Mineral Informatics: Analysis and Visualization of Minerals through Time and Space
Abstract
Minerals provide the most robust, information-rich artifacts of planetary origins and evolution. Each mineral specimen is a time capsule that preserves a record of successive chemical, physical, and ultimately biological environments. If we are to understand the 4.5-billion-year story of Earth and its neighboring planets and moons, then minerals hold the most eloquent testimony of deep time and epic change. Mineral classification is provided by rigorous protocols of the IMA-CNMNC [1]. Each mineral species owes its identity to its unique combination of idealized end-member composition and/or chemical range, plus idealized crystal structure. In this regard, the IMA-CNMNC approach employs the minimum information necessary to unambiguously define each species. More than 5700 species have been approved, while thousands more potential species await discovery and description. However, these criteria do not lend themselves to an exploration of planetary evolution, nor are the idealizations that define IMA species equivalent to mineral natural kinds that represent genuine divisions of nature. We propose a complementary evolutionary system of mineralogy that strives to define mineral natural kinds based on their positions in the evolutionary chronology of a planet, and the process by which they formed. In this system, each mineral natural kind has a distinctive temporal and paragenetic context, as manifest in its unique combination of chemical, structural, physical, and contextual attributes. The evolutionary system is thus data intensive, embracing all of a minerals attributes in identifying natural kinds, for example through cluster analysis [2,3]. Studies of all known minerals and their paragenetic modes suggest that more than 10,000 mineral natural kinds exist, many of which relate to IMA-approved species by lumping and/or splitting criteria [4]. Analytical and visualization methods applied to mineral data reveal striking temporal and spatial trends across the Mineral Kingdom. The bipartite network graph shows 5659 blue nodes representing mineral species linked to 57 green nodes representing different formation processes. [1] Hawthorne et al. (2021) Min. Mag. 85, 125; [2] Hazen & Morrison (2021) Am.Min. 106, in press; [3] Hazen (2019) Ibid. 104, 468; [4] Hazen & Morrison (2022) Ibid. 107, in press.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFMIN13A..01H