Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning
Open magnetic field (OMF) regions are the foot points of the field lines extending to infinity. While they are often considered to coincide with coronal holes (CHs), many studies have reported non-negligible inconsistency between the two. The objective of this study is to develop a deep learning mod...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/ad9a54 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841560729121980416 |
|---|---|
| author | Guan-Han Huang Chia-Hsien Lin |
| author_facet | Guan-Han Huang Chia-Hsien Lin |
| author_sort | Guan-Han Huang |
| collection | DOAJ |
| description | Open magnetic field (OMF) regions are the foot points of the field lines extending to infinity. While they are often considered to coincide with coronal holes (CHs), many studies have reported non-negligible inconsistency between the two. The objective of this study is to develop a deep learning model to predict the OMF regions that coincide with CHs, and to examine the importance of different physical quantities on the prediction. The CHs and OMF regions are both identified from the coronal data computed by a magnetohydrodynamic model developed by Predictive Science. Eight physical quantities, colatitude, magnetic flux expansion factor, magnetic field strength, thermal pressure, electron temperature, electron number density, mass density, and local coronal heating rate, at the OMF footpoints are used as the input to the deep learning model. The results of the test set indicate that the model improves the consistency between CHs and OMF regions from 0.33 to 0.81, and that all eight quantities are similarly important when they are all included in the model although the prediction accuracy would be most affected by excluding the colatitude and/or magnetic field strength. Comparison between the histograms of CH and non-CH open field regions indicates that each quantity alone can be used to distinguish the two within the ranges of values where one histogram is much higher than the other. |
| format | Article |
| id | doaj-art-c1b29bbb8f7d47849199938ef0e8c1c7 |
| institution | Kabale University |
| issn | 1538-4357 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj-art-c1b29bbb8f7d47849199938ef0e8c1c72025-01-03T16:26:00ZengIOP PublishingThe Astrophysical Journal1538-43572025-01-01978212810.3847/1538-4357/ad9a54Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep LearningGuan-Han Huang0https://orcid.org/0000-0003-2297-3279Chia-Hsien Lin1https://orcid.org/0000-0002-5230-0270Department of Space Science and Engineering, National Central University , Taoyuan, Taiwan ; chlin@jupiter.ss.ncu.edu.twDepartment of Space Science and Engineering, National Central University , Taoyuan, Taiwan ; chlin@jupiter.ss.ncu.edu.twOpen magnetic field (OMF) regions are the foot points of the field lines extending to infinity. While they are often considered to coincide with coronal holes (CHs), many studies have reported non-negligible inconsistency between the two. The objective of this study is to develop a deep learning model to predict the OMF regions that coincide with CHs, and to examine the importance of different physical quantities on the prediction. The CHs and OMF regions are both identified from the coronal data computed by a magnetohydrodynamic model developed by Predictive Science. Eight physical quantities, colatitude, magnetic flux expansion factor, magnetic field strength, thermal pressure, electron temperature, electron number density, mass density, and local coronal heating rate, at the OMF footpoints are used as the input to the deep learning model. The results of the test set indicate that the model improves the consistency between CHs and OMF regions from 0.33 to 0.81, and that all eight quantities are similarly important when they are all included in the model although the prediction accuracy would be most affected by excluding the colatitude and/or magnetic field strength. Comparison between the histograms of CH and non-CH open field regions indicates that each quantity alone can be used to distinguish the two within the ranges of values where one histogram is much higher than the other.https://doi.org/10.3847/1538-4357/ad9a54Solar coronaSolar coronal holesSolar magnetic fieldsMagnetohydrodynamicsSolar extreme ultraviolet emissionNeural networks |
| spellingShingle | Guan-Han Huang Chia-Hsien Lin Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning The Astrophysical Journal Solar corona Solar coronal holes Solar magnetic fields Magnetohydrodynamics Solar extreme ultraviolet emission Neural networks |
| title | Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning |
| title_full | Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning |
| title_fullStr | Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning |
| title_full_unstemmed | Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning |
| title_short | Identifying the Coronal Holes in Open Magnetic Field Regions Using Deep Learning |
| title_sort | identifying the coronal holes in open magnetic field regions using deep learning |
| topic | Solar corona Solar coronal holes Solar magnetic fields Magnetohydrodynamics Solar extreme ultraviolet emission Neural networks |
| url | https://doi.org/10.3847/1538-4357/ad9a54 |
| work_keys_str_mv | AT guanhanhuang identifyingthecoronalholesinopenmagneticfieldregionsusingdeeplearning AT chiahsienlin identifyingthecoronalholesinopenmagneticfieldregionsusingdeeplearning |