pith. sign in

arxiv: 2603.04557 · v2 · pith:YU5SGC45new · submitted 2026-03-04 · 🌌 astro-ph.SR

Energetics and Emission in a Simulated Solar Flare Initialised by a Non-Force Free Magnetic Field

W. Bate , M. Gordovskyy , A. Prasad , A. S. Brun , A. Strugarek , M. V. Sieyra , P. Browning , S. Inoue
show 2 more authors
K. Matsumoto A. Roddanavar
This is my paper

Pith reviewed 2026-05-15 15:54 UTC · model grok-4.3

classification 🌌 astro-ph.SR
keywords solar flaresMHD simulationsnon-force-free fieldsNLFF extrapolationmagnetic energy releaseEUV emissionactive region 11283
0
0 comments X p. Extension
pith:YU5SGC45 Add to your LaTeX paper What is a Pith Number? 
\usepackage{pith}
\pithnumber{YU5SGC45}

Prints a linked pith:YU5SGC45 badge after your title and writes the identifier into PDF metadata. Compiles on arXiv with no extra files. Learn more

The pith

Non-force-free initial fields let flare simulations release twice the magnetic energy and match observed EUV emission more closely than standard force-free setups.

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

Solar flare models are typically initialized with nonlinear force-free field extrapolations from photospheric magnetograms, but this neglects plasma forces and may limit available free energy. The paper runs two otherwise identical resistive MHD simulations of the 2011 September 6 X2.1 flare, differing only in the initial coronal field: one uses a conventional NLFF extrapolation and the other a non-force-free extrapolation. The non-force-free case produces more extensive magnetic restructuring and releases roughly twice the magnetic energy, bringing the total closer to expectations for an X-class event. Synthetic 94 Å emission from the non-force-free run is brighter, more spatially extended, and follows the observed light curve and morphology better than the NLFF run. These outcomes indicate that the choice of initial magnetic configuration can substantially change both the energetics and the observable signatures of the simulated flare.

Core claim

The non-force-free model undergoes more extensive magnetic restructuring and releases approximately twice as much magnetic energy (≈4.4 × 10^31 erg) as the NLFF case (≈2.3 × 10^31 erg), while also producing brighter and more spatially extended synthetic EUV emission that more closely resembles the observed flare morphology and light curve.

What carries the argument

The non-force-free extrapolation of the pre-flare coronal magnetic field, which retains plasma forces omitted by the standard NLFF method and thereby supplies a larger reservoir of free magnetic energy for the flare.

If this is right

  • Flare energy budgets in data-constrained models can be brought into better agreement with X-class expectations by relaxing the force-free assumption.
  • Synthetic EUV emission calculated from non-force-free initial conditions reproduces observed flare morphology and time evolution more faithfully.
  • The degree of magnetic restructuring during the flare depends directly on the amount of free energy stored in the initial coronal field.
  • Assumptions used to construct the pre-flare magnetic field can significantly alter both the dynamics and the observable signatures of simulated flares.

Where Pith is reading between the lines

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

  • The same non-force-free initialization approach could be tested on other flare events to check whether it systematically improves energy release estimates across different active regions.
  • Including plasma forces in the initial field may affect the timing and location of reconnection sites in ways that standard NLFF models miss.
  • This result raises the possibility that many existing flare simulations underestimate total energy because they start from overly constrained magnetic fields.

Load-bearing premise

That the only meaningful difference between the two simulations is the initial magnetic configuration and that the non-force-free extrapolation accurately represents the real pre-flare coronal field without introducing uncontrolled artifacts.

What would settle it

An independent measurement or observation showing that the actual energy released in the 2011 September 6 flare was closer to 2.3 × 10^31 erg than to 4.4 × 10^31 erg, or that the real EUV morphology and light curve match the NLFF run better than the non-force-free run.

Figures

Figures reproduced from arXiv: 2603.04557 by A. Prasad, A. Roddanavar, A. S. Brun, A. Strugarek, K. Matsumoto, M. Gordovskyy, M. V. Sieyra, P. Browning, S. Inoue, W. Bate.

