REVIEW 2 major objections 2 minor 62 references
Decomposing cool stellar populations with H-band spectral fluctuations: Long-period variable stars in NGC 5128 and carbon stars in NGC 5102
T0 review · 2 major / 2 minor · reviewed 2026-05-22 · grok-4.3
Pith's one-line read Pixel-to-pixel H-band variations isolate mid-M giant spectra plus components from long-period variables and carbon stars.
desk verdict PCA on H-band IFU pixel variations isolates LPV and carbon-star signals in two nearby galaxies, a useful observational trick that matches known features but needs tighter checks on robustness. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
Principal components analysis of the spectra recorded in individual pixels, which isolates the main modes of spectral variation produced by different types of cool giant stars.
What would settle it
A direct comparison showing that the second spectral component in NGC 5128 does not match the H-band spectra of long-period variable stars observed at their coolest phases, or the detection of coherent large-scale gradients in the spectral variations across either galaxy.
Extended reading notes
Core claim
The paper establishes that in NGC 5128 the pixel-to-pixel variation is dominated by a mid-M giant spectrum with prominent CO bandheads and a second component from cool phases of long-period variable stars, while in NGC 5102 the second eigenspectrum shows a strong 1.77 micron C2 bandhead characteristic of carbon stars; these signals arise in near-resolved point-like sources without large-scale variation, consistent with each pixel sampling stars randomly from a common underlying population.
Load-bearing premise
The pixel-to-pixel signals arise only from random sampling of stars from a uniform population with no significant large-scale spatial changes in stellar content.
Editorial extensions
If this is right
- Models of old stellar populations must incorporate the cool phases of long-period variables to reproduce the observed spectral fluctuations.
- The C2 bandhead feature allows carbon stars to be identified in galaxies that contain post-starburst populations through their contribution to pixel variations.
- Integral-field observations yield information on the mix of giant stars that is lost when only the total spectrum of a galaxy is measured.
- Two-epoch observations can provide independent confirmation of variable stars through direct detection of their brightness changes.
Reading between the lines
- The same pixel-fluctuation technique could be applied to other nearby galaxies to map the spatial distribution of variable stars and carbon stars without resolving individual objects.
- Stellar population synthesis codes could be tested and improved by requiring them to reproduce the second principal components extracted from real integral-field data.
- Extending the analysis to additional wavelength ranges might isolate contributions from other stellar types or dust features that are not prominent in the H band.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper applies principal components analysis to H-band integral-field unit spectra of NGC 5128 and NGC 5102 to decompose pixel-to-pixel variations into stellar population components. In the old bulge of NGC 5128 the dominant component matches a mid-M giant spectrum with CO bandheads while a second smoother component is attributed to cool phases of long-period variables (supported by two-epoch MUSE data in an appendix). In NGC 5102 a post-starburst system, the second eigenspectrum instead shows a prominent 1.77 μm C2 bandhead interpreted as arising from carbon stars. The authors argue that the signals originate from near-resolved point sources sampling a common underlying population without large-scale spatial structure.
Significance. If the attributions hold, the work shows that PCA on IFU data can isolate specific cool stellar types (LPVs, carbon stars) beyond what integrated spectra provide, even at distances where classical resolution of individual stars is marginal. The combination of data-driven decomposition, feature matching to known spectra, and direct MUSE confirmation for LPVs is a concrete strength; the approach could be extended to other nearby galaxies to constrain variable-star fractions or carbon-star populations.
major comments (2)
- [§3 / results on NGC 5128] §3 (or equivalent methods/results section): the statement that the second component in NGC 5128 'is not present in predictions from Poisson-sampled models' is central to the LPV interpretation, yet the manuscript provides no quantitative description of how those models were constructed, what IMF or age-metallicity assumptions were adopted, or the statistical threshold used to declare the component absent. This weakens the claim that the observed component must be attributed to LPVs rather than an incomplete model.
- [Introduction / §2 / discussion of assumptions] The weakest assumption (near-resolved point sources with no large-scale variation) is load-bearing for interpreting the PCA components as stellar-type fluctuations rather than spatial gradients. While the paper states that components lack large-scale structure, no explicit test (e.g., spatial autocorrelation of the PC maps or comparison to simulated gradients) is reported to quantify how well this assumption holds at the ~4 Mpc distance.
minor comments (2)
- [Abstract / §2] The abstract and main text would benefit from a brief statement of the wavelength range, spectral resolution, and typical S/N per pixel of the MUSE and IFU data to allow readers to assess the reliability of the 1.77 μm C2 feature identification.
