Zircon is a common hardy mineral found in rocks that are 4 billion years old. Their structure and texture may reflect the conditions in which they formed, earning them the reputation of “nature’s time capsules.” Using the power of machine learning, scientists can mine zircon textures to identify valuable mineral deposits, according to a new study.
In a new study, Nathwani et al. A method has been developed to distinguish small differences between zircon grains formed in copper-associated rocks and granites. Their method could help scientists search for mineral deposits and determine the origin of various sediments.
The researchers used a machine learning tool called convolutional neural network (CNN), which specializes in image analysis. Using samples collected in southern Peru, where most of the world’s copper is produced, they found that CNN could identify the unique shapes and textures of zircons found near copper deposits. The model was also able to distinguish these copper-related zircons from zircons found in other rock types in the area with 85 percent success.
Copper has a wide range of industrial uses, from electronics to construction, and research shows that combining machine learning with more traditional methods can make it easier to explore and identify copper deposits. (Journal of Geophysical Research: Solid Earth, https://doi.org/10.1029/2022JB025933, 2023)
Hot Research Topics Geophysical Research Letters “Neural Network Tidal Maps of a Small Pond” Sarah Derouen
Editorial Reviews Social Science “Collaboration helps overcome challenges in air quality monitoring”, Muki Haklay
VOX Geophysics Editorial “What We Know and Don’t Know about Critical Climate Drivers” by Seaver Wang
Post time: Oct-30-2023