Figure 1
Figure 1. Figure 1: In the left column, the NLFF extrapolation is shown and in the right, the non-force free. The magnetic field lines shown are seeded at the same locations in both, with the field line density proportional to magnetic field strength at the lower boundary. The lower surface in the top row is coloured according to Bz at the lower boundary. The second row shows the value of slogQ calculated at the lower boundar… view at source ↗
Figure 2
Figure 2. Figure 2: In the first panel, the Lorentz force summed in each layer of the extrapolations is shown on a log scale as a function of height. The solid line represents the NLFF model and the dashed line represents the non-force free model. In the second panel, this is recast as average acceleration per layer using force and density. Both dependent variables are presented in normalised units. ∇ × ∇ × ∇ × B + a1∇ × ∇ × … view at source ↗
Figure 3
Figure 3. Figure 3: The left column shows the results of the simulation initialised with the NLFF extrapolation, and the right shows the non-force free simulation. The rows show the following viewpoints: isometric, x, y, z. The red lines show field lines which are initially closed at t = 0, which become open at t = 100tA and are shown in purple. The blue lines show field lines at t = 0 whose end points move by more than 5 Mm … view at source ↗
Figure 4
Figure 4. Figure 4: In the top row, results of the simulation initialised with the NLFF extrapolation is shown and in the middle row, the non-force free. The left column shows the initial configuration, the next two columns show the simulation at t = 20, 40 tA. The magnetic field lines shown are seeded at the same locations in both, with the field line density proportional to magnetic field strength at the lower boundary. The… view at source ↗
Figure 5
Figure 5. Figure 5: Energy over time for both the NLFF (dashed lines) and non-force free (solid lines) simulations. The first plot (cyan) shows magnetic energy, the second (purple) shows kinetic energy, the third (green) shows thermal energy, and the fourth shows all of them together with a log-scale for comparison. both simulations. Also shown are an overlying set of red field lines which open into the upwards purple field l… view at source ↗
Figure 6
Figure 6. Figure 6: A comparison of the synthetic and measured 94 Å emission. In the left and central columns, synthetic emission is shown from the NLFF and non-force free initialised simulations respectively. The top row shows the simulations at t = 10tA, the middle row at t = 15tA, and the bottom row at t = 20tA. The synthetic emission is scaled to the brightest emission of the six panels (at t = 20tA in the non-force free … view at source ↗
Figure 7
Figure 7. Figure 7: The green, red, and purple lines show the syn￾thetic 94 Å light curves for both of the simulations with density scaling factors corresponding to desity at the upper domain boundary of ρt = 2×108 , 9×108 , and 1010 cm−3 re￾spectively. The solid lines corresponds to the non-force free case, and the dashed lines to the NLFF case. The blue line shows the measured AIA 94 Å light curve. this simulation. This dis… view at source ↗
read the original abstract

Solar flare simulations are commonly initialised using non-linear force free field (NLFF) extrapolations derived from photospheric vector magnetograms. However, the force free assumption neglects plasma forces and may limit the available free magnetic energy. In this work, we perform a controlled comparison of two three-dimensional resistive magnetohydrodynamic simulations of the X2.1-class flare that occurred on 2011 September 06 in NOAA Active Region 11283. The simulations differ only in their initial magnetic configuration: one is based on a conventional NLFF extrapolation, while the other employs a non-force free extrapolation. Both models are evolved in an identical stratified atmosphere using the same numerical framework, enabling direct assessment of how the initial magnetic assumptions influence flare dynamics and energetics. We find that the non-force free model undergoes more extensive magnetic restructuring and releases approximately twice as much magnetic energy ($\approx4.4 \times 10^{31}$ erg) as the NLFF case ($\approx2.3 \times 10^{31}$ erg), bringing the energy budget into closer agreement with expectations for X-class flares. Synthetic extreme ultraviolet emission in the 94A channel is computed for both simulations and compared with observations from the Solar Dynamics Observatory. The non-force free model produces a brighter and more spatially extended emission structure that more closely resembles the observed flare morphology and light curve. These results demonstrate that assumptions made in constructing the pre-flare coronal magnetic field can significantly affect flare energetics and observable signatures, and suggest that non-force free extrapolations provide a promising pathway toward more realistic data-constrained flare modelling.

