NG008
Machine Learning in Space Weather
Session ID#: 72833
Session Description:
In the last few years, several ground-breaking milestones were reached in Artificial Intelligence, such as image recognition at super-human accuracy and the notorious defeat of world champion Lee Sedol in the game of Go.
Advances in Machine Learning (ML) have been fueled by the growth of available data and ubiquity of capable computing technologies. Space physics and geophysics are characterized by impressive amounts of freely-available data and can certainly benefit from data-driven ML applications.
This session will focus on applications of Machine Learning to Space Weather problems across the subdisciplines of the Heliophysics and the physics of the Earth's environment, involving as well the coupling of the interplanetary plasma with the upper atmosphere. Contributions ranging from black-box models to data-driven physics-based simulations are welcome, including (but not limited to) regression and classification problems, inverse problems, dimensionality reduction, automatic event identification, Bayesian inference, feature extraction, deep learning, and reinforcement learning.
Co-Organized:
Index Terms:
1914 Data mining [INFORMATICS]
1942 Machine learning [INFORMATICS]
7599 General or miscellaneous [SOLAR PHYSICS, ASTROPHYSICS, AND ASTRONOMY]
7999 General or miscellaneous [SPACE WEATHER]
Primary Convener: Enrico Camporeale, University of Colorado at Boulder, Boulder, United States; University of Colorado, Boulder, United States
Conveners: Jacob Bortnik, University of California Los Angeles, Atmospheric and Oceanic Sciences, Los Angeles, CA, United States, Tomoko Matsuo, University of Colorado Boulder, Boulder, CO, United States and Ryan Michael McGranaghan, Atmospheric and Space Technology Research Associates (ASTRA), Louisville, CO, United States
Primary Liaison: Enrico Camporeale, University of Colorado at Boulder, Boulder, United States; University of Colorado, Boulder, United States
Dynamical networks: a tool for identifying the evolution of spatio-temporal pattern in spatially irregular, multi (>100) point observations (Invited) (490669)
Sandra C Chapman1, Lauren Orr1 and Jesper W Gjerloev2, (1)University of Warwick, Centre for Fusion, Space and Astrophysics, Department of Physics, Coventry, United Kingdom, (2)Johns Hopkins University - Applied Physics Laboratory, Laurel, MD, United States
Generation of Global TEC maps (498149)
Eun-Young Ji1, Yong-Jae Moon2 and Eunsu Park2, (1)Kyung Hee University, Yongin, Korea, Republic of (South), (2)School of Space Research, Kyung Hee University, Yongin, South Korea
Physics-informed machine learning for estimating the electron flux in the Earth's radiation belts (502140)
Rakesh Sarma1, George John Wilkie1,2, Enrico Camporeale1,3, Alexander Drozdov4 and Yuri Shprits5, (1)Centrum Wiskunde & Informatica, Amsterdam, Netherlands, (2)Princeton Plasma Physics Laboratory, Princeton, United States, (3)University of Colorado, Cooperative Institute for Research in Environmental Sciences, Boulder, United States, (4)University of California Los Angeles, Los Angeles, CA, United States, (5)GFZ German Research Centre for Geosciences, Potsdam, Germany
Application of Deep Learning to Solar and Space Weather Data: 5. Generation of Historic Solar Magnetograms and UV images (502440)
Harim Lee1, Yong-Jae Moon2, Eunsu Park2, Daye Lim2, Gyungin Shin2, Hyunjin Jeong2 and Taeyoung Kim2, (1)Kyung Hee University, Yongin, South Korea, (2)School of Space Research, Kyung Hee University, Yongin, South Korea
Application of Deep Learning to Solar and Space Weather Data: 1. Overview (502898)
Yong-Jae Moon1, Eunsu Park1, Taeyoung Kim1, Hyunjin Jeong1, Daye Lim1, Harim Lee2, Gyungin Shin1, Eun-Young Ji3 and Kangwoo Yi1, (1)School of Space Research, Kyung Hee University, Yongin, South Korea, (2)Kyung Hee University, Yongin, South Korea, (3)Kyung Hee University, Yongin, Korea, Republic of (South)
