pith. sign in

arxiv: 2606.06886 · v1 · pith:2JEAJEJ2new · submitted 2026-06-05 · 🌌 astro-ph.SR

Temperature Diagnostics of Chromospheric Fibrils using DKIST/ViSP Observations: K-Means Clustering Approach

Sanjay Gosain This is my paper

Pith reviewed 2026-06-27 21:16 UTC · model grok-4.3

classification 🌌 astro-ph.SR
keywords chromospheric fibrilstemperature structureDKIST/ViSPCa II 854.2 nmK-means clusteringnon-LTE inversionsolar chromospherehydrostatic equilibrium
0
0 comments X

The pith

Chromospheric fibrils cool smoothly by 1000 K along their length but show abrupt temperature jumps across their sides.

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

The paper applies K-means clustering to non-LTE inversions of high-resolution Ca II 854.2 nm spectra from DKIST/ViSP to extract temperature, density, velocity, and microturbulence along and across solar fibrils near a plage. It establishes that temperature falls gradually from the footpoints inward while dropping sharply across the fibril edges, and that denser fibrils correlate with cooler downflows. These patterns matter for understanding how plasma and energy move between the photosphere and corona. The work relies on spectral line inversion under simplifying assumptions to map thermodynamic structure at megameter scales.

Core claim

Non-LTE inversions of DKIST/ViSP Ca II 854.2 nm profiles combined with K-means clustering show that temperature along fibril length decreases smoothly by about 1000 K from hotter footpoints toward the mid-axis, while temperature across the lateral boundary varies more abruptly by several hundred kelvin over a megameter; denser fibrils link to cooler, downflowing plasma whereas less-dense ones do not, and hotter segments exhibit higher microturbulent velocities.

What carries the argument

K-means clustering of thermodynamic parameters retrieved from non-LTE spectral inversions of the Ca II 854.2 nm line.

If this is right

  • Footpoint heating or reduced cooling must operate to maintain the observed along-fibril temperature gradient.
  • Sharp lateral temperature jumps imply well-defined plasma boundaries that separate fibril interiors from surrounding material.
  • The density-temperature-velocity correlation suggests mass loading influences the thermal state of the denser structures.
  • Higher microturbulence in hotter segments points to increased small-scale motions where temperatures are elevated.

Where Pith is reading between the lines

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

  • The observed gradients could be used to test whether wave damping or magnetic reconnection dominates energy deposition along fibrils.
  • Clustering methods similar to those applied here could be extended to other chromospheric lines to build multi-height temperature maps.
  • The results suggest that models of chromospheric fibrils should incorporate density-dependent cooling rates to reproduce the downflow association.

Load-bearing premise

Thermodynamic parameters are derived assuming hydrostatic equilibrium, so the density and temperature values neglect dynamic and magnetic pressure contributions.

What would settle it

Direct comparison of the derived densities and temperatures against simultaneous magnetic field measurements or time-dependent flow observations that reveal large non-hydrostatic support.

Figures

Figures reproduced from arXiv: 2606.06886 by Sanjay Gosain.