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 / 2 minor

Summary. The manuscript presents a controlled comparison of two 3D resistive MHD simulations of the 2011 September 6 X2.1 flare in AR 11283. The simulations are identical except for the initial coronal magnetic field: one uses a standard NLFF extrapolation from photospheric vector magnetograms, while the other uses a non-force-free extrapolation. Both are evolved in the same stratified atmosphere. The non-force-free run is reported to undergo more extensive reconnection, releasing ~twice the magnetic energy (4.4 × 10^31 erg vs 2.3 × 10^31 erg) and producing synthetic 94 Å EUV emission that more closely matches SDO observations in morphology, brightness, and light curve.

Significance. If the reported energy difference and improved observational match are shown to arise solely from flare reconnection rather than initial-condition relaxation, the work would demonstrate that relaxing the force-free assumption can substantially increase available free energy and improve realism in data-constrained flare models. The direct side-by-side comparison with identical numerics and atmosphere is a strength, as is the quantitative energy numbers and observational comparison.

major comments (3)
  1. [Methods] Methods section: No quantitative checks are reported for initial force balance (e.g., volume-integrated |J × B| / |∇p| or |∇p + ρg| at t = 0) or for magnetic energy dissipation in the first few Alfvén times before the flare trigger. Because only the NLFF field is approximately force-free while the non-force-free field carries net Lorentz forces that cannot be balanced by the given pressure/gravity profiles, immediate relaxation flows and dissipation are expected; without these diagnostics it is impossible to confirm that the factor-of-two energy difference (4.4 vs 2.3 × 10^31 erg) is entirely due to flare reconnection.
  2. [Results] Results, energy-release paragraph: The magnetic energy is stated as ≈4.4 × 10^31 erg (non-force-free) and ≈2.3 × 10^31 erg (NLFF), but the integration volume, height cutoff, and whether the value is total or free energy are not specified. In addition, no error bars or resolution/convergence tests are provided, which is especially important given that the central claim is a quantitative doubling of released energy.
  3. [Section 4] Section 4 (synthetic emission): The computation of 94 Å emission is described only at a high level; missing are the precise temperature response function, density weighting, and line-of-sight integration details. These choices directly affect the claimed better morphological and light-curve match, so they must be documented to allow reproduction and assessment of robustness.
minor comments (2)
  1. [Abstract] Abstract: the ratio 4.4/2.3 ≈ 1.91 is described as 'approximately twice'; a more precise phrasing would avoid slight overstatement.
  2. [Figures] Figure captions: ensure all panels are labeled with simulation name (NLFF vs non-force-free) and time stamps for direct comparison with the observational panels.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their thorough review and valuable comments on our manuscript. We have carefully considered each point and revised the manuscript accordingly to address the concerns about initial conditions, energy calculations, and synthetic observations. Our responses are detailed below.

read point-by-point responses

  1. Referee: [Methods] Methods section: No quantitative checks are reported for initial force balance (e.g., volume-integrated |J × B| / |∇p| or |∇p + ρg| at t = 0) or for magnetic energy dissipation in the first few Alfvén times before the flare trigger. Because only the NLFF field is approximately force-free while the non-force-free field carries net Lorentz forces that cannot be balanced by the given pressure/gravity profiles, immediate relaxation flows and dissipation are expected; without these diagnostics it is impossible to confirm that the factor-of-two energy difference (4.4 vs 2.3 × 10^31 erg) is entirely due to flare reconnection.