A 3D Global Dynamic Plasmapause Model Based on Multi-satellite Measurements Developed by Using a Back Propagation Neural Network (504749)
Zheng Zhiqi1,2, Jiuhou Lei3, Xinan Yue1,2 and Fei He1,2, (1)University of Chinese Academy of Sciences, Beijing, China, (2)Institute of Geology and Geophysics, Chinese Academy of Science, Beijing, China, (3)Univ. of Sci. & Tech. of Chin, Hefei, Anhui, China
A Deep Learning Approach to Forecast Tomorrow's Solar Wind Parameters. (504814)
Carl Shneider1, Enrico Camporeale2,3 and Mandar Chandorkar1, (1)Dutch National Institute for Mathematics and Computer Science (CWI), Multiscale Dynamics, Amsterdam, Netherlands, (2)University of Colorado, Boulder, United States, (3)University of Colorado, Cooperative Institute for Research in Environmental Sciences, Boulder, United States
Application of Deep Learning to Solar and Space Weather Data: 2. Generation of Farside Solar Magnetograms (511893)
Taeyoung Kim1,2, Eunsu Park2, Harim Lee2, Yong-Jae Moon2, Sung-Ho Bae3, Daye Lim2, Soojeong Jang4, Lokwon Kim3, Il-Hyun Cho5, Myungjin Choi1 and Kyung-Suk F Cho4, (1)InSpace Co., Ltd., Daejeon, South Korea, (2)School of Space Research, Kyung Hee University, Yongin, South Korea, (3)Department of Computer Science and Engineering, Kyung Hee University, Yongin, South Korea, (4)Korea Astronomy and Space Science Institute, Daejeon, South Korea, (5)Department of Astronomy and Space Science, Kyung Hee University, Yongin, South Korea
Generative Deep Learning with SDO and SOHO EUV Solar Images (514778)
Ruggero Vasile1, Frederic Effenberger2 and Yuri Shprits2,3, (1)Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Geosciences, Potsdam, Germany, (2)Helmholtz Centre Potsdam GFZ German Research Centre for Geosciences, Potsdam, Germany, (3)University of California Los Angeles, Los Angeles, United States
Chirality Classification for Space Weather Warnings (515578)
Aparna V, Georgia State University, Atlanta, GA, United States, Rafal Angryk, Georgia State University, Computer Science, Atlanta, GA, United States and Petrus C Martens, Georgia State University, Physics and Astronomy, Atlanta, GA, United States
Avoiding More Accurate and Less Robust Models: Mistakes and Pitfalls in Training Flare Forecasting Models (516736)
Azim Ahmadzadeh1, Berkay Aydin2, Dustin Kempton1, Sushant Mahajan3, Manolis K Georgoulis4 and Rafal Angryk2, (1)Georgia State University, Atlanta, GA, United States, (2)Georgia State University, Computer Science, Atlanta, GA, United States, (3)Georgia State University, Physics and Astronomy, Atlanta, United States, (4)RCAAM of the Academy of Athens, Athens, Greece
A Deep Neural Network Model of Global Topside Electron Temperature Modeling Using Incoherent Scatter Radars and Its Application to GPS Radio Occultation (526590)
Andong Hu, Centrum Wiskunde & Informatica, Multi-scale, Amsterdam, Netherlands; RMIT University, Geospatial Science, Melbourne, VIC, Australia, Brett A Carter, RMIT University, SPACE Research Centre, Melbourne, VIC, Australia, Julie Louise Currie, RMIT University, SPACE research Centre, Melbourne, VIC, Australia, Enrico Camporeale, University of Colorado, Cooperative Institute for Research in Environmental Sciences, Boulder, United States; Center for Mathematics and Computer Science (CWI), Amsterdam, Netherlands, Jannis Teunissen, Centrum Wiskunde & Informatica, Amsterdam, Netherlands, Robert Norman, RMIT University, Melbourne, Australia and Kefei Zhang, RMIT Univ, Melbourne, Australia
Forecast of Ionosphere GNSS TEC using CNN and LSTM neural network (526891)
Lei Liu1,2, Shasha Zou2, Yibin Yao1, Zihan Wang2 and Ercha A2, (1)Wuhan University, School of Geodesy and Geomatics, Wuhan, China, (2)University of Michigan, Ann Arbor, MI, United States
