pith. sign in

arxiv: 2605.25634 · v1 · pith:GWUXFCSLnew · submitted 2026-05-25 · 🌌 astro-ph.SR

Machine Learning-based Separation of the He I 10830{AA} Chromospheric Signal: Quantitative Analysis of Chromosphere-Corona Intensity in the Quiet Sun

Huaiming Li , Fangyu Xu , Yi Bi , Zhenyu Jin This is my paper

Pith reviewed 2026-06-29 20:49 UTC · model grok-4.3

classification 🌌 astro-ph.SR
keywords He I 10830Åchromospheric signal separationmachine learning CNNquiet SunEUV correlationmagnetic field polaritychromosphere-corona coupling
0
0 comments X

The pith

Machine learning separates the He I 10830Å chromospheric signal from photospheric background to quantify its negative correlation with EUV intensities in the quiet Sun.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper trains two CNN models on active-region and quiet-Sun data to infer and subtract the photospheric contribution to He I 10830Å images, using TiO images and cross-band information. The cleaned chromospheric signal is then combined with an exponential absorption model for quantitative comparison against simultaneous SDO EUV observations. This separation reveals a strong spatial anticorrelation between the chromospheric intensities and EUV bands, especially R approximately -0.84 with 304Å, together with clear ties to magnetic-field strength and polarity. The work thereby provides a data-driven route to measure vertical coupling across the chromosphere-corona interface without the usual photospheric contamination.

Core claim

After CNN-based removal of the photospheric background, the quiet-Sun He I 10830Å chromospheric component shows a strong negative spatial correlation with EUV radiation (R approx -0.84 in 304Å) and significant layered coupling; strong absorption regions align with strong magnetic fields while 171Å enhancements extend to field edges and mixed-polarity zones, quantifying radiation-intensity relations and heating differences between unipolar and mixed-polarity fields.

What carries the argument

Two CNN models that infer the He I 10830Å photospheric background from TiO and cross-band inputs, followed by an exponential absorption model to isolate the pure chromospheric component.

If this is right

  • The cleaned He I intensities can be used directly to measure chromosphere-corona intensity relationships in the quiet Sun.
  • Strong He I 10830Å absorption areas coincide with strong magnetic fields.
  • 171Å radiative enhancements reach the edges of strong magnetic fields and mixed-polarity regions.
  • The quantified intensity relations distinguish heating characteristics between unipolar and mixed-polarity magnetic configurations.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same separation technique could be tested on other optically thin lines that suffer photospheric contamination to check whether similar anticorrelation patterns appear.
  • If the anticorrelation holds, models of coronal heating must account for suppression or modulation of EUV emission in regions of strong chromospheric absorption.
  • The polarity-dependent differences suggest that future multi-height observations should separate unipolar and mixed-polarity patches when mapping energy transport.

Load-bearing premise

The CNN models recover the true photospheric background in quiet-Sun regions without introducing spatially correlated residuals that would bias the later correlation measurements with EUV bands.

What would settle it

Repeating the correlation analysis after subtracting the photospheric background with an independent, non-ML method and obtaining substantially weaker or positive correlations with the same EUV channels would falsify the reported layered-coupling result.

Figures

Figures reproduced from arXiv: 2605.25634 by Fangyu Xu, Huaiming Li, Yi Bi, Zhenyu Jin.