Figure 1
Figure 1. Figure 1: The top panel marks the region scanned by the ViSP instrument over a context image of the Sun by SDO AIA in 1600˚A. The bottom panel shows the map of Ca II 854 nm line core intensity and a sub-region marked by a white rectangle, which is chosen for further analysis. 1.1. Observations and Data Reduction The observations analyzed here were obtained with the ViSP instrument (de Wijn et al. 2022) at the DKIST … view at source ↗
Figure 2
Figure 2. Figure 2: The left panel shows a map of cluster labels, color-coded in grayscale (a total of 50 clusters), corresponding to the observed region shown by white rectangle in the lower panel of [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The top row shows the maps of Ca II line core/wing intensity of the observed and the best-fit spectra, as indicated by the title on top of each panel, respectively. Four locations marked by ’x’ symbols and labeled ’1’ through ’4’ on the top right panel, are chosen to show individual observed (black symbols) and best-fit (red curve) spectral profiles. The bottom row shows the models of temperature versus op… view at source ↗
Figure 4
Figure 4. Figure 4: The inverted maps of temperature, gas density, Doppler velocity and microturbulent velocity, averaged over an optical depth range of logτ -4.5 to -5.5, are shown from left to right in the top row, respectively. Bottom row shows histograms of the corresponding maps. the temperature is less than or greater than 7000 K. This temperature is represented by white color on the tem￾perature colorbar and dilineates… view at source ↗
Figure 5
Figure 5. Figure 5: Map of Ca II line core intensity is shown over￾laid with four dashed/dotted lines which mark the location along which we sample the profile of physical parameters. These dashed (red) and dotted (yellow) lines are oriented such that they sample the fibrils across and along their long axis, respectively. These cuts are labeled alphabetically for reference. 1.3.2. Thermodynamic profile along and across fibril… view at source ↗
Figure 6
Figure 6. Figure 6: The panels on the top row show the profile of line core intensity, temperature, density, Doppler velocity and microturbulent velocity across the fibrils, corresponding to the cuts ‘A’ (top left panel) and ‘B’ (top right panel), as labeled in the [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Left panel shows the normalized NSOKP/FTS atlas spectrum at disk center (symbols) and the corresponding SOLIS/VSM spectrum (blue curve). Middle panel shows the quiet-sun spectra at various heliocentric angles. Right panel shows the quiet-sun ViSP/DKIST spectrum (symbols) normalized to the corresponding VSM/SOLIS spectrum (blue curve) at µ=0.85. B. COMPARISON OF CHI SQUARE BETWEEN K-MEANS AND HSRA MODEL INI… view at source ↗
Figure 8
Figure 8. Figure 8: Map fo chi-square statistic when the inversions are initialized with K-means approach (left panel), and when they are initialized with HSRA model for each pixel (right panel) [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The histogram of chi-square statistic when the inversions are initialized using the K-means approach and when they are initialized with HSRA model for each pixel (right panel). REFERENCES Asensio Ramos, A., de la Cruz Rodr´ıguez, J., Mart´ınez Gonz´alez, M. J., & Socas-Navarro, H. 2017, A&A, 599, A133, doi: 10.1051/0004-6361/201629755 Astropy Collaboration, Price-Whelan, A. M., Sip˝ocz, B. M., et al. 2018,… view at source ↗
read the original abstract

The chromosphere is a critical layer of the solar atmosphere situated between the photosphere and the corona. Studying its temperature structure is important to understand the complex dynamics and energy-transfer processes between these layers. We investigate the thermodynamic properties of chromospheric fibrils adjacent to a plage region using high-resolution DKIST/ViSP observations of the Ca II 854.2 nm spectral line. We analyze the spectral profiles with the non-LTE inversion code NICOLE combined with K-means clustering. The high spectral and spatial resolution of the DKIST observations allows us to trace thermodynamic properties temperature, density, line-of-sight velocity, and microturbulent velocity along and across the fibrils. We note that while the thermodynamic parameters are retrieved under the assumption of hydrostatic equilibrium, the resulting density and temperature values should be interpreted with the caveat that dynamic and magnetic terms are neglected. The temperature along the fibril length decreases smoothly by about 1000 K from the hotter footpoints toward the mid-axis. The temperature variation across the lateral boundary of fibrils is more abrupt and can vary by several hundreds of degree Kelvin across a megameter. Denser fibrils tend to be associated with cooler, downflowing plasma, while less-dense fibrils do not show this trend. Furthermore, the hotter parts of the fibrils tend to exhibit higher microturbulent velocities than the cooler parts.

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

1 major / 2 minor

Summary. The paper investigates the thermodynamic properties of chromospheric fibrils using high-resolution DKIST/ViSP observations of the Ca II 854.2 nm line. It applies non-LTE inversions with the NICOLE code under the assumption of hydrostatic equilibrium, followed by K-means clustering to analyze temperature, density, line-of-sight velocity, and microturbulent velocity along and across the fibrils. The main results include a smooth temperature decrease of about 1000 K from the footpoints to the mid-axis, abrupt lateral temperature changes of several hundred K over a megameter, a correlation between denser fibrils and cooler downflowing plasma, and higher microturbulent velocities in hotter regions, with an explicit caveat on the hydrostatic assumption.

Significance. If the results hold under the stated assumptions, the study provides important high-resolution diagnostics of chromospheric fibril thermodynamics, which can inform models of chromospheric heating and dynamics. The application of K-means clustering to inverted parameters is a useful approach for pattern identification in complex observational data. The explicit acknowledgment of the hydrostatic equilibrium caveat is a strength, though it limits the definitiveness of the quantitative claims.

major comments (1)
  1. [Abstract] Abstract: The temperature decrease along the fibril length (~1000 K) and the density-downflow association are derived from NICOLE inversions that assume hydrostatic equilibrium. Given that the abstract notes dynamic and magnetic terms are neglected, and considering observed LOS velocities in fibrils, the paper should demonstrate through additional analysis (e.g., comparison to non-hydrostatic models) that these trends remain robust when the assumption is relaxed.
minor comments (2)
  1. [Abstract] Abstract: The abstract does not include any mention of uncertainty estimates or error bars on the reported temperature variations, densities, or velocities.
  2. The description of the K-means clustering method and the number of clusters used could be expanded for clarity on how the patterns were identified.

