Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning

We develop two artificial intelligence–based models (Models A and B) to predict time-evolving photospheric magnetic fields with an adjustable time step, ranging up to one solar rotation later. Model A predicts future magnetic field data using three consecutive radial magnetic field data sets, each 1...

Full description

Saved in:
Bibliographic Details
Main Authors: Hyun-Jin Jeong, Stefaan Poedts, Haopeng Wang, Ekatarina Dineva, Panagiotis Gonidakis, Francesco Carella, George Miloshevich, Luis Linan, Harim Lee
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:The Astrophysical Journal Letters
Subjects:
Online Access:https://doi.org/10.3847/2041-8213/adf5c7
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849770866791940096
author Hyun-Jin Jeong
Stefaan Poedts
Haopeng Wang
Ekatarina Dineva
Panagiotis Gonidakis
Francesco Carella
George Miloshevich
Luis Linan
Harim Lee
author_facet Hyun-Jin Jeong
Stefaan Poedts
Haopeng Wang
Ekatarina Dineva
Panagiotis Gonidakis
Francesco Carella
George Miloshevich
Luis Linan
Harim Lee
author_sort Hyun-Jin Jeong
collection DOAJ
description We develop two artificial intelligence–based models (Models A and B) to predict time-evolving photospheric magnetic fields with an adjustable time step, ranging up to one solar rotation later. Model A predicts future magnetic field data using three consecutive radial magnetic field data sets, each 12 hr apart. Model B reconstructs evolving magnetic fields using two sets of three consecutive data sets, one from a given date and the other from 27 days later, each 12 hr apart. To train and evaluate our models, we use a Pix2PixCC-based architecture and radial magnetic field data sets from the Solar Dynamics Observatory/Helioseismic and Magnetic Imager during the solar maximum periods of 2012–2016 and 2021–2023. Models A and B successfully generate magnetic field data corresponding to the input prediction time step. Based on quantitative comparison, Model A outperforms the persistence model and performs comparably to the classical surface flux transport model. Model B shows improved performance compared to both models. We also compare predictions for National Oceanic and Atmospheric Administration Active Regions 12673 and 13664, which rapidly emerged and produced powerful solar flares. Notably, both Models A and B can predict increases in total unsigned magnetic fluxes several days in advance if their evolution is already captured in the input data. Model B also reconstructs radial magnetic fluxes that smoothly overlap between the input data and the model outputs. These results demonstrate the ability of our models to generate time-evolving magnetic field data flexibly and capture emerging magnetic fluxes, which are crucial for heliophysics research and space weather forecasting.
format Article
id doaj-art-9b23192030c944288d2ca507aa62ee55
institution DOAJ
issn 2041-8205
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series The Astrophysical Journal Letters
spelling doaj-art-9b23192030c944288d2ca507aa62ee552025-08-20T03:02:51ZengIOP PublishingThe Astrophysical Journal Letters2041-82052025-01-019892L3110.3847/2041-8213/adf5c7Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep LearningHyun-Jin Jeong0https://orcid.org/0000-0003-4616-947XStefaan Poedts1https://orcid.org/0000-0002-1743-0651Haopeng Wang2https://orcid.org/0000-0002-4217-6990Ekatarina Dineva3https://orcid.org/0000-0002-4645-4492Panagiotis Gonidakis4https://orcid.org/0000-0001-5797-0794Francesco Carella5https://orcid.org/0009-0009-3284-4340George Miloshevich6https://orcid.org/0000-0001-9896-1704Luis Linan7https://orcid.org/0000-0002-4014-1815Harim Lee8https://orcid.org/0000-0002-9300-8073Centre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.kr; School of Space Research, Kyung Hee University , Yongin, 17104, Republic of KoreaCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.kr; Institute of Physics, University of Maria Curie-Skłodowska , ul. Radziszewskiego 10, 20-031 Lublin, PolandCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCentre for mathematical Plasma Astrophysics, Department of Mathematics, KU Leuven , Celestijnenlaan 200B, 3001 Leuven, Belgium ; hyun-jin.jeong@kuleuven.be, jeong_hj@khu.ac.krCenter for Solar-Terrestrial Research, New Jersey Institute of Technology , Newark, NJ 07102, USA; Department of Astronomy and Space Science, College of Applied Science, Kyung Hee University , Yongin, 17104, Republic of KoreaWe develop two artificial intelligence–based models (Models A and B) to predict time-evolving photospheric magnetic fields with an adjustable time step, ranging up to one solar rotation later. Model A predicts future magnetic field data using three consecutive radial magnetic field data sets, each 12 hr apart. Model B reconstructs evolving magnetic fields using two sets of three consecutive data sets, one from a given date and the other from 27 days later, each 12 hr apart. To train and evaluate our models, we use a Pix2PixCC-based architecture and radial magnetic field data sets from the Solar Dynamics Observatory/Helioseismic and Magnetic Imager during the solar maximum periods of 2012–2016 and 2021–2023. Models A and B successfully generate magnetic field data corresponding to the input prediction time step. Based on quantitative comparison, Model A outperforms the persistence model and performs comparably to the classical surface flux transport model. Model B shows improved performance compared to both models. We also compare predictions for National Oceanic and Atmospheric Administration Active Regions 12673 and 13664, which rapidly emerged and produced powerful solar flares. Notably, both Models A and B can predict increases in total unsigned magnetic fluxes several days in advance if their evolution is already captured in the input data. Model B also reconstructs radial magnetic fluxes that smoothly overlap between the input data and the model outputs. These results demonstrate the ability of our models to generate time-evolving magnetic field data flexibly and capture emerging magnetic fluxes, which are crucial for heliophysics research and space weather forecasting.https://doi.org/10.3847/2041-8213/adf5c7Solar magnetic fieldsThe SunAstronomy data analysisConvolutional neural networksSolar magnetic flux emergence
spellingShingle Hyun-Jin Jeong
Stefaan Poedts
Haopeng Wang
Ekatarina Dineva
Panagiotis Gonidakis
Francesco Carella
George Miloshevich
Luis Linan
Harim Lee
Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
The Astrophysical Journal Letters
Solar magnetic fields
The Sun
Astronomy data analysis
Convolutional neural networks
Solar magnetic flux emergence
title Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
title_full Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
title_fullStr Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
title_full_unstemmed Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
title_short Prediction of Time-evolving Radial Magnetic Fields on the Solar Surface Using Deep Learning
title_sort prediction of time evolving radial magnetic fields on the solar surface using deep learning
topic Solar magnetic fields
The Sun
Astronomy data analysis
Convolutional neural networks
Solar magnetic flux emergence
url https://doi.org/10.3847/2041-8213/adf5c7
work_keys_str_mv AT hyunjinjeong predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT stefaanpoedts predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT haopengwang predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT ekatarinadineva predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT panagiotisgonidakis predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT francescocarella predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT georgemiloshevich predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT luislinan predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning
AT harimlee predictionoftimeevolvingradialmagneticfieldsonthesolarsurfaceusingdeeplearning