Figure 1
Figure 1. Figure 1: The first row (A1, A2) displays the high-contrast features of the Active Region dataset. The second row (B1, B2) shows the low-contrast, fine-scale structures of the Quiet Sun dataset. Panels (A1, B1) are the TiO photospheric signals, and Panels (A2, B2) are the observed He i 10830 ˚A signals. All images have a spatial size of 400 × 400 pixels. As shown in [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Training data for Model A from NVST observations. From left to right: 2024-11-25, 2024-11-30, 2025-01-13, 2025-04-25, and 2025-05-12. The first row shows TiO images, and the second row shows 10829 ˚A images representing the photospheric component of He i 10830 ˚A. Model B addresses the challenge posed by QS and weak absorption regions, where the He i 10830 ˚A chromospheric absorption is often too weak to b… view at source ↗
Figure 3
Figure 3. Figure 3: Comparison of Loss for Model A and Model B. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Testing cases of our proposed models: The top row displays results from Model A for active regions, and the bottom row shows results from Model B for the quiet Sun. The columns from left to right represent: (1) the observed intensity used as the training label (Obs); (2) the predicted 10830 ˚A photospheric signal generated by the model (Gen); (3) the ratio map indicating relative deviations (Obs/Gen); (4) … view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of photospheric and chromospheric components. The first row shows the active region; the second row shows the quiet region. Columns 1–2 display the original TiO and He i 10830 ˚A observations; columns 3–4 present the generated photospheric reference and extracted chromospheric absorption maps. All images have a spatial size of 400 × 400 pixels. Animations of the aligned sequences are available i… view at source ↗
Figure 6
Figure 6. Figure 6: Multi-wavelength observations and quantitative analysis of the quiet-Sun region. The images used in this figure are 45-minute time-averaged data. (A) SDO/AIA 171 ˚A image, with the blue box indicating the area of interest for this study; (B) HMI magnetogram; (C) Chromospheric structure separated from the He i 10830 ˚A line; (D)-(F) 304 ˚A, 171 ˚A, and 193 ˚A observations within the study region; (G)-(I) Sc… view at source ↗
Figure 7
Figure 7. Figure 7: The images used in this figure are 45-minute time-averaged data. (A) The He i 10830 ˚A chromospheric signal; (B)-(D) AIA band images at 304 ˚A, 171 ˚A, and 193 ˚A, respectively; (E) The HMI magnetogram. (F)-(H) Scatter plots and linear relationships between the three EUV bands and the He i 10830 ˚A chromospheric signal. Legend Details: In panels (A)-(E), the blue dashed line outlines the strong absorption … view at source ↗
Figure 8
Figure 8. Figure 8: (A) The HMI magnetogram; (B) TheHe i 10830 ˚A chromospheric signal; (C) The AIA 193 ˚A image. The red dashed line outlines the regions where the 193 ˚A intensity is greater than 200 DN/s, representing the strongly heated regions observed in the 193 ˚A band. 4.3. Decoupled He i 10830 ˚A Absorption and EUV Brightening We first examine vertical differences in heating. The red dashed line in [PITH_FULL_IMAGE:… view at source ↗
Figure 9
Figure 9. Figure 9: (A) The He i 10830 ˚A chromospheric signal; (B) The 171 ˚A image. The area below the yellow-green line indicates our primary analysis region. We next examine the spatial structure of heating, as revealed by the He i 10830 ˚A absorption and AIA 171 ˚A emission. Although strong magnetic field regions generally correspond to stronger He i 10830 ˚A absorption, they do not necessarily exhibit strong AIA 171 ˚A … view at source ↗
Figure 10
Figure 10. Figure 10: Basic residual block incorporating Squeeze-and-Excitation (SE) attention mechanism, used in Model A. Input Image 400×400 ResBlock 64ch 400×400 ResBlock 64ch 400×400 ResBlock 64ch 400×400 ResBlock 64ch 400×400 ResBlock 64ch 400×400 ResBlock 64ch 400×400 Downsample (stride 2) 200×200 ResBlock 64ch 200×200 ResBlock 64ch 200×200 ResBlock 64ch 200×200 ResBlock 64ch 200×200 Upsample (Bilinear) 400×400 Concatena… view at source ↗
Figure 11
Figure 11. Figure 11: Detailed architecture of the Model A. The model maps TiO input to 10829 ˚A to leverage the photospheric information of 10829 ˚A as a proxy for the 10830 ˚A photosphere [PITH_FULL_IMAGE:figures/full_fig_p014_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Modified residual block for Model B, with BN and Dropout removed to better fit continuous observations of a specific region rather than cross-date generalization. Input Image (1, 200×200) Fourier Features (1 256ch) Conv1x1/BN/ReLU (256 128ch) ResBlock (128ch) (200×200) ResBlock (128ch) (200×200) ResBlock (128ch) (200×200) ResBlock (128ch) (200×200) ResBlock (128ch) (200×200) ResBlock (128ch) (200×200) Dow… view at source ↗
Figure 13
Figure 13. Figure 13: Detailed architecture of the Model B for Quiet Sun (QS) regions. The model directly maps TiO to 10830 ˚A by treating the weak and rapidly varying chromospheric signals as residuals, thereby establishing a mapping relationship between TiO and the photospheric information of 10830 ˚A [PITH_FULL_IMAGE:figures/full_fig_p015_13.png] view at source ↗
read the original abstract