    Authors: We acknowledge the importance of these diagnostics. In the revised version, we have added quantitative checks for the initial force balance using the volume-integrated |J × B| / |∇p + ρg| ratio for both initial magnetic fields. Additionally, we report the magnetic energy loss during the pre-flare relaxation phase over the first few Alfvén times. These additions demonstrate that the initial relaxation dissipates only a minor fraction of the energy compared to the flare-related release, supporting that the factor-of-two difference is due to the flare reconnection. revision: yes

  2. Referee: [Results] Results, energy-release paragraph: The magnetic energy is stated as ≈4.4 × 10^31 erg (non-force-free) and ≈2.3 × 10^31 erg (NLFF), but the integration volume, height cutoff, and whether the value is total or free energy are not specified. In addition, no error bars or resolution/convergence tests are provided, which is especially important given that the central claim is a quantitative doubling of released energy.

    Authors: We agree that more details are needed. We have revised the text to specify that the energies represent the change in total magnetic energy integrated over the entire simulation volume (from the photosphere to the top boundary). There is no height cutoff. We clarify these are the released energies (initial minus final). We have included error bars estimated from numerical resolution and added a brief convergence study showing the results are robust to grid resolution. revision: yes

  3. Referee: [Section 4] Section 4 (synthetic emission): The computation of 94 Å emission is described only at a high level; missing are the precise temperature response function, density weighting, and line-of-sight integration details. These choices directly affect the claimed better morphological and light-curve match, so they must be documented to allow reproduction and assessment of robustness.

    Authors: We have expanded the description in Section 4. The 94 Å emission is computed using the AIA temperature response function from the CHIANTI database via SolarSoft, assuming optically thin emission with n^2 weighting for density. The line-of-sight integration is performed by summing the emissivity along the observer's line of sight through the 3D domain. These details are now fully documented in the revised manuscript to ensure reproducibility. revision: yes

Circularity Check

0 steps flagged

No circularity: results are direct simulation outputs benchmarked to independent observations

full rationale

The paper runs two controlled 3D resistive MHD simulations that differ solely in the initial magnetic field (NLFF vs non-force-free extrapolation) while sharing identical atmosphere, numerics, and flare trigger. Magnetic energy release values (4.4 vs 2.3 × 10^31 erg) and synthetic 94 Å emission are computed outputs, not quantities defined in terms of each other or fitted to the target observables. No equations reduce by construction to inputs, no self-citation chain supports a uniqueness claim, and the comparison is externally falsifiable against SDO data. This is the standard non-circular case for data-constrained simulation studies.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on standard resistive MHD equations and the assumption that the two extrapolations differ only in their force-free property; no new physical entities are introduced and no free parameters are fitted to the flare data itself.

axioms (2)
  • standard math The evolution of the coronal plasma is governed by the resistive MHD equations in a stratified atmosphere.
    Invoked throughout the simulation description in the abstract.
  • domain assumption The only controlled difference between the two runs is the initial magnetic field extrapolation method.
    Explicitly stated as the basis for the comparison.

pith-pipeline@v0.9.0 · 5637 in / 1372 out tokens · 55185 ms · 2026-05-15T15:54:51.397102+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

35 extracted references · 35 canonical work pages

  1. [1]

    D., Longbottom, A

    Arber, T. D., Longbottom, A. W., Gerrard, C. L., & Milne, A. M. 2001, Journal of Computational Physics, 171, 151, doi: 10.1006/jcph.2001.6780

    work page doi:10.1006/jcph.2001.6780 2001

  2. [2]

    B., Wilkins, C

    Aslanyan, V., Scott, R. B., Wilkins, C. P., et al. 2024, ApJ, 971, 137, doi: 10.3847/1538-4357/ad55ca

    work page doi:10.3847/1538-4357/ad55ca 2024

  3. [3]

    G., Sun, X., Hoeksema, J

    Bobra, M. G., Sun, X., Hoeksema, J. T., et al. 2014, SoPh, 289, 3549, doi: 10.1007/s11207-014-0529-3 Bárta, M., Büchner, J., Karlický, M., & Kotrč, P. 2011, ApJ, 730, 47, doi: 10.1088/0004-637X/730/1/47

    work page doi:10.1007/s11207-014-0529-3 2014

  4. [4]

    S., Barnes, W

    Community, T. S., Barnes, W. T., Bobra, M. G., et al. 2020, ApJ, 890, 68, doi: 10.3847/1538-4357/ab4f7a

    work page doi:10.3847/1538-4357/ab4f7a 2020

  5. [5]

