GSA Connects 2023 Meeting in Pittsburgh, Pennsylvania

Paper No. 141-1
Presentation Time: 8:05 AM

AN EVOLUTIONARY SYSTEM OF MINERALOGY: AN INFORMATICS APPROACH TO MINERAL CLASSIFICATION (Invited Presentation)


HAZEN, Robert, MORRISON, Shaunna M. and PRABHU, Anirudh, Earth and Planets Laboratory, Carnegie Institution for Science, 5251 Broad Branch Road NW, Washington, DC 20015

The story of Earth is a 4.5-billion-year saga of dramatic transformations, driven by physical, chemical, and—based on a fascinating growing body of evidence—biological processes. The co-evolution of life and rocks unfolds in an irreversible sequence of evolutionary stages. Each stage re-sculpted our planet’s surface, while introducing new planetary processes and phenomena. This grand and intertwined tale of Earth’s living and non-living spheres is coming into ever-sharper focus. Sequential changes of terrestrial planets and moons are best preserved in their rich mineral record. “Mineral evolution,” the study of our planet’s diversifying near-surface environment, began with a score of different mineral species that formed in the cooling envelopes of exploding stars. Dust and gas from those stars clumped together to form our stellar nebula, the nebula formed the Sun and countless planetesimals, and alteration of planetesimals by water and heat resulted in the 300 minerals found today in meteorites that fall to Earth. Earth’s evolution progressed by a sequence of chemical and physical processes, which ultimately led to the origin-of-life. Once life emerged, mineralogy and biology co-evolved, as changes in the chemistry of oceans, the atmosphere, and the crust dramatically increased Earth’s mineral diversity to the more than 5900 species known today.

To capture the sweep of Earth’s changing mineralogy, we have developed an “evolutionary system” of mineral classification that is based on age and formation process, as well as the traditional attributes of idealized composition and structure. Large and growing open-access mineral data resources facilitate the analysis, visualization, and prediction of mineral systems. Accordingly, we employ association analysis, network analysis, cluster analysis, community detection, and other methods to characterize and classify minerals. This informatics approach ties a planet’s mineralogy to changing environmental conditions, including the origins and evolution of life.