Geochemical Mechanics and Deep Neural Network Modeling: Applications to Earthquake Prediction

Toriumi, Mitsuhiro

ISBN 10: 9811936587 / ISBN 13: 9789811936586

Published by Springer Nature, 2022

About this Item

287 pages. 9.25x6.10x0.79 inches. In Stock.

Seller Inventory # x-9811936587

Bibliographic Details

• Title: Geochemical Mechanics and Deep Neural Network Modeling: Applications to Earthquake Prediction
• Publisher: Springer Nature
• Publication Date: 2022
• Language: English
• ISBN 10: 9811936587
• ISBN 13: 9789811936586
• Binding: Hardcover
• Condition: Brand New
• Weight: 0.61 kilograms

About this title

Synopsis

The recent understandings about global earth mechanics are widely based on huge amounts of monitoring data accumulated using global networks of precise seismic stations, satellite monitoring of gravity, very large baseline interferometry, and the Global Positioning System. New discoveries in materials sciences of rocks and minerals and of rock deformation with fluid water in the earth also provide essential information. This book presents recent work on natural geometry, spatial and temporal distribution patterns of various cracks sealed by minerals, and time scales of their crack sealing in the plate boundary. Furthermore, the book includes a challenging investigation of stochastic earthquake prediction testing by means of the updated deep machine learning of a convolutional neural network with multi-labeling of large earthquakes and of the generative autoencoder modeling of global correlated seismicity. Their manifestation in this book contributes to the development of human societyresilient from natural hazards. Presented here are (1) mechanics of natural crack sealing and fluid flow in the plate boundary regions, (2) large-scale permeable convection of the plate boundary, (3) the rapid process of massive extrusion of plate boundary rocks, (4) synchronous satellite gravity and global correlated seismicity, (5) Gaussian network dynamics of global correlated seismicity, and (6) prediction testing of plate boundary earthquakes by machine learning and generative autoencoders.

About the Author

Mitsuhiro Toriumi is a senior researcher at the Research Institute of Marine Geodynamics, the Japan Agency for Marine–Earth Science and Technology (JAMSTEC). He studies physical processes regarding metamorphism, metasomatism, tectonics, and rheology of plate boundary zones. He was the scientific director of the Institute for Research on Earth Evolution (IFREE) between 2011 and 2013, the chief leader of the Earth and Marine Biology Research Group (2014–2015), and a senior researcher of the Innovation Center (2016–2018) of JAMSTEC. He is a fellow of the Japan Geoscience Union.