S015
Extracting Information from Geophysical Signals with Machine Learning
Session ID#: 80960
Session Description:
Recent advances in machine learning show great promise in geophysical applications, which include earthquake detection, phase picking, source characterization, ground motion prediction, signal denoising, and subsurface imaging. Future breakthroughs in geophysics should be enabled as more researchers take advantage of the full spectrum of capabilities that machine learning and other data science techniques have to offer. Owing to the inherent complexity of machine learning methods, they are prone to misapplication, may produce uninterpretable models, and can easily be insufficiently documented. This combination of attributes hinders reliable assessment of model validity and consistent interpretation of model outputs. This session aims to engage geophysical communities to share their experience with machine learning, to address the needs (e.g., well documented datasets), and to accelerate progress in the application of data science to geophysics. We welcome contributions that are related to applying machine learning techniques to across the full range of geophysical problems.
Index Terms:
0555 Neural networks, fuzzy logic, machine learning [COMPUTATIONAL GEOPHYSICS]
1910 Data assimilation, integration and fusion [INFORMATICS]
1914 Data mining [INFORMATICS]
1942 Machine learning [INFORMATICS]
Primary Convener: Ting Chen, Los Alamos National Laboratory, Los Alamos, NM, United States
Conveners: Laura J Pyrak-Nolte, Purdue University, Department of Physics and Astronomy, West Lafayette, IN, United States, Paul A Johnson, Los Alamos National Laboratory, Earth and Environmental Sciences: Geophysics (EES-17), Los Alamos, NM, United States and Gregory C Beroza, Stanford University, Department of Geophysics, Stanford, CA, United States
Primary Liaison: Ting Chen, Los Alamos National Laboratory, Los Alamos, NM, United States
Seismic signals and noises clustering with unsupervised deep representation learning (Invited) (490642)
Leonard Seydoux, Universite Grenoble Alpes, ISTerre, Grenoble, France, Randall Balestriero, Rice University, Electrical and Computer Engineering Department, Houston, TX, United States, Piero Poli, CNRS, Grenoble Cedex 9, France, Maarten V De Hoop, Rice University, Computational and Applied Mathematics, Houston, TX, United States, Richard Baraniuk, Rice University, Department of Electrical and Computer Engineering, Houston, TX, United States and Michel Campillo, ISTerre Institute of Earth Sciences, Saint Martin d'Hères, France
Machine Learning Techniques for Detecting Slow, Sustained Deformation in InSAR Time Series. (Invited) (492084)
Juliet Biggs1, Nantheera Anantrasirichai2, Fabien Albino3, Paul Hill4 and Victor Ponce Lopez4, (1)University of Bristol, COMET, School of Earth Sciences, Bristol, United Kingdom, (2)University of Bristol, Bristol, BS8, United Kingdom, (3)University of Miami, Marine Geology and Geophysics, Miami, FL, United States, (4)University of Bristol, Bristol, United Kingdom
A deep-learning framework for inference in geomechanics (493862)
Ehsan Haghighat1, Maziar Raissi2, Adrian Moure Rosende3, Hector Gomez3 and Ruben Juanes1, (1)Massachusetts Institute of Technology, Cambridge, MA, United States, (2)Brown University, Providence, United States, (3)Purdue University, West Lafayette, United States
Machine learning approaches to microseismic event detection in Distributed Acoustic Sensing (DAS) fibre-optic data (496008)