The He I 10830{\AA} line, a crucial optically thin chromospheric line, is frequently used to study coronal heating and vertical coupling across the chromosphere-corona interface. However, its images are severely contaminated by the strong photospheric background signal, hindering the analysis of fine chromospheric structures. Given the morphological differences between the Active Region (AR) and the Quiet Sun (QS), we proposed separating the He I 10830{\AA} chromospheric signal using two deep learning CNN models. Our model utilizes TiO images and cross-band learning to infer the He I 10830{\AA} photospheric background. The output is combined with an exponential absorption model to achieve quantitative analysis of the pure chromospheric component. Joint analysis of Solar Dynamics Observatory (SDO) data and the separated QS structures reveals a strong spatial negative correlation between chromospheric He I 10830{\AA} intensities(R approx -0.84 in 304{\AA} ), and significant layered coupling with EUV (171, 193, and 304{\AA}) radiation. Furthermore, strong He I 10830{\AA} absorption areas are highly correlated with regions of strong magnetic fields, while 171{\AA} radiative enhancement areas extend to the strong magnetic field edges and the mixed-polarity regions. These findings quantify the radiation intensity relationship between He I 10830{\AA} and EUV bands in the Quiet Sun. It also demonstrates the differences in heating characteristics between unipolar and mixed-polarity magnetic fields.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 0 minor

Summary. The manuscript proposes two CNN models, trained on TiO images and cross-band inputs from SDO data, to separate the photospheric background from He I 10830Å observations in active regions and quiet Sun. An exponential absorption model is applied to the separated signal to enable quantitative analysis of the chromospheric component. The central claims are a strong negative spatial correlation (R ≈ −0.84) between the isolated quiet-Sun He I 10830Å intensities and 304Å EUV, plus layered coupling to 171Å/193Å bands, correlations with magnetic field strength, and differences in heating between unipolar and mixed-polarity regions.

Significance. If the CNN separation is shown to be free of spatially correlated residuals with respect to the EUV bands, the quantitative correlation results would supply useful observational constraints on chromosphere-corona coupling and differential heating in the quiet Sun. The work illustrates an ML route to multi-layer signal disentanglement but currently lacks the validation steps needed to support those constraints.

major comments (3)
  1. [Abstract] Abstract: the reported correlation R ≈ −0.84 is given without error bars, cross-validation statistics, or residual maps, preventing assessment of whether the central quantitative claim is supported by the data.
  2. [Abstract] Abstract: the parameters of the exponential absorption model are not stated, so it is impossible to judge the quantitative analysis of the pure chromospheric component or to reproduce the intensity relationships.
  3. [Separation procedure (CNN models)] Separation procedure (CNN models): no independent validation (synthetic data tests, radiative-transfer forward modeling, or residual–EUV cross-correlation statistics) is described to demonstrate that CNN-inferred photospheric residuals are spatially white with respect to the 171/193/304Å bands used in the correlation step; this assumption is load-bearing for the reported R value and layered-coupling results.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive feedback emphasizing validation and reproducibility. We address each major comment below and commit to revisions that strengthen the quantitative claims without overstating the current manuscript content.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the reported correlation R ≈ −0.84 is given without error bars, cross-validation statistics, or residual maps, preventing assessment of whether the central quantitative claim is supported by the data.

