Harnessing the Power of Artificial Intelligence and Machine Learning in Mineral Exploration—Opportunities and Cautionary Notes

Editor’s note: The aim of the Geology and Mining series is to introduce early-career professionals and students to various aspects of mineral exploration, development, and mining in order to share the experiences and insight of each author on the myriad of topics involved with the mineral industry and the ways in which geoscientists contribute to each. Abstract Artificial intelligence (AI), and machine learning (ML) have emerged in the last few years from relative obscurity in the mineral exploration sector and they now attract significant attention from people in both industry and academia. However, due to the novelty of AI and ML applications, their practical use and potential remain enigmatic to many beyond a relatively few expert practitioners. We introduce this subject for the nonexpert and review some of the current applications and evolving uses. For the most traditionally minded geologist, we argue that ML can be an invaluable new tool, contributing to topics that range from exploratory data analysis to automated core logging and mineral prospectivity mapping, such that it will have a substantial impact on how exploration is conducted in the future. However, ML algorithms perform best with a large amount of homogeneously distributed clean data for a well-constrained objective. For this reason, the application to exploration strategy, especially for optimizing target selection, will be a challenge where data are heterogeneous, multiscale, amorphous, and discontinuous. For the more tech-savvy geologist and data scientist, we provide notes of caution regarding the limitations of ML applied to geoscience data, and reasons to temper expectations. Nonetheless, we project that such technologies, if used in an appropriate manner, will eventually be part of the full range of exploration tasks, allowing explorers to do more with their data in less time. However, whether this will tip the scales in favor of higher discovery rates remains to be demonstrated.

SEG Discovery 127 (October)

This file includes the entire issue in PDF format. The HTML versions of the peer-reviewed articles must be viewed and/or purchased separately.

Battery and Energy Metals: Future Drivers of the Minerals Industry?

A wide range of metals and minerals are currently used in battery and energy technology, meaning that an increasing number of these commodities are being considered as potentially viable primary products by the minerals industry. A select group of these minerals and elements that are vital for energy and battery technologies, including Al, Cr, Co, Cu, graphite, In, Li, Mn, Mo, the rare earth elements (REEs; primarily Dy and Nd), Ni, Ag, Ti, and V, are also likely to undergo rapid increases in demand as a result of the move toward low- and zero-CO2 energy and transportation technology (often termed the energy transition) driven by climate change mitigation and consumer and investor concerns and demands. Increased levels of mineral exploration, discovery, and production will be needed to meet this rising demand. However, several of these key metals and minerals are produced as co- and by-products of other elements. This means that their production is inherently linked to the production of main product elements that may not undergo similar increases in demand, creating issues related to security of supply. It is also not simple to just produce more metal and minerals given the environmental, social, and governmental challenges the global mining industry currently faces. Finally, there are uncertainties over exactly what technologies will dominate the energy transition, meaning that robust demand predictions are still relatively problematic. Quantifying these and other uncertainties and addressing issues over by-and coproduct supply will help ensure that mineral deposits are used sustainably. In addition, understanding the deportment and processing behavior of key critical metals and minerals that are produced as by- or coproducts of main metals such as copper will allow these to actually be extracted from mineral deposits being mined now and into the future rather than be lost to waste. Both of these are vital steps in terms of ensuring that future increases in metal and mineral demand can be met. The impact of these changes on metal and mineral demand and pricing also needs to be examined to ensure the economics of these changes relating to the energy transition are fully understood. All of this means that the mineral industry must act and plan for this transition accordingly in coordination with governments and other organizations. This is especially true given the long lead-in times related to the vast majority of mineral exploration and mining projects compared to the potentially rapid increase in demand for certain battery and energy metals and minerals. This is somewhat analogous to the technology sector, where software (analogous to battery and energy technology) can advance rapidly, creating significant demand that puts pressure on associated hardware (in this case, the development of new mines or changes in mineral processing) that advances more slowly. Failing to ensure mineral and metal supply meets increasing (and potentially rapidly varying) demand may lead to situations where demand far exceeds supply, causing preventable issues related to supply chain continuity and further delaying climate change mitigation, with potential global consequences.

SEG Discovery 126 (July)

This file includes the entire issue in PDF format. The HTML versions of the peer-reviewed articles must be viewed and/or purchased separately.