Anna Louise Stork1, Steve A Horne2, Sacha Lapins3, Alan F Baird1, Garth Naldrett4, J Michael Kendall5, James M Wookey5, Andy Clarke4 and James Verdon6, (1)University of Bristol, Bristol, BS8, United Kingdom, (2)Chevron ETC, London, United Kingdom, (3)University of Bristol, Bristol, United Kingdom, (4)Silixa Ltd., Hertfordshire, United Kingdom, (5)University of Bristol, School of Earth Sciences, Bristol, United Kingdom, (6)University of Bristol, School of Earth Science, Bristol, United Kingdom
The detection of Ms 6.0 Changning aftershock sequence based on deep learning algorithms (502215)
Ming Zhao1, David A Yuen2, Lihua Fang3, Bei Zhang Sr4, Yonggang Wei1, Shi Chen1 and Changsheng Jiang5, (1)IGP Institute of Geophysics, China Earthquake Administration, Beijing, China, (2)Univ of Minnesota, Minneapolis, MN, United States, (3)Institute of Geophysics, China Earthquake Administration, Beijing, WI, China, (4)IGP Institute of Geophysics, China Earthquake Administration, int, Beijing, China, (5)Institute of Geophysics, China Earthquake Administration, Beijing, China
A focal mechanism catalog for Southern California derived with deep learning algorithms (504527)
Yifang Cheng1, Zachary E. Ross2, Egill Hauksson2 and Yehuda Ben-Zion1, (1)University of Southern California, Los Angeles, CA, United States, (2)California Institute of Technology, Pasadena, CA, United States
Towards Systematic Classification of Seismic Signals with Deep Neural Networks (504929)
Eric Beauce, Massachusetts Institute of Technology, Cambridge, MA, United States, Glenn Cougoulat, University Joseph Fourier Grenoble, ISTerre, Grenboble, France, Piero Poli, CNRS, Grenoble Cedex 9, France, Leonard Seydoux, Université Grenoble Alpes, Grenoble, France, Robert D van der Hilst, Massachusetts Institute of Technology, Cambridge, United States and Michel Campillo, ISTerre Institute of Earth Sciences, CNRS, Université Grenoble Alpes, Saint Martin d'Hères, France
Predicting the distance to system-scale failure throughout dynamic in situ X-ray tomography experiments (505249)
Jessica McBeck1, John M. Aiken1, Yehuda Ben-Zion2, Benoit Cordonnier1 and Francois Renard1,3, (1)University of Oslo, Oslo, Norway, (2)University of Southern California, Los Angeles, CA, United States, (3)University Joseph Fourier Grenoble, Grenboble, France
Transfer Learning for Automated Seismic Phase Arrival Detection on Local or Temporary Volcano-Seismic Networks with Limited Seismic Catalogue (506176)
Sacha Lapins1, Berhe Goitom1, J Michael Kendall2, Katharine V Cashman1 and James Hammond3, (1)University of Bristol, Bristol, United Kingdom, (2)University of Bristol, Bristol, BS8, United Kingdom, (3)Birkbeck, University of London, London, United Kingdom
Earthquake Detection Using PhaseNet in the Pecos, Texas Region of the Delaware Basin (508711)
Jenna Faith1, Dmitrii Merzlikin2, Weiqiang Zhu3, Marianne S. Karplus1, Stephen A Veitch4, Alexandros Savvaidis2, Diane Irene Doser1, Gregory C Beroza3, William L Ellsworth3 and S. Mostafa Mousavi3, (1)University of Texas at El Paso, Department of Geological Sciences, El Paso, TX, United States, (2)University of Texas at Austin, Bureau of Economic Geology, Austin, TX, United States, (3)Stanford University, Department of Geophysics, Stanford, CA, United States, (4)University of Texas at El Paso, Geological Sciences, El Paso, TX, United States
Phases detection on earthquake and microseismic recordings by a universal CNN-based classifier (510216)
Liang Ding1,2, Wei Qian1, Ergen Gao3 and Qinya Liu2,4, (1)Hohai University, School of Earth Sciences and Engineering, Nanjing, China, (2)University of Toronto, Department of Physics, Toronto, ON, Canada, (3)Anhui Jianzhu University, Hefei, China, (4)University of Toronto, Department of Earth Sciences, Toronto, ON, Canada