    Journal of Computational Physics , author =

    Dedner, A., Kemm, F., Kröner, D., et al. 2002, Journal of Computational Physics, 175, 645, doi: 10.1006/jcph.2001.6961 Del Zanna, G., Dere, K. P., Young, P. R., & Landi, E. 2021, ApJ, 909, 38, doi: 10.3847/1538-4357/abd8ce

    work page doi:10.1006/jcph.2001.6961 2002

  6. [6]

    Young, P. R. 1997, Astron. Astrophys. Suppl. Ser., 125, 149, doi: 10.1051/aas:1997368

    work page doi:10.1051/aas:1997368 1997

  7. [7]

    L., & Handy, B

    Freeland, S. L., & Handy, B. N. 1998, SoPh, 182, 497, doi: 10.1023/A:1005038224881

    work page doi:10.1023/a:1005038224881 1998

  8. [8]

    Gary, G. A. 2001, Solar Physics, 203, 71, doi: 10.1023/A:1012722021820

    work page doi:10.1023/a:1012722021820 2001

  9. [9]

    Gary, G. A. 2009, SoPh, 257, 271, doi: 10.1007/s11207-009-9376-z

    work page doi:10.1007/s11207-009-9376-z 2009

  10. [10]

    K., Inoue, S., et al

    Gordovskyy, M., Browning, P. K., Inoue, S., et al. 2020, ApJ, 902, 147, doi: 10.3847/1538-4357/abb60e

    work page doi:10.3847/1538-4357/abb60e 2020

  11. [11]

    K., Kontar, E

    Gordovskyy, M., Browning, P. K., Kontar, E. P., & Bian, N. H. 2014, A&A, 561, A72, doi: 10.1051/0004-6361/201321715

    work page doi:10.1051/0004-6361/201321715 2014

  12. [12]

    Vekstein, G. E. 2023, ApJ, 952, 75, doi: 10.3847/1538-4357/acdb4d

    work page doi:10.3847/1538-4357/acdb4d 2023

  13. [13]

    2008, Sol Phys, 247, 87, doi: 10.1007/s11207-007-9090-7

    Hu, Q., & Dasgupta, B. 2008, Sol Phys, 247, 87, doi: 10.1007/s11207-007-9090-7

    work page doi:10.1007/s11207-007-9090-7 2008

  14. [14]

    P., & Büchner, J

    Hu, Q., Dasgupta, B., Choudhary, D. P., & Büchner, J. 2008, ApJ, 679, 848, doi: 10.1086/587639

    work page doi:10.1086/587639 2008

  15. [15]

    L., Büchner, J., & Gary, G

    Hu, Q., Dasgupta, B., DeRosa, M. L., Büchner, J., & Gary, G. A. 2010, Journal of Atmospheric and Solar-Terrestrial Physics, 72, 219, doi: 10.1016/j.jastp.2009.11.014

    work page doi:10.1016/j.jastp.2009.11.014 2010

  16. [17]

    S., et al

    Inoue, S., Magara, T., Pandey, V. S., et al. 2014, ApJ, 780, 101, doi: 10.1088/0004-637X/780/1/101

    work page doi:10.1088/0004-637x/780/1/101 2014

  17. [18]

    2022, Front

    Kumar, S., Prasad, A., Sarkar, R., & Bhattacharyya, R. 2022, Front. Astron. Space Sci., 9, doi: 10.3389/fspas.2022.1039061

    work page doi:10.3389/fspas.2022.1039061 2022

  18. [19]

    , keywords =

    Lemen, J. R., Title, A. M., Akin, D. J., et al. 2012, SoPh, 275, 17, doi: 10.1007/s11207-011-9776-8 Extrapolations and MHD Simulations 13

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

  19. [20]

    2017, A&A, 604, A76, doi: 10.1051/0004-6361/201629654

    Masson, S., Pariat, E., Valori, G., et al. 2017, A&A, 604, A76, doi: 10.1051/0004-6361/201629654

    work page doi:10.1051/0004-6361/201629654 2017

  20. [21]

