Involvement of Slab‐Derived Fluid in the Generation of Cenozoic Basalts in Northeast China Inferred From Machine Learning

The origin and involvement of fluid in the generation of Cenozoic basalts in Northeast China are still under debate. Here we apply the machine learning methods of random forest and deep neural network to train models using data sets of global island arc and ocean island basalts. The trained models p...

Full description

Saved in:
Bibliographic Details
Published in:Geophysical research letters Vol. 46; no. 10; pp. 5234 - 5242
Main Authors: Zhao, Yong, Zhang, Yigang, Geng, Ming, Jiang, Jilian, Zou, Xinyu
Format: Journal Article
Language:English
Published: Washington John Wiley & Sons, Inc 28.05.2019
Subjects:
Artificial intelligence
Artificial neural networks
Basalt
Cenozoic
Chemical composition
Island arcs
Isotopes
Learning algorithms
Machine learning
Magma
Magnesium
Magnesium isotopes
Neural networks
Ocean models
Organic chemistry
Ratios
Seismic tomography
Subduction (geology)
Tomography
Volcanic eruptions
Volcanoes
China
ISSN:0094-8276, 1944-8007
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The origin and involvement of fluid in the generation of Cenozoic basalts in Northeast China are still under debate. Here we apply the machine learning methods of random forest and deep neural network to train models using data sets of global island arc and ocean island basalts. The trained models predict that most Cenozoic basalts in Northeast China are influenced by fluid and that the fluid activity decreases from east to west. The boundary defined by fluid activity coincides with the westernmost edge of the present‐day stagnant Pacific slab determined by seismic tomography and with the geochemical boundary defined by magnesium isotopes. These observations support the view that the fluid involved in the generation of the basalts is controlled by the stagnant Pacific slab instead of driven by the plume induced by the sinking Izanagi Plate. Plain Language Summary Many volcanoes are erupted in Northeast China, and it is of great interest to study the origin of these volcanoes. Previously, the chemical compositions of these volcanoes and typically elemental ratios such as Ba/Th are used to find the answer. In the present study, modern machine learning methods called random forest and deep neural network are used to do the work. The advantage of the new methods is that they can obtain a whole picture of the chemical compositional data instead of a particular one from an elemental ratio. The new methods find that the generation of these basalts is closely related to the Pacific slab, subducting downward at Japan, reaching ~600‐km depth at the border of eastern China, and extending horizontally to the Mongolia border. Materials released from the slab, such as fluid and melt and the elements dissolved in them, move upward and trigger the volcanoes we see on the surface. The boundary between volcanoes affected dominantly by the slab‐derived fluid and those that are not coincides with the westernmost edge of the deeply buried Pacific slab. These findings deepen our understanding of the generation of these volcanoes. Key Points Global geochemical data of island arc and ocean island basalts are used to train machine learning models The models predict that most basalts in Northeast China have been influenced by fluid derived from the stagnant Pacific slab underneath The boundary defined by fluid activity in basalts coincides with those defined by seismic tomography and Mg isotopes
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0094-8276
1944-8007
DOI:10.1029/2019GL082322