Editors’ Highlights are summaries of recent papers by AGU journal editors.
sauce: Geophysical Research Journal: Solid Earth
The thickness of the continental crust changed significantly over geological time as the continents collided and split apart. However, quantifying past crustal thickness and how it evolved during different crustal periods remains a challenging task. Conventional methods rely on geochemical surrogates containing trace element ratios that are susceptible to minerals such as garnet, which are stable only in regions of thick crust. However, such substitutes can break in regions of unusually thick crust, such as continental collisions.
Guo and Yang [2023] We train machine learning algorithms on crustal thickness data obtained from large geochemical datasets and geophysical surveys. They applied this method to several thick crustal regions associated with subduction and continental collisions, such as the Kohistan-Ladakh Arc (Pakistan) and the Talkeetna Arc (Alaska). The authors found that machine learning results overlap well with geochemical proxy data. Machine learning-based models trained on large-scale geochemical and geophysical datasets are promising tools for future studies of crustal evolution.
Quote: Guo, P., Yang, T. (2023). Quantify continental crust thickness using machine learning techniques. Geophysical Research Journal: Solid Earth128, e2022JB025970. https://doi.org/10.1029/2022JB025970
— Emily Chin, Deputy Editor, JGR: Solid Earth
Text © 2023. Author. CC BY-NC-ND 3.0
Images are subject to copyright unless otherwise stated. Reuse without the express permission of the copyright owner is prohibited.
Related