    N., & Mikic, Z

    McClymont, A. N., & Mikic, Z. 1994, ApJ, 422, 899, doi: 10.1086/173781

    work page doi:10.1086/173781 1994

  21. [22]

    Mikic, Z., & McClymont, A. N. 1994, in Astronomical Society of the Pacific Conference Series, Vol. 68, Solar Active Region Evolution: Comparing Models with Observations, ed. K. S. Balasubramaniam & G. W. Simon, 225

    work page 1994

  22. [23]

    S., Chamberlin, P

    Moore, C. S., Chamberlin, P. C., & Hock, R. 2014, ApJ, 787, 32, doi: 10.1088/0004-637X/787/1/32

    work page doi:10.1088/0004-637x/787/1/32 2014

  23. [24]

    Assessments in

    Morton, R. 2024, in Oxford Research Encyclopedia of Physics, doi: 10.1093/acrefore/9780190871994.013.10

    work page doi:10.1093/acrefore/9780190871994.013.10 2024

  24. [25]

    S., Bhattacharyya, R., Prasad, A., et al

    Nayak, S. S., Bhattacharyya, R., Prasad, A., et al. 2019, ApJ, 875, 10, doi: 10.3847/1538-4357/ab0a0b

    work page doi:10.3847/1538-4357/ab0a0b 2019

  25. [26]

    , keywords =

    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

  26. [27]

    Nayak, S. S. 2018, ApJ, 860, 96, doi: 10.3847/1538-4357/aac265

    work page doi:10.3847/1538-4357/aac265 2018

  27. [28]

    2020, ApJ, 903, 129, doi: 10.3847/1538-4357/abb8d2

    Prasad, A., Dissauer, K., Hu, Q., et al. 2020, ApJ, 903, 129, doi: 10.3847/1538-4357/abb8d2

    work page doi:10.3847/1538-4357/abb8d2 2020

  28. [29]

    C., et al

    Prasad, A., Kumar, S., Sterling, A. C., et al. 2023, A&A, 677, A43, doi: 10.1051/0004-6361/202346267

    work page doi:10.1051/0004-6361/202346267 2023

  29. [30]

    M., Norton, A

    Schou, J., Borrero, J. M., Norton, A. A., et al. 2012, SoPh, 275, 327, doi: 10.1007/s11207-010-9639-8

    work page doi:10.1007/s11207-010-9639-8 2012

  30. [31]

    J., DeRosa, M

    Schrijver, C. J., DeRosa, M. L., Metcalf, T., et al. 2008, ApJ, 675, 1637, doi: 10.1086/527413

    work page doi:10.1086/527413 2008

  31. [32]

    V., Strugarek, A., Prasad, A., et al

    Sieyra, M. V., Strugarek, A., Prasad, A., et al. 2026, A&A

    work page 2026

  32. [33]

    S., Hornig, G., & Démoulin, P

    Titov, V. S., Hornig, G., & Démoulin, P. 2002, Journal of Geophysical Research: Space Physics, 107, SSH 3, doi: 10.1029/2001JA000278

    work page doi:10.1029/2001ja000278 2002

  33. [34]

    2019, Living Reviews in Solar Physics, 16, 3, doi: 10.1007/s41116-019-0019-7

    Toriumi, S., & Wang, H. 2019, Living Rev Sol Phys, 16, 3, doi: 10.1007/s41116-019-0019-7

    work page doi:10.1007/s41116-019-0019-7 2019

  34. [35]

    Wiegelmann and T

    Wiegelmann, T., & Sakurai, T. 2012, Living Reviews in Solar Physics, 9, 5, doi: 10.12942/lrsp-2012-5

    work page doi:10.12942/lrsp-2012-5 2012

  35. [36]

    L., & Metcalf, T

    DeRosa, M. L., & Metcalf, T. R. 2008, Sol Phys, 247, 249, doi: 10.1007/s11207-008-9130-y

    work page doi:10.1007/s11207-008-9130-y 2008