- [Figures 2–4 / results] Figure captions (or the relevant results section) should clarify whether the eigenspectra are normalized in a specific way and whether the reported bandhead strengths include formal uncertainties.
Simulated Author's Rebuttal
We thank the referee for their constructive comments and positive recommendation. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation.
read point-by-point responses
-
Referee: [§3 / results on NGC 5128] §3 (or equivalent methods/results section): the statement that the second component in NGC 5128 'is not present in predictions from Poisson-sampled models' is central to the LPV interpretation, yet the manuscript provides no quantitative description of how those models were constructed, what IMF or age-metallicity assumptions were adopted, or the statistical threshold used to declare the component absent. This weakens the claim that the observed component must be attributed to LPVs rather than an incomplete model.
Authors: We agree that the current description of the Poisson-sampled models lacks sufficient quantitative detail. In the revised manuscript we will expand §3 to specify the model construction (Monte Carlo sampling of stellar luminosities from an old population), the adopted IMF, the age-metallicity assumptions used for NGC 5128, and the precise statistical criterion (e.g., variance threshold or residual comparison) applied to conclude that the second observed component is absent from the models. This addition will clarify the robustness of the LPV interpretation. revision: yes
-
Referee: The weakest assumption (near-resolved point sources with no large-scale variation) is load-bearing for interpreting the PCA components as stellar-type fluctuations rather than spatial gradients. While the paper states that components lack large-scale structure, no explicit test (e.g., spatial autocorrelation of the PC maps or comparison to simulated gradients) is reported to quantify how well this assumption holds at the ~4 Mpc distance.
Authors: We acknowledge that an explicit quantitative test of the no-large-scale-variation assumption would strengthen the paper. In the revision we will add such a test, including computation of the spatial autocorrelation of the principal-component maps together with a comparison against mock data containing imposed large-scale gradients. This will provide a direct metric confirming that the observed signals are consistent with near-resolved point-source sampling at the distance of the galaxies. revision: yes
Circularity Check
No significant circularity detected in data-driven PCA decomposition
full rationale
The paper applies principal components analysis directly to observed H-band integral-field unit spectra of NGC 5128 and NGC 5102. The resulting eigenspectra are interpreted by matching prominent features (CO bandheads at known wavelengths, 1.77 micron C2) to established stellar templates, with the second component in NGC 5128 attributed to LPV cool phases based on absence from Poisson models and supporting MUSE evidence in the appendix. All steps are empirical extractions from the data itself; no parameter is fitted to a subset and then relabeled as a prediction, no uniqueness theorem is imported from self-citation, and no ansatz is smuggled via prior work. The derivation chain remains self-contained against external spectral libraries and direct observations.
Assumptions & free parameters
assumptions (2)
- standard math Principal component analysis separates the dominant independent sources of spectral variation across pixels.
- domain assumption Each pixel samples stars randomly from a shared underlying population with no large-scale spatial gradients in the signal.
Cite this review
Pith. "Pith review of Decomposing cool stellar populations with H-band spectral fluctuations: Long-period variable stars in NGC 5128 and carbon stars in NGC 5102." pith.science (2026). https://pith.science/paper/WLNKOM5R
@misc{pith2026260521588,
author = {Pith},
title = {Pith review of: Decomposing cool stellar populations with H-band spectral fluctuations: Long-period variable stars in NGC 5128 and carbon stars in NGC 5102},
year = {2026},
howpublished = {\url{https://pith.science/paper/WLNKOM5R}},
note = {Machine review of arXiv:2605.21588}
}
read the original abstract
We analyse new H-band integral-field unit observations of two galaxies at ~4 Mpc, using a principal components analysis of pixel spectra to probe their giant star content. In both galaxies, the signals arise in near-resolved point-like sources without large-scale variation, consistent with each pixel sampling stars randomly from a common underlying population. In the (mostly) old bulge of NGC 5128, the observed pixel-to-pixel variation is dominated by a component with a mid-M giant spectrum with prominent CO bandheads. We also recover a smoother second spectral component, apparently driven by contributions from later spectral types. This component is not present in predictions from Poisson-sampled models of old stellar populations; we suggest that it arises from the cool phases of long-period variable stars. (An appendix provides direct evidence for such variables in complementary two-epoch MUSE observations.) In the contrasting galaxy NGC 5102, where a post-starburst stellar population is known to be present, we again find two distinct components. As before, the first component carries the CO bands typical of M-giants. The second eigenspectrum in this younger galaxy shows a strong 1.77 micron C2 bandhead, a feature which is characteristic of carbon stars. Our results highlight the ability of integral field data to access information beyond the total spectrum, even when individual stars cannot be classically resolved.