Weathering of Copper Deposits and Copper Mobility: Mineralogy, Geochemical Stratigraphy, and Exploration Implications

Weathering of pyrite and copper sulfide-bearing rocks produces a consistent series of iron- and copper-bearing minerals that reflect vertical and lateral geochemical changes as supergene solutions react with rock and experience loss of oxidizing capacity. Reactive host rocks, comprising feldspars, mafic minerals, chlorite, and carbonates, buffer pH values that limit copper mineral destruction, thus restricting the supergene transport of copper. Generally, rocks that have undergone well-developed hypogene or supergene hydrolysis of aluminosilicates promote copper mobility because they do not react substantially with low-pH supergene solutions generated by oxidation of pyrite and associated copper sulfides. Development of geochemical stratigraphy is characterized by physical and geochemical parameters that determine the maturity of a supergene profile, with cyclical leaching and enrichment periods critical for the production of economically significant copper accumulation. Evidence for multicycle enrichment is recorded by hematite after chalcocite, hanging zones of copper oxides that replace chalcocite, and hematitic capping overlying immature goethitic-pyritic capping. Because pyrite is the most refractory sulfide with respect to chalcocite replacement, geochemically strong supergene enrichment is independent of total copper added to protore and instead is indicated qualitatively by the degree to which chalcocite replaces pyrite. Covellite replacement of chalcopyrite indicates weak copper addition to protore and generally represents the base and lateral extent of supergene enrichment; covellite replacement of chalcocite indicates incipient copper removal by copper-impoverished supergene solutions. Exploration for, and delineation of, mature supergene enrichment profiles benefits from the recognition of paleoweathering cycles and consequent development of mature geochemical stratigraphy.

Comminution and Mineral Separation—Geological Input to Metallurgy

Editor’s note: The aim of the Geology and Mining series is to introduce early-career professionals and students to various aspects of mineral exploration, development, and mining, in order to share the experiences and insight of each author on the myriad of topics involved with the mineral industry and the ways in which geoscientists contribute to each. Abstract Communication and collaboration during mine development and operation are essential if the maximum value of a mineral deposit is to be realized, since there are many links between the geology and mineralogy of an orebody and the complex task of an effective plant design. This is only achieved when geologists, metallurgists, and mining and environmental engineers jointly assess the results of metallurgical characterization. This requirement is examined here, albeit for only two of the three metallurgical ore-processing activities—comminution and mineral separation. Wealth is not captured (i.e., is destroyed) unless the most efficient and effective methods for comminuting and separating the mineral(s) of value in a deposit are identified. Benchmarking metallurgical test work requirements for the next mine development based solely on past experience does not address the variability that is unique to the mineralogy of each mineral deposit. Metallurgists are now slowly advancing from using a few (so-called) representative samples to assess the processing characteristics of a deposit to applying metallurgical testing to tens, or hundreds, of samples, with the increase in number of samples allowed by technological advances. More still needs to be done. Identifying the characteristics of different mineralization types of a deposit and grouping it into domains are crucially important. These steps simplify processing by separating ore into relatively few (4–6) types with similar expected metallurgical performance. Understanding what metallurgical tests are measuring and how representative the samples and tests are of the orebody domains are essential considerations for a testing program. No knowledge is bad; some is better or more useful than other. Testing for penalty elements (As, Bi, Hg, F, etc.) and, more importantly, for penalty-element minerals allows their effects to be mitigated during design of the processing plant; this should start during the early exploration stage. Continued evolution of orebody knowledge and confidence in processing ores will lead to better performance of the processing plant, thereby reducing investment risk.

Epithermal Precious Metal Deposits in South Korea—History and Pursuit

The gold and silver endowment of Korea has historically been well known, with records alluding to production as far back as 1122 BC. The main gold production period was from 1925 to 1943 during the Japanese occupation of Korea, with more than 1 Moz recorded in 1939. Muguk was the most productive gold mining operation, located within the central region of South Korea, with a recorded 590 koz of gold produced from 1934 to 1998 (first mined in AD 912). The majority of the historical mining operations were closed by government order in 1943 during the Second World War and never reopened. A number of small mines operated between 1971 and 1998, with limited production during a period of gold prices generally lower than at present (~25–50% of current inflation adjusted prices, apart from a four-year period 1979–83). It is likely that significant resources remain within these historical mining areas. Gold-silver deposit types historically recognized and exploited in Korea include placers and orogenic and intrusion-related vein systems. Only more recently have epithermal vein and breccia systems been recognized. This is not surprising, given that the geologic and tectonic setting of the Southern Korean peninsula is prospective for epithermal precious metal deposits, spatially associated with basin-scale brittle fault systems in Cretaceous volcanic terranes. South Korea is an underexplored jurisdiction, with limited modern exploration and drilling until the mid-1990s, when Ivanhoe Mines Ltd. discovered the Gasado, Eunsan, and Moisan epithermal gold-silver deposits, all of which became mines. Exploration was limited for another 20 years until Southern Gold Ltd., an Australian Securities Exchange (ASX)-listed company, commenced regional-scale exploration for epithermal deposits, using a strategy similar to that successfully employed by Ivanhoe.