Nowcasting of Ionospheric Total Electron Content Using the Random Forest and Deep Neural Network Machine Learning Methods (531146)
Gebreab Zewdie1, Morris Cohen1, Cesar Enrique Valladares2 and David J Lary2, (1)Georgia Institute of Technology Main Campus, School of Electrical and Computer Engineering, Atlanta, GA, United States, (2)University of Texas at Dallas, Richardson, TX, United States
Medium energy electron fluxes in Earth’s outer radiation belt: a machine learning model (531296)
Artem Smirnov, Ludwig Maximilians University of Munich, Earth and Environmental Sciences, Munich, Germany, Elena A Kronberg, Max Planck Institute for Solar System Research, Gottingen, Germany, Max Berrendorf, Ludwig Maximilians University of Munich, Database Systems and Data Mining, Munich, Germany, Steven Morley, Los Alamos National Laboratory, Los Alamos, NM, United States and Yuri Shprits, GFZ German Research Centre for Geosciences, Potsdam, Germany
Joint Classification and Instance Segmentation of Auroral Images for Automatic Computation of Arc Width (532300)
Chuang Niu1, Yiping Tang1, Shenghan Ren1, Ze-Jun Hu2, Han Desheng3 and Jimin Liang1, (1)Xidian University, Xi'an, China, (2)Polar Research Institute of China, Shanghai, China, (3)School of Ocean and Earth Science, Tongji University, Shanghai, China
Poleward Moving Aurora Detection based on Deep Learning (539951)
Yiping Tang1, Chuang Niu1, Shenghan Ren1, Ze-Jun Hu2, Han Desheng3 and Jimin Liang1, (1)Xidian University, Xi'an, China, (2)Polar Research Institute of China, Shanghai, China, (3)School of Ocean and Earth Science, Tongji University, Shanghai, China
Neural Network Approaches To Find Evidences Of EMIC Waves And Their Contribution To Pitch Angle Distribution Reconfiguration In The Earth's Outer Van Allen Radiation Belts (549735)
Claudia Medeiros1, Vitor Souza2, Luis Vieira3, David G Sibeck4, Ligia Alves Da Silva5, Livia Ribeiro Alves6, Jose Marchezi2, Paulo Ricardo Jauer7, Marlos Rockenbach2, Alisson Dal Lago2, Shrikanth G Kanekal8, Dan N Baker9 and Craig Kletzing10, (1)National Institute for Space Research (INPE), Sao Jose dos Campos, Brazil, (2)INPE National Institute for Space Research, Sao Jose dos Campos, Brazil, (3)INPE National Institute for Space Research, The Space Geophysics Division, Sao Jose dos Campos, Brazil, (4)NASA/GSFC, Greenbelt, MD, United States, (5)Inst Nac Pesquisas Espaciais, Sao José dos Campos, Brazil, (6)INPE National Institute for Space Research, Space Science Division, Sao Jose dos Campos, Brazil, (7)Inst Nac Pesquisas Espaciais, são josé dos campos - SP, Brazil, (8)NASA GSFC, Greenbelt, MD, United States, (9)University of Colorado Boulder, Laboratory for Atmospheric and Space Physics, Boulder, CO, United States, (10)University of Iowa, Iowa City, IA, United States
Super-Resolution Maps of the Solar Magnetic Field Covering 40 Years of Space Weather Events (549829)
Paul James Wright, Stanford University, W. W. Hansen Experimental Physics Laboratory, Stanford, CA, United States, Xavier Gitiaux, George Mason University, Fairfax, VA, United States, Anna Jungbluth, University of Oxford, Oxford, United Kingdom, Shane Maloney, Trinity College Dublin, Dublin, Ireland; Dublin Institute for Advanced Studies, Dublin, Ireland, Carl Shneider, Dutch National Institute for Mathematics and Computer Science (CWI), Multiscale Dynamics, Amsterdam, Netherlands and Andres Munoz-Jaramillo, SouthWest Research Institute, Boulder, CO, United States
Dynamic Time Lag Regression: Predicting Time Lagged Effects of Solar Activity (552891)
Mandar Chandorkar1, Enrico Camporeale2, Bala Poduval3, Cyril Furthlener4 and Michélè Sebag4, (1)Centrum Wiskunde & Informatica, Amsterdam, Netherlands, (2)University of Colorado, Cooperative Institute for Research in Environmental Sciences, Boulder, United States, (3)Space Science Institute, Boulder, CO, United States, (4)INRIA Paris-Saclay, Paris, France
