Scientists are testing machine learning algorithms to predict deposits on Earth and other planets.
Researchers from the Carnegie Institution for Science in Washington, D.C., and other universities across the U.S. have developed an algorithm that predicts the location of minerals on Earth utilizing patterns in mineral associations. The hence-called mineral association analysis could prove useful to locate specific critical raw materials that would have otherwise only been possible to find through “accumulated experience in the field and laboratory” or “implemented by perseverance and luck,” according to the research team.
In a paper published in the open-access journal Proceedings of the National Academy of Sciences (PNAS) Nexus, the group explains their tool which uses data from the Mineral Evolution Database, consisting of 295,583 mineral localities of 5,478 mineral species. By utilizing association rules, a method that is already a common instrument in advertisement and marketing fields, the tool was able to successfully predict the location of uranium minerals. Additionally, the model also identified areas with rare earth element (REE) and lithium containing minerals. While the tool was tested on Earth, the researchers emphasize that it could also be used on extraterrestrial objects such as planets or asteroids.
The new model could join existing methods and form an arsenal to locate critical minerals outside of our home planet. For example, just in April, terbium was discovered in the atmosphere of an exoplanet also using a new tool. However, the fields of use on Earth are not to be despised, as the supply of REEs and other critical minerals like lithium are heavily concentrated in individual countries like China. Because these raw materials are essential components in electronic devices and green energy infrastructure and technologies, the dependency on imports could present issues in the context of geopolitical tensions and American policymakers are seeking alternatives. The mineral association analysis demonstrated the ability to find new deposits of critical minerals in the United States, according to the researchers, and could hence help to limit the dependency on imports.