Geology and Mining: Mineral Resources and Reserves: Their Estimation, Use, and Abuse

Editor’s note: The Geology and Mining series, edited by Dan Wood and Jeffrey Hedenquist, is designed to introduce early-career professionals and students to a variety of topics in mineral exploration, development, and mining, in order to provide insight into the many ways in which geoscientists contribute to the mineral industry. Abstract Resource and reserve estimation is a critical step in mine development and the progression from mineral exploration to commodity production. The data inputs typically change over time and reflect variations in geoscientific knowledge as well as the modifying factors required by regulation for estimating a reserve. These factors include mineral (ore) processing, metallurgical treatment of the ore, infrastructure requirements for mine and workforce, and the transportation of processed products to buyers; others that will affect the production of metals and/or minerals from a deposit include economic, marketing, legal, environmental, social, and governmental factors. All are needed by the mining industry to quantify the contained mineralization within mineral deposits that likely warrant the significant capital investment required to build a mine. However, these resource and reserve data are estimates that change over time due to unpredicted variations in the initial inputs. Paramount to the two estimates are the quality and accuracy of the geologic inputs and the communication of these to the professionals tasked with making each estimate. Geostatistical processing of the grade of the resource has become a dominant element of the estimation process, but this requires transparent and informed communication between geologists and mining engineers with the geostatistician responsible for mathematically processing the grade data. Regulatory constraints also mean that estimated resources and reserves seldom capture the full extent of a mineral deposit. Similarly, co- and by-product metals and minerals that are commonly produced by mines may not be captured by resource and reserve estimates because of their limited economic contribution. This suggests that reporting standards for co- and by-products—particularly for the critical metals that may have a sharp increase in demand—need improvement. Finally, the importance of these data to the mining industry is such that informing investors and the broader public about the nature of resource and reserve estimates, and the meaning of associated terminology, is also essential when considering the global metal and mineral supply, and the role of mining in modern society.

SEG Discovery 125 (April)

This file includes the entire issue in PDF format. The HTML versions of the peer-reviewed articles must be viewed and/or purchased separately.

Mesozoic Metallogenesis of Peru: A Reality Check on Geodynamic Models

The Andean Cordillera is generally regarded as the product of easterly subduction of oceanic lithosphere below South America since the Late Triassic, but recent syntheses have challenged this paradigm. In one model, W-dipping oceanic subduction pulls the continent west until it collides with a ribbon continent that now forms the coastal region and Western Cordillera of the Peruvian Andes. A second model involves westerly oceanic subduction until 120 to 100 Ma, without the involvement of a ribbon continent, to explain deep, subducted slabs revealed by mantle tomographic images. Both assume that “Andean-style” E-dipping subduction did not exist during the Jurassic and Early Cretaceous. Another model, also involving mantle tomography, assumes that a back-arc basin opened inboard of the trench between 145 and 100 Ma, displacing the E-dipping subduction zone offshore without changing its polarity. This article examines the implications of these hypotheses for southern Peruvian metallogenesis during the Mesozoic, when marginal basins opened and closed and were thrust eastward and then were intruded, between 110 and ~50 Ma, by a linear belt of multiple plutons known as the Coastal Batholith. The earliest mineralization in southern Peru is located on the coast and comprises major iron oxide and minor porphyry copper deposits emplaced between 180 and 110 Ma. This was followed by Cu-rich iron oxide copper-gold deposits and a large Zn-rich volcanogenic massive sulfide (VMS) deposit between 115 and 95 Ma, then minor porphyry Cu deposits at ~80 Ma. A second episode of localized VMS mineralization followed at 70 to 68 Ma, then a group of at least five giant porphyry Cu-Mo deposits in southernmost Peru formed between 62 and 53 Ma. The conventional model of Andean-style subduction, which explains many features of Mesozoic Andean metallogenesis in terms of changing plate vectors and velocities, is a poor fit with mantle tomographic anomalies that are thought to record the paleopositions of ancient trenches. A ribbon-continent model requires some plutons of the Coastal Batholith to have been separated from others by an ocean basin. West-dipping oceanic subduction does not account for Jurassic mineralization and magmatism in southern Peru. A model involving a back-arc basin that opened inboard of the existing trench, forcing E-dipping subduction to retreat offshore between 145 and 100 Ma, seems to best explain the metallogenic and tomographic data.