Simulated Author's Rebuttal

1 responses · 1 unresolved

We thank the referee for their constructive review and for recognizing the value of the K-means approach and the explicit hydrostatic caveat. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The temperature decrease along the fibril length (~1000 K) and the density-downflow association are derived from NICOLE inversions that assume hydrostatic equilibrium. Given that the abstract notes dynamic and magnetic terms are neglected, and considering observed LOS velocities in fibrils, the paper should demonstrate through additional analysis (e.g., comparison to non-hydrostatic models) that these trends remain robust when the assumption is relaxed.

    Authors: We agree that the hydrostatic-equilibrium assumption is a fundamental limitation of the NICOLE inversions and that the reported temperature drop and density-downflow correlation must be interpreted in that context. The manuscript already states this caveat explicitly in the abstract and in the methods section. Performing the suggested additional analysis—i.e., repeating the inversions or post-processing with non-hydrostatic or MHD-constrained models—would require an entirely different inversion framework and is beyond the scope of the present observational study. We therefore maintain the results as valid under the stated assumptions while clearly flagging their limitations for the reader. revision: no

standing simulated objections not resolved
  • Demonstration that the reported temperature and density trends remain robust when the hydrostatic assumption is relaxed (would require new non-hydrostatic inversions not performed in this work)

Circularity Check

0 steps flagged

No circularity: observational inversion + clustering with explicit caveat on hydrostatic assumption

full rationale

The paper reports results from applying the external NICOLE inversion code to DKIST/ViSP spectra, followed by K-means clustering on the retrieved parameters. The hydrostatic-equilibrium assumption is stated once as a caveat applying to all outputs; it is not derived from or equivalent to any quantity defined inside the paper. No equations, fitted parameters, or self-citations are shown that would make reported temperature/density trends reduce by construction to the paper's own inputs. This matches the default expectation of a self-contained data-analysis study.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central results rest on the NICOLE inversion code outputs and the hydrostatic equilibrium assumption stated in the abstract; no free parameters or invented entities are introduced in the provided text.

axioms (1)
  • domain assumption Hydrostatic equilibrium for retrieval of thermodynamic parameters from spectral inversions
    Explicitly invoked in the abstract as the basis for temperature and density values, with noted caveat on neglected dynamic and magnetic terms.

pith-pipeline@v0.9.1-grok · 5771 in / 1246 out tokens · 21212 ms · 2026-06-27T21:16:02.937333+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

33 extracted references · 26 canonical work pages

  1. [1]

    J., & Socas-Navarro, H

    Asensio Ramos, A., de la Cruz Rodr´ ıguez, J., Mart´ ınez Gonz´ alez, M. J., & Socas-Navarro, H. 2017, A&A, 599, A133, doi: 10.1051/0004-6361/201629755 Astropy Collaboration, Price-Whelan, A. M., Sip˝ ocz, B. M., et al. 2018, AJ, 156, 123, doi: 10.3847/1538-3881/aabc4f Bjørgen, J. P., Leenaarts, J., Rempel, M., et al. 2019, A&A, 631, A33, doi: 10.1051/000...

    work page doi:10.1051/0004-6361/201629755 2017

  2. [2]

    A., & Pereira, T

    Chappell, B. A., & Pereira, T. M. D. 2022, A&A, 658, A182, doi: 10.1051/0004-6361/202142625 da Silva Santos, J. M., Reardon, K., Cauzzi, G., et al. 2023, ApJL, 954, L35, doi: 10.3847/2041-8213/acf21f de la Cruz Rodr´ ıguez, J., Leenaarts, J., Danilovic, S., &

    work page doi:10.1051/0004-6361/202142625 2022

  3. [3]

    2019, A&A, 623, A74, doi: 10.1051/0004-6361/201834464 de la Cruz Rodr´ ıguez, J., & Socas-Navarro, H

    Uitenbroek, H. 2019, A&A, 623, A74, doi: 10.1051/0004-6361/201834464 de la Cruz Rodr´ ıguez, J., & Socas-Navarro, H. 2011, A&A, 527, L8, doi: 10.1051/0004-6361/201016018 de la Cruz Rodr´ ıguez, J., & van Noort, M. 2017, SSRv, 210, 109, doi: 10.1007/s11214-016-0294-8 de Wijn, A. G., Casini, R., Carlile, A., et al. 2022, SoPh, 297, 22, doi: 10.1007/s11207-0...