Forecasting Substorms using Recurrent Neural Networks (567913)
Charles Topliff, Georgia Institute of Technology Main Campus, Atlanta, GA, United States, Morris Cohen, Georgia Institute of Technology Main Campus, School of Electrical and Computer Engineering, Atlanta, GA, United States, William A Bristow, University of Alaska Fairbanks, Fairbanks, AK, United States, Ryan Michael McGranaghan, University of Colorado at Boulder, Boulder, CO, United States and Jacob Bortnik, University of California Los Angeles, Atmospheric and Oceanic Sciences, Los Angeles, CA, United States
Using Deep Learning for Spectropolarimetric Inversions (570477)
Serena Flint, University of Rochester, Rochester, NY, United States and Ivan Milic, National Solar Observatory, Boulder, CO, United States; University of Colorado at Boulder, Boulder, CO, United States
Visualization Far-Side Magnetic Images from Helioseismic Holography (571125)
Rasha Alshehhi, New York University Abu Dhabi, Abu Dhabi, United Arab Emirates, Christopher Hanson, New York University, Abu Dhabi, United Arab Emirates and Laurent C. Gizon, Max Planck Institute for Solar System Research, Katlenburg-Lindau, Germany
Auroral Image Clustering Based on Deep Learning Technology (578193)
Qiu-Ju Yang, School of Physics and Information Technology, Shaanxi Normal University, Xi'an, Shaanxi Province, China and Chang Liu, Northeastern University, Mechanical & Industrial Engineering, Boston, MA, United States
Automatic classification of plasma regions using 3D energy distributions (579410)
Vyacheslav Olshevsky1, Yuri Khotyaintsev2, Andrey V Divin3, Gian Luca Delzanno4, Sven Anderzen1, Pawel Herman1, Steven W.D. Chien1, Levon A Avanov5 and Stefano Markidis1, (1)KTH Royal Institute of Technology, Stockholm, Sweden, (2)IRF Swedish Institute of Space Physics, Uppsala, Sweden, (3)St. Petersburg State university, St Petersburg, Russia, (4)Los Alamos National Laboratory, Los Alamos, NM, United States, (5)NASA Goddard Space Flight Center, Greenbelt, MD, United States
A Robust System Science Approach to the Complex Magnetospheric dynamics (579937)
Juan Alejandro Valdivia, University of Chile, Department of Physics, Faculty of Sciences, Santiago, Chile, Benjamin Toledo, University of Chile, Departament of Physics, Faculty of Sciences, Santiago, Chile, Jose Rogan, Universidad de Chile, Santiago, Chile and Sylvain Blunier, Universidad de Chile, Physics, Santiago, Chile
Neural Network Based Model for Global Total Electron Content Forecasting (581867)
Claudio Cesaroni1, Luca Spogli2,3, Angela Aragon-Angel4, Michele Fiocca5, Varuliantor Dear6, Giorgiana De Franceschi2 and Vincenzo Romano2,3, (1)Istituto Nazionale di Geofisica e Vulcanologia, Palermo, Italy, (2)Istituto Nazionale di Geofisica e Vulcanologia, Rome, Italy, (3)SpacEarth Technology, Rome, Italy, (4)Joint Research Center Ispra, Ispra, Italy, (5)Independent researcher, Rome, Italy, (6)National Institute of Aeronautics and Space LAPAN, Bandung, Indonesia
A Neural Network Model for Plasma Sheet Electrons from Solar Wind Observations (583383)
Brian Michael Swiger1, Michael Warren Liemohn2 and Natalia Y Ganushkina1, (1)University of Michigan Ann Arbor, Ann Arbor, MI, United States, (2)University of Michigan, Climate and Space Sciences and Engineering, Ann Arbor, MI, United States
Towards the Objective Classification of Coronal Hole and Streamer Belt Solar Wind (589781)
Téo Bloch1, Clare Watt2, Mathew James Owens3, Leland McInnes4 and Allan Macneil1, (1)University of Reading, Reading, United Kingdom, (2)University of Reading, Reading, RG6, United Kingdom, (3)Uinversity of Reading, Reading, United Kingdom, (4)Tutte Institute for Mathematics and Computing, Ottawa, ON, Canada