Figures
Figures from the paper (1 more)
Reference graph
Works this paper leans on
-
[1]
Unmixing Hyperspectral Data , url =
Parra, Lucas and Spence, Clay and Sajda, Paul and Ziehe, Andreas and M\". Unmixing Hyperspectral Data , url =. Advances in Neural Information Processing Systems , editor =
-
[2]
Evolutionary population synthesis: models, analysis of the ingredients and application to high-z galaxies. , keywords =. doi:10.1111/j.1365-2966.2005.09270.x , archivePrefix =. astro-ph/0410207 , primaryClass =
-
[3]
Near-infrared spectroscopy of post-starburst galaxies: a limited impact of TP-AGB stars on galaxy spectral energy distributions. , keywords =. doi:10.1093/mnras/sts126 , archivePrefix =. 1205.4717 , primaryClass =
-
[4]
, year = 1993, month = aug, volume =
Statistical properties of stellar populations and surface-brightness fluctuations in galaxies. , year = 1993, month = aug, volume =
work page 1993
-
[5]
Astronomy and Astrophysics , author =
Modelling simple stellar populations in the near-ultraviolet to near-infrared with the X-shooter Spectral Library (XSL). , keywords =. doi:10.1051/0004-6361/202142387 , archivePrefix =. 2110.10190 , primaryClass =
- [6]
-
[7]
Dominant dark matter and a counter rotating disc: MUSE view of the low luminosity S0 galaxy NGC 5102
Dominant dark matter and a counter-rotating disc: MUSE view of the low-luminosity S0 galaxy NGC 5102. , keywords =. doi:10.1093/mnras/stw2677 , archivePrefix =. 1610.04516 , primaryClass =
-
[8]
Ghosts in the Attic: Mapping the Stellar Content of the S0 Galaxy NGC 5102
Ghosts in the Attic: Mapping the Stellar Content of the S0 Galaxy NGC 5102. , keywords =. doi:10.1088/0004-6256/139/2/680 , archivePrefix =. 0908.2263 , primaryClass =
Show all 62 references
-
[9]
Nature Astronomy , keywords =
Strong spectral features from asymptotic giant branch stars in distant quiescent galaxies. Nature Astronomy , keywords =. doi:10.1038/s41550-024-02391-9 , archivePrefix =. 2403.07414 , primaryClass =
-
[10]
, keywords =
A library of 0.5 to 2.5 mu m spectra of luminous cool stars. , keywords =. doi:10.1051/aas:2000269 , adsurl =
- [11]
- [12]
- [13]
-
[14]
, keywords =
Probing the Stellar Populations and Star Formation History of Early-type Galaxies at 0 < z < 1.1 in the Rest-frame Ultraviolet. , keywords =. doi:10.3847/1538-4357/ad3209 , archivePrefix =. 2403.08301 , primaryClass =
- [15]
- [16]
- [17]
-
[18]
, year = 2020, month = aug, volume =
Evidence for Initial Mass Function Variation in Massive Early-Type Galaxies. , year = 2020, month = aug, volume =. doi:10.1146/annurev-astro-032620-020217 , adsurl =
2020 doi
- [19]
-
[20]
, keywords =
The Enhanced Resolution Imager and Spectrograph for the VLT. , keywords =. doi:10.1051/0004-6361/202346559 , archivePrefix =. 2304.02343 , primaryClass =
-
[21]
Ground-based and Airborne Instrumentation for Astronomy III , year = 2010, editor =
The MUSE second-generation VLT instrument. Ground-based and Airborne Instrumentation for Astronomy III , year = 2010, editor =. doi:10.1117/12.856027 , archivePrefix =. 2211.16795 , primaryClass =
2010 doi
-
[22]
Ground-based and Airborne Instrumentation for Astronomy IX , year = 2022, editor =
HARMONI at ELT: overview of the capabilities and expected performance of the ELT's first light, adaptive optics assisted integral field spectrograph. Ground-based and Airborne Instrumentation for Astronomy IX , year = 2022, editor =. doi:10.1117/12.2628834 , adsurl =
2022 doi
-
[23]
, keywords =
PCA Tomography: how to extract information from data cubes. , keywords =. doi:10.1111/j.1365-2966.2009.14530.x , archivePrefix =. 0901.2266 , primaryClass =
2009 doi
- [24]
- [25]
- [26]
- [27]
- [28]
-
[29]
, keywords =