    Authors: We agree that additional statistical support is needed. The revised manuscript will report the correlation with uncertainty estimates obtained via spatial bootstrapping over multiple quiet-Sun patches and will include cross-validation across independent observation dates. A supplementary figure will display residual maps together with their cross-correlation coefficients against the 304 Å, 171 Å and 193 Å bands. revision: yes

  2. Referee: [Abstract] Abstract: the parameters of the exponential absorption model are not stated, so it is impossible to judge the quantitative analysis of the pure chromospheric component or to reproduce the intensity relationships.

    Authors: The exponential absorption model and its fitted coefficients are given in Section 3.2, yet we accept that they should be summarized for immediate reference. The revision will state the key parameters (absorption coefficient and characteristic length) directly in the abstract and will add a compact table listing all model coefficients. revision: yes

  3. Referee: [Separation procedure (CNN models)] Separation procedure (CNN models): no independent validation (synthetic data tests, radiative-transfer forward modeling, or residual–EUV cross-correlation statistics) is described to demonstrate that CNN-inferred photospheric residuals are spatially white with respect to the 171/193/304Å bands used in the correlation step; this assumption is load-bearing for the reported R value and layered-coupling results.

    Authors: We acknowledge the absence of explicit residual–EUV cross-correlation tests in the submitted version. The revised manuscript will include these statistics, showing that the residuals lack significant spatial correlation with the EUV bands. While synthetic-data and full radiative-transfer forward-modeling validations were not performed (owing to the lack of suitable multi-layer training simulations), the added empirical residual analysis directly addresses the load-bearing assumption for the reported correlations. revision: partial

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper trains two CNN models on AR and QS data to infer the photospheric background component of He I 10830Å images from TiO and cross-band inputs, subtracts this background, and applies an exponential absorption model to isolate the chromospheric signal. The headline quantitative result (spatial correlation R ≈ −0.84 with 304 Å and layered coupling with 171/193 Å) is then computed directly as a statistical measure on the separated QS maps. No equation, definition, or step in the provided text reduces this measured correlation to the CNN training inputs, to a fitted parameter renamed as a prediction, or to a self-citation chain. The derivation remains self-contained; the correlation is an independent post-processing statistic on the output maps rather than a quantity forced by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that the CNN-inferred photospheric background is accurate enough that residuals do not drive the reported EUV correlations, plus the validity of the exponential absorption model for quiet-Sun conditions. No free parameters are explicitly listed in the abstract, but the absorption model necessarily contains at least one scale factor.

free parameters (1)
  • exponential absorption scale factor
    Required to convert the separated intensity into a quantitative chromospheric quantity; value not stated in abstract.
axioms (1)
  • domain assumption The photospheric background in He I 10830Å can be accurately predicted from TiO images via CNN cross-band learning without introducing spatially structured errors.
    Invoked when the model output is subtracted to isolate the chromospheric component.

pith-pipeline@v0.9.1-grok · 5830 in / 1492 out tokens · 19688 ms · 2026-06-29T20:49:35.471789+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

41 extracted references · 38 canonical work pages · 1 internal anchor

  1. [1]

    S., & Jones, H

    Andretta, V., Giampapa, M. S., & Jones, H. P. 1995, Irish Astronomical Journal, 22, 177

    1995

  2. [2]

    Andretta, V., & Jones, H. P. 1997, ApJ, 489, 375, doi: 10.1086/304760 Astropy Collaboration, Robitaille, T. P., Tollerud, E. J., et al. 2013, A&A, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f Astropy Collaboration, Price-Whelan, A. M., Lim, ...