Applying Machine Learning to Locate Earthquakes in Sichuan, China (511619)
Kang Wang, School of Earth and Space Sciences, USTC, Hefei, China and Jie Zhang, USTC, China, Hefei, Anhui, China
Classifying emergent and impulsive signals in continuous seismic waveforms (512409)
Christopher W Johnson1, Yehuda Ben-Zion2, Haoran Meng2 and Frank Vernon3, (1)Scripps Institution of Oceanography, La Jolla, CA, United States, (2)University of Southern California, Los Angeles, CA, United States, (3)University of California San Diego, La Jolla, CA, United States
Machine-learning-based interpretation of upper mantle structures beneath the Korean Peninsula (514633)
Youngseok Song, Hanyang University, Seoul, South Korea, Jaewook Lee, University of Texas at Dallas, Richardson, TX, United States and Joongmoo Byun, Hanyang University, Seoul, Korea, Republic of (South)
An unsupervised automatic classification algorithm for continuous seismic records using a nonparametric Bayesian approach (514798)
Yuki Kodera, Meteorological Research Institute, Japan Meteorological Agency, Ibaraki, Japan, Shin'ichi Sakai, Univ Tokyo, Bunkyo-Ku Tokyo, Japan and Takahito Nishiimiya, Meteorological Research Institute, Japan Meteorological Agency, Tsukuba, Japan
Apply a Deep Learning Method to Identify Anomalous Patterns in Ambient Noise Data prior to Major Earthquakes in Sichuan, China (514999)
Zhangyu Wang, School of Earth and Space Sciences, USTC, Hefei, China and Jie Zhang, USTC, China, Hefei, Anhui, China
A Global Data Set of Seismic Signals for AI (522276)
S. Mostafa Mousavi1, Weiqiang Zhu1, Yixiao Sheng2 and Gregory C Beroza1, (1)Stanford University, Department of Geophysics, Stanford, CA, United States, (2)Stanford University, Stanford, CA, United States
A Deep Convolutional Neural Network Model for Earthquake Phase Picking in Taiwan (523478)
Lindsay Yuling Chuang1, Lijun Zhu2, Qiushi Zhai1, Zhigang Peng3, James H. McClellan2 and Kate Huihsuan Chen4, (1)Georgia Institute of Technology, School of Earth and Atmospheric Sciences, Atlanta, GA, United States, (2)Georgia Institute of Technology Main Campus, School of Electrical and Computer Engineering, Atlanta, GA, United States, (3)Georgia Institute of Technology, School of Earth and Atmospheric Sciences, Atlanta, United States, (4)Department of Earth Sciences, National Taiwan Normal University, Taipei, Taiwan
Using Generative Adversarial Networks (GANs) to invert for local density field from gravity data (530148)
Bei Zhang1, Shi Chen1, Ming Zhao1 and David A Yuen2,3, (1)IGP Institute of Geophysics, China Earthquake Administration, Beijing, China, (2)Columbia University, Dept. of Applied Physics and Applied Mathematics, New York, NY, United States, (3)China University of Geosciences Wuhan, Department of Big Data, Wuhan, China
An end-to-end earthquake monitoring method for joint earthquake detection and association using deep learning (531180)
Weiqiang Zhu1, Kai Sheng Tai2, S. Mostafa Mousavi1 and Gregory C Beroza1, (1)Stanford University, Department of Geophysics, Stanford, CA, United States, (2)Stanford University, Computer Science Department, Stanford, CA, United States
Convolutional event embeddings for fast probabilistic earthquake assessment (533028)
Jannes Münchmeyer1,2, Dino Bindi1, Ulf Leser2 and Frederik J Tilmann1,3, (1)Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Potsdam, Germany, (2)Humboldt University of Berlin, Computer Science, Berlin, Germany, (3)Freie Universität Berlin, Geophysics, Berlin, Germany
Automatic mapping of fractures and faults in optical images and Digital Surface Models with deep learning (533965)