Identifying Flux Ropes signatures using Deep Learning (601964)
Luiz Fernando Guedes dos Santos1, Ayris Narock2, Teresa Nieves-Chinchilla3, Marlon Nunez4, Tiago Pinho da Silva5, Michael S Kirk3 and Barbara J Thompson6, (1)Catholic University of America, Washington, DC, United States, (2)ADNET Systems Inc. Greenbelt, Greenbelt, MD, United States, (3)NASA Goddard Space Flight Center, Greenbelt, MD, United States, (4)Universidad de malaga, Malaga, Spain, (5)USP University of Sao Paulo, Mathematical Sciences and Computational Institute, São Paulo, Brazil, (6)NASA/GSFC, Greenbelt, MD, United States
Leveraging Topological Data Analysis and Deep Learning for Solar Flare Prediction (615286)
Varad Deshmukh, University of Colorado Boulder, Computer Science, Boulder, United States, Elizabeth Bradley, University of Colorado at Boulder, Computer Science, Boulder, CO, United States, Thomas E Berger, University of Colorado, Space Weather Technology, Research, and Education Center, Boulder, CO, United States, James D Meiss, University of Colorado Boulder, Applied Mathematics, Boulder, United States and Naoto Nishizuka, National Institute of Information and Communications Technology, Tokyo, Japan
Using Unsupervised Machine Learning to Explore New Classification of Sunspot Active Regions (619773)
Sara Nicole Housseal, Millersville University of Pennsylvania, Millersville, PA, United States, Thomas E Berger, NOAA Space Weather Prediction Center, Boulder, CO, United States and Varad Deshmukh, University of Colorado Boulder, Computer Science, Boulder, United States
Emulating Numerical Simulations of the Sun to Infer Synthetic Plasma Motions from Observations of the Quiet Sun and Sunspots (624387)
Benoit Tremblay, University of Montreal, Physics department, Montreal, QC, Canada; Laboratory for Atmospheric and Space Physics, Boulder, CO, United States, Thierry Roudier, IRAP, Toulouse, France, Raphael Attié, NASA Goddard Space Flight Center, Greenbelt, MD, United States, Jean-Francois Cossette, Michel Rieutord, Université de Toulouse, UPS-OMP, IRAP, Toulouse, France and Alain P Vincent, Universite de Montreal, Montreal, QC, Canada
A deep-learning based approach for predicting high latitude ionospheric scintillations using geospace data and auroral imagery (625667)
Garima Malhotra, University of Michigan Ann Arbor, Ann Arbor, MI, United States, Athanasios Vlontzos, Imperial College London, London, United Kingdom, Edward Wagstaff, University of Oxford, Oxford, United Kingdom, Kara Lamb, NOAA Earth System Research Laboratory, Chemical Sciences Division, Boulder, CO, United States and Asti Bhatt, SRI International, Menlo Park, CA, United States
Auto-calibration and reconstruction of SDO’s Atmospheric Imaging Assembly channels with Deep Learning (628427)
Chun Ming Mark Cheung1, Luiz Fernando Guedes dos Santos2, Souvik Bose3, Brad Neuberg4, Valentina Salvatelli5, Atılım Güneş Baydin6, Miho Janvier7 and Meng Jin5, (1)Lockheed Martin Solar and Astrophysics Laboratory, Palo Alto, CA, United States, (2)Catholic University of America, Washington, DC, United States, (3)University of Oslo, Rosseland Center for Solar Physics, Oslo, Norway, (4)SETI Institute, Mountain View, United States, (5)SETI Institute, Mountain View, CA, United States, (6)University of Oxford, Department of Engineering Science, Oxford, United Kingdom, (7)Institut d'Astrophysique Spatiale, Orsay, France
A Machine Learning Approach to Forecasting Solar Radiation Storms at Earth (Invited) (490319)
Hazel M Bain, University of Colorado at Boulder, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States; NOAA Space Weather Prediction Center, Boulder, CO, United States, Pedro Brea, University of Texas at Dallas, Richardson, TX, United States and Eric T Adamson, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, United States