    work page doi:10.1086/304760 1997

  3. [3]

    H., Fontenla, J

    Avrett, E. H., Fontenla, J. M., & Loeser, R. 1994, in In: Infrared Solar Physics, Vol. 154, 35–47

    1994

  4. [4]

    2008, ApJ, 677, 742, doi: 10.1086/528680

    Collados, M. 2008, ApJ, 677, 742, doi: 10.1086/528680

    work page doi:10.1086/528680 2008

  5. [5]

    P., Peter, H., Solanki, S

    Chitta, L. P., Peter, H., Solanki, S. K., et al. 2017, ApJS, 229, 4, doi: 10.3847/1538-4365/229/1/4 De Pontieu, B., Tarbell, T., & Erd´ elyi, R. 2003, ApJ, 590, 502, doi: 10.1086/374928

    work page doi:10.3847/1538-4365/229/1/4 2017

  6. [6]

    R., & Cao, W

    Goode, P. R., & Cao, W. 2013, SoPh, 287, 315, doi: 10.1007/s11207-013-0235-6

    work page doi:10.1007/s11207-013-0235-6 2013

  7. [7]

    N., Acton, L

    Handy, B. N., Acton, L. W., Kankelborg, C. C., et al. 1999, SoPh, 187, 229, doi: 10.1023/A:1005166902804

    work page doi:10.1023/a:1005166902804 1999

  8. [8]

    2021, RAA, 21, 105, doi: 10.1088/1674-4527/21/4/105

    Hashim, P., Hong, Z.-X., Ji, H.-S., et al. 2021, RAA, 21, 105, doi: 10.1088/1674-4527/21/4/105

    work page doi:10.1088/1674-4527/21/4/105 2021

  9. [9]

    2024, ApJ, 964, 157, doi: 10.3847/1538-4357/ad2e9d

    Hashim, P., Xu, F., Wang, Y., et al. 2024, ApJ, 964, 157, doi: 10.3847/1538-4357/ad2e9d

    work page doi:10.3847/1538-4357/ad2e9d 2024

  10. [10]

    Deep residual learning for image recognition

    He, K., Zhang, X., Ren, S., & Sun, J. 2016, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1, doi: 10.1109/CVPR.2016.90

    work page doi:10.1109/cvpr.2016.90 2016

  11. [11]

    2022, ApJ, 928, 153, doi: 10.3847/1538-4357/ac590c

    Hong, Z., Wang, Y., & Ji, H. 2022, ApJ, 928, 153, doi: 10.3847/1538-4357/ac590c

    work page doi:10.3847/1538-4357/ac590c 2022

  12. [12]

    2017, RAA, 17, 25, doi: 10.1088/1674-4527/17/3/25

    Hong, Z.-X., Yang, X., Wang, Y., et al. 2017, RAA, 17, 25, doi: 10.1088/1674-4527/17/3/25

    work page doi:10.1088/1674-4527/17/3/25 2017

  13. [13]

    Ji, H., Cao, W., & Goode, P. R. 2012, ApJL, 750, L25, doi: 10.1088/2041-8205/750/1/L25

    work page doi:10.1088/2041-8205/750/1/l25 2012

  14. [14]

    Neupert, W. M. 1993, ApJ, 406, 346, doi: 10.1086/172444

    work page doi:10.1086/172444 1993

  15. [15]

    2025, A&A, 696, A3, doi: 10.1051/0004-6361/202453355

    Leenaarts, J., van Noort, M., de la Cruz Rodr´ ıguez, J., et al. 2025, A&A, 696, A3, doi: 10.1051/0004-6361/202453355

    work page doi:10.1051/0004-6361/202453355 2025

  16. [16]

    The Atmospheric Imaging Assembly (AIA) on the Solar Dynamics Observatory (SDO).Sol

