Porosity evolution of mafic crystal mush during reactive flow

This is a Preprint and has not been peer reviewed. The published version of this Preprint is available: https://doi.org/10.1038/s41467-023-38136-x. 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

Authors

Matthew Lloyd Morgan Gleeson , C. Johan Lissenberg, Paula Antoshechkina

Abstract

The emergence of the “mush paradigm” has raised several questions for conventional models of magma storage and extraction: how are melts extracted to form eruptible liquid-rich domains? What mechanism controls melt transport in mush-rich systems? Recently, reactive flow has been proposed as a major contributing factor in the formation of high porosity, melt-rich regions. Yet, owing to the absence of accurate geochemical simulations, the influence of reactive flow on the porosity of natural mush systems remains under-constrained. Here, we use a thermodynamically constrained model of melt-mush reaction to simulate the chemical, mineralogical, and physical consequences of reactive flow in a multi-component mush system. Our results demonstrate that reactive flow within troctolitic to gabbroic mushes can drive large changes in mush porosity. For example, primitive magma recharge drives an increase in the system porosity and could trigger melt channelization or mush destabilization, aiding rapid melt transfer through low-porosity mush reservoirs.

DOI

https://doi.org/10.31223/X5BD45

Subjects

Physical Sciences and Mathematics

Keywords

Mush, Reactive Flow, gabbro, Melt Transport

Dates

Published: 2023-02-19 11:05

Last Updated: 2023-02-19 19:05

License

CC BY Attribution 4.0 International

Additional Metadata

Data Availability (Reason not available):
All code developed in this study is available via the lead-author’s GitHub (https://github.com/gleesonm1/MeltMushRxn) and the current version used in this publication has been archived using Zenodo (https://doi.org/10.5281/zenodo.7626521). The GitHub repository is set up with a Results folder containing MATLAB script that will recreate the figures shown in this manuscript, allowing readers to investigate the results of our models in detail.