    work page doi:10.1051/0004-6361/201834464 2019

  4. [4]

    2025, MNRAS, 543, 1303, doi: 10.1093/mnras/staf1567

    Dong, Q., Yan, X., Xue, Z., et al. 2025, MNRAS, 543, 1303, doi: 10.1093/mnras/staf1567

    work page doi:10.1093/mnras/staf1567 2025

  5. [5]

    K., Leenaarts, J., Carlsson, M., & Szydlarski, M

    Druett, M. K., Leenaarts, J., Carlsson, M., & Szydlarski, M. 2022, A&A, 665, A6, doi: 10.1051/0004-6361/202142399

    work page doi:10.1051/0004-6361/202142399 2022

  6. [6]

    2021, A&A, 651, A31, doi: 10.1051/0004-6361/201936910

    Gafeira, R., Orozco Su´ arez, D., Mili´ c, I., et al. 2021, A&A, 651, A31, doi: 10.1051/0004-6361/201936910

    work page doi:10.1051/0004-6361/201936910 2021

  7. [7]

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

    work page doi:10.1023/a:1012722021820 2001

  8. [8]

    W., Kalkofen, W., & Cuny, Y

    Gingerich, O., Noyes, R. W., Kalkofen, W., & Cuny, Y. 1971, SoPh, 18, 347, doi: 10.1007/BF00149057

    work page doi:10.1007/bf00149057 1971

  9. [9]

    V., Carlsson, M., Hansteen, V

    Gudiksen, B. V., Carlsson, M., Hansteen, V. H., et al. 2011, A&A, 531, A154, doi: 10.1051/0004-6361/201116520

    work page doi:10.1051/0004-6361/201116520 2011

  10. [10]

    R., Millman, K

    Harris, C. R., & ... 2020, Nature, doi: 10.1038/s41586-020-2649-2

    work page doi:10.1038/s41586-020-2649-2 2020

  11. [11]

    Henriques, V. M. J., Nelson, C. J., Rouppe van der Voort, L. H. M., & Mathioudakis, M. 2020, A&A, 642, A215, doi: 10.1051/0004-6361/202038538

    work page doi:10.1051/0004-6361/202038538 2020

  12. [12]

    Hunter, J. D. 2007, Computing in Science & Engineering

    2007

  13. [13]

    U., & Nso Staff

    Keller, C. U., & Nso Staff. 1998, in Astronomical Society of the Pacific Conference Series, Vol. 154, Cool Stars, Stellar Systems, and the Sun, ed. R. A. Donahue & J. A. Bookbinder, 636

    1998

  14. [14]

    2023, A&A, 672, A89, doi: 10.1051/0004-6361/202245527

    Kriginsky, M., Oliver, R., & Kuridze, D. 2023, A&A, 672, A89, doi: 10.1051/0004-6361/202245527

    work page doi:10.1051/0004-6361/202245527 2023

  15. [15]

    2024, ApJ, 965, 15, doi: 10.3847/1538-4357/ad2702

    Kuridze, D., Uitenbroek, H., W¨ oger, F., et al. 2024, ApJ, 965, 15, doi: 10.3847/1538-4357/ad2702

    work page doi:10.3847/1538-4357/ad2702 2024

  16. [16]

    L., Furenlid, I., Brault, J., & Testerman, L

    Kurucz, R. L., Furenlid, I., Brault, J., & Testerman, L. 1984, Solar flux atlas from 296 to 1300 nm

    1984

  17. [17]

    2012, ApJ, 749, 136, doi: 10.1088/0004-637X/749/2/136

    Leenaarts, J., Carlsson, M., & Rouppe van der Voort, L. 2012, ApJ, 749, 136, doi: 10.1088/0004-637X/749/2/136

    work page doi:10.1088/0004-637x/749/2/136 2012

  18. [18]

    1982, IEEE transactions on information theory, 28, 129

    Lloyd, S. 1982, IEEE transactions on information theory, 28, 129

    1982

  19. [19]

    1967, in Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics, Vol

    MacQueen, J. 1967, in Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics, Vol. 5, University of California press, 281–298 14 Mart´ ınez-Sykora, J., De Pontieu, B., Carlsson, M., &

    1967

  20. [20]