    Lemen, J. R., Title, A. M., Akin, D. J., et al. 2012, SoPh, 275, 17, doi: 10.1007/s11207-011-9776-8

    work page doi:10.1007/s11207-011-9776-8 2012

  17. [17]

    2016, ApJ, 826, 217, doi: 10.3847/0004-637X/826/2/217

    Li, Z., Fang, C., Guo, Y., et al. 2016, ApJ, 826, 217, doi: 10.3847/0004-637X/826/2/217

    work page doi:10.3847/0004-637x/826/2/217 2016

  18. [18]

    P., Leenaarts, J., et al

    Libbrecht, T., Bjørgen, J. P., Leenaarts, J., et al. 2021, A&A, 652, A146, doi: 10.1051/0004-6361/202039788

    work page doi:10.1051/0004-6361/202039788 2021

  19. [19]

    Leenaarts, J., & Ramos, A. A. 2017, A&A, 598, A33, doi: 10.1051/0004-6361/201629266

    work page doi:10.1051/0004-6361/201629266 2017

  20. [20]

    2022, Research in Astronomy and Astrophysics, 22, 095005, doi: 10.1088/1674-4527/ac7cba

    Liu, H., Jin, Z., Xiang, Y., & Ji, K. 2022, Research in Astronomy and Astrophysics, 22, 095005, doi: 10.1088/1674-4527/ac7cba

    work page doi:10.1088/1674-4527/ac7cba 2022

  21. [21]

    2014, RAA, 14, 705, doi: 10.1088/1674-4527/14/6/009

    Liu, Z., Xu, J., Gu, B.-Z., et al. 2014, RAA, 14, 705, doi: 10.1088/1674-4527/14/6/009

    work page doi:10.1088/1674-4527/14/6/009 2014

  22. [22]

    Lowe, D. G. 1999, in Proceedings of the Seventh IEEE International Conference on Computer Vision (ICCV), 1150–1157, doi: 10.1109/ICCV.1999.790410

    work page doi:10.1109/iccv.1999.790410 1999

  23. [23]

    Lowe, D. G. 2004, International Journal of Computer Vision, 60, 91, doi: 10.1023/B:VISI.0000029664.99615.94

    work page doi:10.1023/b:visi.0000029664.99615.94 2004

  24. [24]

    2024, RAA, 24, 055008, doi: 10.1088/1674-4527/ad37f4

    Meng, W.-J., Xu, F.-Y., & Jin, Z.-Y. 2024, RAA, 24, 055008, doi: 10.1088/1674-4527/ad37f4

    work page doi:10.1088/1674-4527/ad37f4 2024

  25. [25]

    Parker, E. N. 1988, ApJ, 330, 474, doi: 10.1086/166485

    work page doi:10.1086/166485 1988

  26. [26]

    PyTorch: An Imperative Style, High-Performance Deep Learning Library

    Paszke, A., Gross, S., Massa, F., et al. 2019, arXiv e-prints, arXiv:1912.01703, doi: 10.48550/arXiv.1912.01703

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1912.01703 2019

  27. [27]

    , Thompson , B.J

    Pesnell, W. D., Thompson, B. J., & Chamberlin, P. C. 2012, SoPh, 275, 3, doi: 10.1007/s11207-011-9841-3

    work page doi:10.1007/s11207-011-9841-3 2012

  28. [28]

    B., Bjelksjo, K., Korhonen, T

    Scharmer, G. B., Bjelksjo, K., Korhonen, T. K., Lindberg, B., & Petterson, B. 2003, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4853, Innovative Telescopes and Instrumentation for Solar Astrophysics, ed. S. L. Keil & S. V. Avakyan, 341–350, doi: 10.1117/12.460377

    work page doi:10.1117/12.460377 2003

  29. [29]

