluids and highlight their critical role in the formation of orogenic gold deposits, a pattern also observed in other mining camps across the Superior craton. These results and associated mineral prospectivity maps are integral for greenfields exploration in the Malartic region and may offer valuable insights for mineral exploration in other greenstone belts of the Superior craton.</div></div>","PeriodicalId":19644,"journal":{"name":"Ore Geology Reviews","volume":"178 ","pages":"Article 106498"},"PeriodicalIF":3.6000,"publicationDate":"2025-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ore Geology Reviews","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0169136825000587","RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/2/13 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"GEOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Mineral Prospectivity Mapping has been applied to define exploration targets for orogenic gold mineralization in the world-class Malartic-Val-d’Or area (Quebec) of the Abitibi greenstone belt, a region that contributes significantly to Canada’s annual gold production. This research utilizes lithological, geophysical, and structural data from the Malartic transect, collected as part of the Metal Earth project, to investigate factors controlling gold distribution in the region. Stepwise logistic regression and random forest algorithms were employed to map mineral prospectivity for gold potential. The overall accuracy indicates that the random forest method has outperformed logistic regression, although the latter produced an acceptable model. This conclusion is supported by the classification accuracy validated using an independent Au occurrence database and the performance metrics generated by the random forest and logistic regression models. Crustal density, shear zones, and faults are strong predictors for distinguishing mineralized and non-mineralized locations, as identified by both Random Forest and Logistic Regression models. In the Malartic-Val-d’Or region, subvertical conductive anomalies revealed by magnetotelluric data align closely with major shear zones and crustal-scale structures, such as the Cadillac-Larder Lake deformation zone, which hosts significant gold deposits. These anomalies suggest a paleo-hydrothermal footprint of mineralizing fluids and highlight their critical role in the formation of orogenic gold deposits, a pattern also observed in other mining camps across the Superior craton. These results and associated mineral prospectivity maps are integral for greenfields exploration in the Malartic region and may offer valuable insights for mineral exploration in other greenstone belts of the Superior craton.
在世界级的malartici - val -d 'Or地区(魁北克)的Abitibi绿岩带,该地区对加拿大的年度黄金产量贡献巨大,已应用矿产勘探测绘来确定造山带金矿化的勘探目标。本研究利用来自Malartic样带的岩性、地球物理和构造数据(作为Metal Earth项目的一部分)来研究控制该地区黄金分布的因素。采用逐步逻辑回归和随机森林算法绘制金矿远景图。总体精度表明,随机森林方法优于逻辑回归,尽管后者产生了一个可接受的模型。这一结论得到了独立金产率数据库验证的分类精度以及随机森林和逻辑回归模型生成的性能指标的支持。通过随机森林和逻辑回归模型,地壳密度、剪切带和断层是区分矿化和非矿化位置的有力预测因素。在malartici - val -d 'Or地区,大地电磁资料显示的亚垂直导电异常与主要剪切带和地壳尺度构造(如Cadillac-Larder湖变形带)密切相关,这些构造带具有重要的金矿床。这些异常表明了成矿流体的古热液足迹,并突出了它们在造山带金矿床形成中的关键作用,这种模式也在整个上必胜克拉通的其他采矿营地中观察到。这些结果和相关的矿产找矿图对马拉蒂地区的绿地勘查具有重要意义,并可能为上克拉通其他绿岩带的矿产勘查提供有价值的参考。
期刊介绍:
Ore Geology Reviews aims to familiarize all earth scientists with recent advances in a number of interconnected disciplines related to the study of, and search for, ore deposits. The reviews range from brief to longer contributions, but the journal preferentially publishes manuscripts that fill the niche between the commonly shorter journal articles and the comprehensive book coverages, and thus has a special appeal to many authors and readers.