Skip to main content
A Bayesian Inversion for the Generation of Felsic Archean Crust

A Bayesian Inversion for the Generation of Felsic Archean Crust

This is a Preprint and has not been peer reviewed. This is version 1 of this Preprint.

Add a Comment

You must log in to post a comment.


Comments

There are no comments or no comments have been made public for this article.

Downloads

Download Preprint

Supplementary Files

Authors

Jannitta Yao , C. Brenhin Keller 

Abstract

Archean continental crust is compositionally distinct from post-Archean continental crust; its felsic components predominantly comprise the tonalite, trondhjemite, and granodiorite suite, while its mafic components record evidence of secular mantle cooling. Although the geochemistry of these rocks have been well characterized, the production of Archean felsic magmas remains contested because the early Earth's dominant geodynamic regime is uncertain. Here, a novel integration of parallel Markov chain Monte Carlo methods with the Perple_X thermodynamic forward model inverts for the H2O content and the P–T evolution required of primitive magmas to generate differentiation trends recorded by the Archean igneous rock record. Resulting P–T paths show a period of extended differentiation at 4-5 kbar with a H2O content of ~3.4 wt%. These results are consistent with the depth and water content of mantle-derived mafic magmas in modern arcs, and point towards subduction as a potential tectonic setting for the production of these magmas.

DOI

https://doi.org/10.31223/X5RN49

Subjects

Earth Sciences, Geology

Keywords

Archean, Bayesian statistics, Markov chain Monte Carlo methods, thermodynamic modeling

Dates

Published: 2026-07-02 08:32

Last Updated: 2026-07-02 08:32

License

CC BY Attribution 4.0 International

Additional Metadata

Data Availability:
The inverse model and dataset can be found on GitHub (https://github.com/jannittayao/PerpleX-MCMC-MPI) and is preserved on Zenodo (https://doi.org/10.5281/zenodo.20673031)

Metrics

Views: 36

Downloads: 1