    2016, ApJL, 831, L1, doi: 10.3847/2041-8205/831/1/L1 Mili´ c, I., & Gafeira, R

    Hansteen, V. 2016, ApJL, 831, L1, doi: 10.3847/2041-8205/831/1/L1 Mili´ c, I., & Gafeira, R. 2020, A&A, 644, A129, doi: 10.1051/0004-6361/201936537

    work page doi:10.3847/2041-8205/831/1/l1 2016

  21. [21]

    2005, SoPh, 229, 13, doi: 10.1007/s11207-005-4081-z

    Neckel, H. 2005, SoPh, 229, 13, doi: 10.1007/s11207-005-4081-z

    work page doi:10.1007/s11207-005-4081-z 2005

  22. [22]

    2011, Journal of Machine Learning Research, 12, 2825

    Pedregosa, F., Varoquaux, G., Gramfort, A., et al. 2011, Journal of Machine Learning Research, 12, 2825

    2011

  23. [23]

    2007, ApJ, 663, 1386, doi: 10.1086/518714

    Pietarila, A., Socas-Navarro, H., & Bogdan, T. 2007, ApJ, 663, 1386, doi: 10.1086/518714

    work page doi:10.1086/518714 2007

  24. [24]

    Flannery, B. P. 1992, Numerical recipes in C (2nd ed.): the art of scientific computing (USA: Cambridge University Press) Quintero Noda, C., Shimizu, T., de la Cruz Rodr´ ıguez, J., et al. 2016, MNRAS, 459, 3363, doi: 10.1093/mnras/stw867

    work page doi:10.1093/mnras/stw867 1992

  25. [25]

    2017, ApJ, 834, 10, doi: 10.3847/1538-4357/834/1/10

    Rempel, M. 2017, ApJ, 834, 10, doi: 10.3847/1538-4357/834/1/10

    work page doi:10.3847/1538-4357/834/1/10 2017

  26. [26]

    R., Warner, M., Keil, S

    Rimmele, T. R., Warner, M., Keil, S. L., et al. 2020, SoPh, 295, 172, doi: 10.1007/s11207-020-01736-7 Sainz Dalda, A., de la Cruz Rodr´ ıguez, J., De Pontieu, B., & Goˇ si´ c, M. 2019, ApJL, 875, L18, doi: 10.3847/2041-8213/ab15d9

    work page doi:10.1007/s11207-020-01736-7 2020

  27. [27]

    A., Penn, M

    Schad, T. A., Penn, M. J., & Lin, H. 2013, ApJ, 768, 111, doi: 10.1088/0004-637X/768/2/111

    work page doi:10.1088/0004-637x/768/2/111 2013

  28. [28]

    2021, arXiv e-prints, arXiv:2101.11445, doi: 10.48550/arXiv.2101.11445

    Socas-Navarro, H., & Asensio Ramos, A. 2021, arXiv e-prints, arXiv:2101.11445, doi: 10.48550/arXiv.2101.11445

    work page doi:10.48550/arxiv.2101.11445 2021

  29. [29]

    2015, A&A, 577, A7, doi: 10.1051/0004-6361/201424860

    Ramos, A., Trujillo Bueno, J., & Ruiz Cobo, B. 2015, A&A, 577, A7, doi: 10.1051/0004-6361/201424860

    work page doi:10.1051/0004-6361/201424860 2015

  30. [30]

    2000, ApJ, 530, 977, doi: 10.1086/308414 The SunPy Community, Barnes, W

    Socas-Navarro, H., Trujillo Bueno, J., & Ruiz Cobo, B. 2000, ApJ, 530, 977, doi: 10.1086/308414 The SunPy Community, Barnes, W. T., Bobra, M. G., et al. 2020, The Astrophysical Journal, 890, 68, doi: 10.3847/1538-4357/ab4f7a

    work page doi:10.1086/308414 2000

  31. [31]

    Thorndike, R. L. 1953, Psychometrika, 18, 267

    1953

  32. [32]

    1989, A&A, 213, 360 Vicente Ar´ evalo, A., Borrero, J

    Uitenbroek, H. 1989, A&A, 213, 360 Vicente Ar´ evalo, A., Borrero, J. M., Mili´ c, I., et al. 2026, A&A, 708, A351, doi: 10.1051/0004-6361/202659286 Wedemeyer-B¨ ohm, S., & Carlsson, M. 2011, A&A, 528, A1, doi: 10.1051/0004-6361/201016186

    work page doi:10.1051/0004-6361/202659286 1989

  33. [33]

    V., & Erd´ elyi, R

    Zaqarashvili, T. V., & Erd´ elyi, R. 2009, SSRv, 149, 355, doi: 10.1007/s11214-009-9549-y

    work page doi:10.1007/s11214-009-9549-y 2009