DI002
Advances in Machine Learning for Solid Earth Geoscience
Session ID#: 248661
Session Description:
With the rapid growth of observational datasets, advancements in machine learning algorithms, and expanding computational power, machine learning is playing an increasingly significant role in Solid Earth Geosciences. These tools are transforming our understanding of physical and chemical processes across spatial and temporal scales, both on Earth and other planetary bodies.This session invites contributions across a broad spectrum of methods and applications, including but not limited to: data compilation and mining, statistical modeling, classical and deep learning, explainable AI, and generative models applied to geophysics, geodynamics, geochemistry, structural geology, volcanology, petrology, and mineral physics. We encourage both novel machine learning methodologies addressing geoscientific challenges and innovative applications that provide new insights into Earth processes. Example topics include dimensionality reduction or clustering analysis on geochemical or volcanological data, uncertainty-aware geophysical inversion, seismic event detection with neural networks, geodynamical emulators, and machine-learning assisted multiscale modeling.
Co-Sponsor(s):
Index Terms:
0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
1027 Composition of the planets [GEOCHEMISTRY]
3919 Equations of state [MINERAL PHYSICS]
7290 Computational seismology [SEISMOLOGY]
Primary Convener: Xiyuan Bao, Harvard University, Department of Earth and Planetary Sciences, Cambridge, MA, United States
Conveners: Jie Deng, Princeton University, Department of Geosciences, Princeton, United States, Karianne J. Bergen, Brown University, Data Science Institute and Department of Earth, Environmental and Planetary Sciences, Providence, United States and Maurizio Petrelli, University of Perugia, Department of Physics and Geology, Perugia, Italy
Student/Early Career Convener: Caifeng Zou, California Institute of Technology, Pasadena, United States
The sub-arc mantle has remained oxidized since the Neoproterozoic Oxygenation: Insights from Machine Learning (1854348)
J ZhangZhou1, Chuntao Liu2, Chenyang Ye2 and Qunke Xia2, (1)Zhejiang University, School of Earth Sciences, Hangzhou, China, (2)Zhejiang University, Earth Sciences, Hangzhou, China
Three-dimensional Lithospheric Thermal Structure Study of the Tibetan Plateau Using Data-driven Methods (1858110)
Zhengliang Zhang1, Zhenhong Du1, Prof. Sensen Wu1, Baohua Zhang1 and Qunke Xia2, (1)Zhejiang University, Hangzhou, China, (2)Zhejiang University, Earth Sciences, Hangzhou, China
Intermediate-Depth Earthquakes Beneath East Antarctica Reveal Lithosphere-Asthenosphere Interactions Control Intraplate Deformation (1872367)
Long Ho1, José Luis Sánchez-Roldán2, Samantha Hansen1 and Jacob I Walter3, (1)University of Alabama, Department of Geological Sciences, Tuscaloosa, United States, (2)Complutense University of Madrid, Department of Geodynamics, Stratigraphy & Paleontology, Madrid, Spain, (3)University of Oklahoma, Oklahoma Geological Survey, Norman, United States
Machine Learning–Driven Acceleration of Integrated Geodynamic-Geochemical Models (1875416)
Qian Yuan, Texas A&M University College Station, College Station, TX, United States, Paul D Asimow, California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, CA, United States, Michael Gurnis, California Institute of Technology, Seismological Laboratory, Pasadena, CA, United States, Paula M Antoshechkina, California Inst Tech, Pasadena, United States and Junjie Dong, California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, United States
Detecting Seismic PS Wave Arrivals in the Alaska Peninsula Through Machine Learning (1891437)
Nick Turner, Winona State University, Department of Geoscience, Winona, United States, Andrew Birkey, University of Delaware, Department of Earth Sciences, Newark, DE, United States, Yaqi Jie, Michigan State University, Department of Earth and Environmental Sciences, East Lansing, United States and Songqiao Shawn Wei, Michigan State University, Department of Earth and Environmental Sciences, East Lansing, MI, United States
Towards Physics-based Machine Learning Framework of Seismic Wavefield Modeling and Full-waveform Inversion (Invited) (1908248)
Shiqian Li1, Zhi Li2, Zhancun Mu1, Shiji Xin3, Zhixiang Dai4, Kuangdai Leng5, Ruihua Zhang4, Yixin Zhu1 and Xiaodong Song1,2, (1)Institute for Artificial Intelligence, Peking University, Beijing, China, (2)School of Earth and Space Sciences, Peking University, Beijing, China, (3)Harvard University, Cambridge, United States, (4)NVIDIA, Beijing, China, (5)Rutherford Appleton Laboratory, Harwell Campus, Didcot, United Kingdom
Adapting Deep Learning Models using an Instance-Weighting Approach for Seismic Fault Detection (1908741)
Tiash Ghosh, Indian Institute of Technology Kharagpur, Department of Electrical Engineering, Kharagpur, India, Mamata Jenamani, Indian Institute of Technology, Kharagpur, Department of Industrial & Systems Engineering, Kharagpur, India, Aurobinda Routray, Indian Institute of Technology, Kharagpur, Department of Electrical Engineering, Kharagpur, India and Sanjai Kumar Singh, GEOPIC, ONGC, Dehradun, India
P-wave First-Motion Polarity Classification for Seafloor Microearthquakes Using Transfer Learning at Axial Seamount (1928747)
Maochuan Zhang, University of Washington, School of Oceanography, Seattle, United States, William S D Wilcock, University of Washington Seattle Campus, School of Oceanography, Seattle, United States, Marine Denolle, University of Washington Seattle Campus, Department of Earth and Space Sciences, Seattle, United States, Felix Waldhauser, Columbia University, Lamont-Doherty Earth Observatory, Palisades, United States, Kaiwen Wang, Stanford University, Department of Geophysics, Stanford, United States, Maya Tolstoy, University of Washington, College of the Environment, Seattle, United States and Yen Joe Tan, The Chinese University of Hong Kong, Department of Earth and Environmental Sciences, Hong Kong, Hong Kong
Optimizing earthquake phase association performance through semi-supervised calibration of synthetic training data generator (1938671)
Ian W. McBrearty, Stanford University, Department of Geophysics, Stanford, United States, Gabriel Dupuis, ENSTA, Institut Polytechnique de Paris, Paris, France and Gregory C Beroza, Stanford University, Department of Geophysics, Stanford, CA, United States
2D Seismic Datasets for Training Signal Processing Neural Networks: Fast Convolutional Synthetics for Ground-Truth Labels (1943951)
Giacomo Roncoroni, University of Trieste, Trieste, Italy, Ivan Deiana, Stanford University, Stanford, CA, United States and Robert Graham Clapp, Stanford University, Stanford, United States
A Deep Learning-Based Joint Inversion Method for Multiphysics Data with Dual Constraints on Spatial Structures and Petrophysical Properties (1947732)
Biao Xi1,2 and Zhongxing Wang1,2, (1)Institute of Geology and Geophysics, Chinese Academy of Sciences, Key Laboratory of Deep Petroleum Intelligent Exploration and Development, Beijing, China, (2)University of Chinese Academy of Sciences, College of Earth and Planetary Sciences, Beijing, China
