GSA Annual Meeting in Phoenix, Arizona, USA - 2019

Paper No. 234-2
Presentation Time: 8:15 AM

MINERAL EVOLUTION DATABASE: DATA-DRIVEN AGE ASSIGNMENT, HOW DOES A MINERAL GET AN AGE?


GOLDEN, Joshua J.1, DOWNS, Robert T.1, HAZEN, Robert M.2, PIRES, Alexander J.1 and RALPH, Jolyon3, (1)Department of Geosciences, University of Arizona, Tucson, AZ 85721, (2)Geophysical Laboratory, Carnegie Institution for Science, Washington, DC 20015, (3)mindat.org, Surrey, CR4 4FD, United Kingdom

The concept of mineral evolution, that Earth’s mineralogy has changed throughout Earth’s history in response to changing atmospheric and crustal conditions, requires large amounts of spatial and temporal data to be collected and correlated in order to test. The Mineral Evolution Database (MED; rruff.info/Evolution/) is designed to host the temporal distribution of each occurrence of the 5,477 unique mineral species approved by the International Mineralogical Association (IMA). Age data has been collected from over 1,600 references as well as pre-existing databases.

Mindat.org is spearheading the effort to record the spatial distribution of mineral species on Earth’s surface based on primary literature and labels from mineral specimens. Their repository now contains nearly 300,000 localities, of which more than half contain minerals resulting in a total of nearly 811,000 mineral occurrence data (mineral-locality pairs).

Spatial and temporal data regarding specific mineral occurrences rarely occur together in the same repository, so it is necessary to establish correlations between the literature and repositories. Ideally, each mineral would be directly dated at each of its occurrences, however, few mineral species can currently be directly dated. Therefore, assigning an age to each mineral necessitates the use of other geologic contextual information and also primary author interpretations. This data is curated and classified to allow an algorithm to assign each mineral an age based on the data’s relevance to localities, minerals, or elements. The spatiotemporal mineral distribution can then be correlated with major geologic events in Earth’s history such as the GOE and supercontinent formation and breakup.

The MED’s age algorithm currently considers 15,906 unique age records which directly reference 6,253 localities, and assigns an age to 23,969 of the 155,112 localities that have minerals listed directly at them on mindat.org, resulting in a total of 194,090 total dated mineral-locality occurrences. At least one age has been calculated for 4,756 of the 5,477 IMA-approved mineral species. The unprocessed age data used to populate the MED can be accessed in the Open Data Repository (https://odr.io/MEDAges).