Lionel Mattéo1,2, Isabelle Manighetti1, Onur Tasar2, Frédérique Leclerc1, Nicolas Girard2, Yuliya Tarabalka3, Tiziano Giampietro4, Chelsea Scott5, Stephane Dominguez6, Martijn van den Ende4, Ramon Arrowsmith7, Alexandre Vincendeau4, Tyler Scott5 and Jacques Malavieille8, (1)Géoazur-Université Côte d'Azur, Nice, France, (2)Inria-Université Côte d'Azur, Sophia Antipolis, France, (3)LuxCarta, Sophia Antipolis, France, (4)Géoazur-Université Côte d'Azur, Sophia Antipolis, France, (5)Arizona State University, Tempe, AZ, United States, (6)University of Montpellier, Montpellier, France, (7)Arizona State Univ, Tempe, AZ, United States, (8)Universite Montpellier 2, Montpellier, France
Earthquake Magnitude, Distance, First Motion Polarity, and Noise/Event Determination Using AE, CNN1D, RNN Networks (539557)
Payman Janbakhsh, Russell Pysklywec and M Hosein Shahnas, University of Toronto, Department of Earth Sciences, Toronto, ON, Canada
How robust is deep learning-based earthquake detection? Insights from adversarial machine learning (539566)
Karianne J. Bergen1,2, Flavio P. Calmon2 and David C. Parkes2, (1)Harvard University, Earth and Planetary Sciences, Cambridge, MA, United States, (2)Harvard University, School of Engineering and Applied Sciences, Cambridge, MA, United States
Automatic seismic phase picking using machine learning for the EGS Collab project (542386)
Chengping Chai1, Monica Maceira1, Hector J. Santos-Villalobos1, Singanallur V. Venkatakrishnan2, Martin Schoenball3 and EGS Collab Team, (1)Oak Ridge National Laboratory, Oak Ridge, TN, United States, (2)Oak Ridge National Laboratory, Oak Ridge, United States, (3)Lawrence Berkeley National Laboratory, Berkeley, CA, United States
CONTRIBUTION OF MACHINE LEARNING TO EARTHQUAKE DETECTION IN HIGH ANTHROPOGENIC CONTEXT (553552)
Alexandra Renouard1,2, Marc Grunberg3, Cecile Doubre1 and Alessia Maggi2, (1)University of Strasbourg, Strasbourg Cedex, France, (2)EOST École et Observatoire des Sciences de la Terre, Strasbourg Cedex, France, (3)Université de Strasbourg, CNRS, EOST UMS830, Strasbourg, France
Machine Learning Reveals the Seismic Signature of Eruptive Behavior at Piton de la Fournaise (555193)
Christopher Ren, Los Alamos National Laboratory, Los Alamos, NM, United States, Florent Brenguier, University Joseph Fourier Grenoble, Grenboble, France, Aline Peltier, Université de Paris, Institut de Physique du Globe de Paris (IPGP), UMR 7154, Paris, France, Valerie Ferrazzini, Institut de Physique du Globe de Paris, Paris, France and Paul A Johnson, Los Alamos National Laboratory, Earth and Environmental Sciences: Geophysics (EES-17), Los Alamos, NM, United States
Accurate automatic Phase picking for continuous, noisy microseismic recordings without prefiltering. (557226)
Angelos Mongolias, University of Strathclyde, Civil & Environmental Engineering, Glasgow, United Kingdom, Stella Pytharouli, University of Strathclyde, Cumbernauld, United Kingdom and Rebecca Jane Lunn, University of Strathclyde, Glasgow, United Kingdom
Dense Northern Chile Catalog with Graph-Theory-Based Earthquake Arrival Associations (559828)
Ian W. McBrearty1, Joan S Gomberg2, Andrew Delorey3 and Paul A Johnson1, (1)Los Alamos National Laboratory, Earth and Environmental Sciences: Geophysics (EES-17), Los Alamos, NM, United States, (2)US Geological Survey, Seattle, WA, United States, (3)Los Alamos National Laboratory, Los Alamos, NM, United States
Identifying P and S phases using supervised learning algorithms (578641)
Bjorn Lund1, Gunnar A Eggertsson1, Peter Schmidt2 and Reynir Bodvarsson3, (1)Uppsala University, Uppsala, Sweden, (2)Uppsala university, Uppsala, Sweden, (3)Swedish National Seismic Network, Uppsala, Sweden