    H., Bush, R

    Schou, J., Scherrer, P. H., Bush, R. I., et al. 2012, SoPh, 275, 229, doi: 10.1007/s11207-011-9842-2

    work page doi:10.1007/s11207-011-9842-2 2012

  30. [30]

    and Mildenhall, Ben and Fridovich-Keil, Sara and Raghavan, Nithin and Singhal, Utkarsh and Ramamoorthi, Ravi and Barron, Jonathan T

    Tancik, M., Srinivasan, P. P., Mildenhall, B., et al. 2020, arXiv e-prints, arXiv:2006.10739, doi: 10.48550/arXiv.2006.10739

    work page doi:10.48550/arxiv.2006.10739 2020

  31. [31]

    E., Del Zanna, G., & Young, P

    Tripathi, D., Mason, H. E., Del Zanna, G., & Young, P. R. 2010, A&A, 518, A42, doi: 10.1051/0004-6361/200913883 Trujillo Bueno, J., ˇStˇ ep´ an, J., & Belluzzi, L. 2012, ApJL, 746, L9, doi: 10.1088/2041-8205/746/1/L9

    work page doi:10.1051/0004-6361/200913883 2010

  32. [32]

    E., Avrett, E

    Vernazza, J. E., Avrett, E. H., & Loeser, R. 1981, ApJS, 45, 635, doi: 10.1086/190731

    work page doi:10.1086/190731 1981

  33. [33]

    2025, ApJ, 982, 161, doi: 10.3847/1538-4357/adbaec

    Wang, Y., Ji, K., Jin, Z., & Liu, H. 2025, ApJ, 982, 161, doi: 10.3847/1538-4357/adbaec

    work page doi:10.3847/1538-4357/adbaec 2025

  34. [34]

    2023, A&A, 672, A173, doi: 10.1051/0004-6361/202244607

    Wang, Y., Zhang, Q., Hong, Z., et al. 2023, A&A, 672, A173, doi: 10.1051/0004-6361/202244607

    work page doi:10.1051/0004-6361/202244607 2023

  35. [35]

    2021, ApJ, 913, 59, doi: 10.3847/1538-4357/abf2b9

    Wang, Y., Zhang, Q., & Ji, H. 2021, ApJ, 913, 59, doi: 10.3847/1538-4357/abf2b9

    work page doi:10.3847/1538-4357/abf2b9 2021

  36. [36]

    2016, ApJL, 820, L13, doi: 10.3847/2041-8205/820/1/L13

    Wang, Y.-M. 2016, ApJL, 820, L13, doi: 10.3847/2041-8205/820/1/L13

    work page doi:10.3847/2041-8205/820/1/l13 2016

  37. [37]

    L., & Fuhr, J

    Wiese, W. L., & Fuhr, J. R. 2009, Journal of Physical and Chemical Reference Data, 38, 565, doi: 10.1063/1.3077727

    work page doi:10.1063/1.3077727 2009

  38. [38]

    S., et al

    Xu, Y., Yang, X., Kerr, G. S., et al. 2022, ApJL, 924, L18, doi: 10.3847/2041-8213/ac447c 17

    work page doi:10.3847/2041-8213/ac447c 2022

  39. [39]

    2016, PhD thesis, New Jersey Institute of Technology

    Zeng, Z. 2016, PhD thesis, New Jersey Institute of Technology

    2016

  40. [40]

    2013, ApJL, 769, L33, doi: 10.1088/2041-8205/769/2/L33

    Zeng, Z., Cao, W., & Ji, H. 2013, ApJL, 769, L33, doi: 10.1088/2041-8205/769/2/L33

    work page doi:10.1088/2041-8205/769/2/l33 2013

  41. [41]

    R., & Cao, W

    Zeng, Z., Chen, B., Ji, H., Goode, P. R., & Cao, W. 2016, ApJL, 819, L3, doi: 10.3847/2041-8205/819/1/L3

    work page doi:10.3847/2041-8205/819/1/l3 2016