HEX: Hyperbolic Event eXtractor, a seismic phase association algorithm for regions of intense seismicity (579210)
Jack Woollam1,2, Andreas Rietbrock1, Stefan Hinz1 and Jens Leitloff1, (1)Karlsruhe Institute of Technology, Karlsruhe, Germany, (2)University of Liverpool, Liverpool, United Kingdom
An Unsupervised-Machine-Learning Approach to Understanding Seismicity at an Alpine Glacier (584572)
Theresa Sawi, Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY, United States, Meredith Nettles, Columbia University, Lamont-Doherty Earth Observatory, Palisades, NY, United States, Benjamin K Holtzman, Lamont Doherty Earth Observatory, Columbia Univ., Palisades, NY, United States, John Paisley, Columbia University of New York, Data Science Institute, Dept. of Electrical Engineering, New York, NY, United States and Fabian Walter, ETH Swiss Federal Institute of Technology Zurich, Laboratory of Hydraulics, Hydrology and Glaciology, Zurich, Switzerland
Seismic Clipped Waveform Reconstruction and Noise Attenuation Using Deep Learning (587558)
Madison Spencer Sheridan, Texas A&M University College Station, Math, College Station, TX, United States and Cleat Philip Zeiler, Mission Support and Test Services, Defense Nuclear Nonproliferation, North Las Vegas, NV, United States
Convolutional Neural Network to Detect Strombolian Eruptions at Mount Erebus, Antarctica (589313)
Taylor Privat, Self Employed, Washington, DC, United States, Brian Dye, University of Louisiana at Lafayette, Lafayette, LA, United States and Gabriele Morra, University of Louisiana at Lafayette, Department of Physics and School of Geosciences, Lafayette, LA, United States
Unsupervised Classification of Acoustic Emissions from Catalogs and Fault Time-to-Failure Prediction (603595)
Hope Abigail Jasperson, Rice University, Earth, Environmental and Planetary Sciences, Houston, TX, United States, David Chas Bolton, Pennsylvania State University, University Park, PA, United States, Paul A Johnson, Los Alamos National Laboratory, Earth and Environmental Sciences: Geophysics (EES-17), Los Alamos, NM, United States, Chris Marone, Pennsylvania State University, Department of Geosciences, University Park, PA, United States and Maarten Dehoop, Rice University, Computational and applied mathematics, Houston, TX, United States
Characterizing dynamics in materials through automated learning and classification of acoustic signals emitted during drying of geo-architected rocks (605127)
Charles Fieseler, University of Washington Seattle Campus, Physics, Seattle, WA, United States, Chven A Mitchell, Purdue University, Earth, Atmospheric and Planetary sciences, West Lafayette, IN, United States, Nathan Kutz, University of Washington Seattle Campus, Applied Mathematics, Seattle, WA, United States and Laura J Pyrak-Nolte, Purdue University, Department of Physics and Astronomy, West Lafayette, IN, United States
Event Classification of the Mount St Helens Nodal dataset Using Unsupervised Machine Learning (622426)
Steven M Hansen, Macquarie University, Sydney, NSW, Australia, Brandon Schmandt, University of New Mexico, Department of Earth & Planetary Sciences, Albuquerque, NM, United States and Margaret Elizabeth Glasgow, University of New Mexico Main Campus, Albuquerque, NM, United States
Incorporating machine learning to aid global earthquake monitoring at the USGS National Earthquake Information Center (629042)
William L Yeck1, Paul S Earle1, Nicholas Ambruz2, Harley Benz1, John Patton1, Gavin P Hayes1 and Michelle R Guy1, (1)USGS National Earthquake Information Center Golden, Golden, CO, United States, (2)USGS Earthquake Science Center, Golden, CO, United States