Surface brightness fluctuation spectrum: a new probe of evolved stars in unresolved stellar populations. , keywords =. doi:10.1093/mnras/sty1840 , adsurl =
- [30]
- [31]
-
[32]
, keywords =
A New Technique for Measuring Extragalactic Distances. , keywords =. doi:10.1086/114847 , adsurl =
- [33]
-
[34]
The TRGB SBF Project. II. Resolving the Virgo Cluster with JWST. , keywords =. doi:10.3847/1538-4357/adb399 , archivePrefix =. 2408.16810 , primaryClass =
- [35]
- [36]
-
[37]
, year = 1963, month = jan, volume =
The A'\^\ 3\ \_\ g\ - X'\^\ 3\ \_\ u\ Band System of the C\_\ 2\ Molecule. , year = 1963, month = jan, volume =. doi:10.1086/147485 , adsurl =
1963 doi
- [38]
- [39]
-
[40]
, keywords =
Peering through the OH forest: a new technique to remove residual sky features from Sloan Digital Sky Survey spectra. , keywords =. doi:10.1111/j.1365-2966.2005.08844.x , archivePrefix =. astro-ph/0501460 , primaryClass =
2005 doi
-
[41]
, keywords =
Nitrogen enhancements 440 Myr after the big bang: supersolar N/O, a tidal disruption event, or a dense stellar cluster in GN-z11?. , keywords =. doi:10.1093/mnras/stad1579 , archivePrefix =. 2302.10142 , primaryClass =
-
[42]
and Plaza, Antonio and Dobigeon, Nicolas and Parente, Mario and Du, Qian and Gader, Paul and Chanussot, Jocelyn , journal=
Bioucas-Dias, José M. and Plaza, Antonio and Dobigeon, Nicolas and Parente, Mario and Du, Qian and Gader, Paul and Chanussot, Jocelyn , journal=. Hyperspectral Unmixing Overview: Geometrical, Statistical, and Sparse Regression-Based Approaches , year=
-
[43]
IEEE signal processing magazine , volume=
Spectral unmixing , author=. IEEE signal processing magazine , volume=. 2002 , publisher=
2002
-
[44]
, keywords =
Probing cool giants in unresolved galaxies using fluctuation eigenspectra: A demonstration using high-resolution MUSE observations of NGC 5128. , keywords =. doi:10.1093/mnras/stab3415 , archivePrefix =. 2111.09317 , primaryClass =
- [45]
-
[46]
, keywords =
Comprehensive Stellar Population Models and the Disentanglement of Age and Metallicity Effects. , keywords =. doi:10.1086/192096 , adsurl =
-
[47]
, keywords =
PARSEC: stellar tracks and isochrones with the PAdova and TRieste Stellar Evolution Code. , keywords =. doi:10.1111/j.1365-2966.2012.21948.x , archivePrefix =. 1208.4498 , primaryClass =
2012 doi
-
[48]
, keywords =
Constraining the thermally pulsing asymptotic giant branch phase with resolved stellar populations in the Large Magellanic Cloud. , keywords =. doi:10.1093/mnras/staa2565 , archivePrefix =. 2008.08595 , primaryClass =
2008 doi
- [49]
-
[50]
and Peletier, R
Sánchez-Blázquez, P. and Peletier, R. F. and Jiménez-Vicente, J. and Cardiel, N. and Cenarro, A. J. and Falcón-Barroso, J. and Gorgas, J. and Selam, S. and Vazdekis, A. , title =. Monthly Notices of the Royal Astronomical Society , volume =. 2006 , month =. doi:10.1111/j.1365-...
2006 doi
-
[51]
, year = 1955, month = jan, volume =
The Luminosity Function and Stellar Evolution. , year = 1955, month = jan, volume =. doi:10.1086/145971 , adsurl =
1955 doi
-
[52]
The Open Journal of Astrophysics , keywords =
Early Bright Galaxies from Helium Enhancements in High-Redshift Star Clusters. The Open Journal of Astrophysics , keywords =. doi:10.33232/001c.126253 , archivePrefix =. 2410.14846 , primaryClass =
-
[53]
, keywords =
Strong CO absorption features in massive ETGs. , keywords =. doi:10.1093/mnras/stac471 , archivePrefix =. 2202.08651 , primaryClass =
- [54]
- [55]
-
[56]
, keywords =
Puzzling radial gradients of K-band absorption features in the giant elliptical galaxy M87. , keywords =. doi:10.1051/0004-6361/202554315 , archivePrefix =. 2506.05986 , primaryClass =
- [57]
- [58]
-
[59]
Galaxies , keywords =
Applications of Stellar Population Synthesis in the Distant Universe. Galaxies , keywords =. doi:10.3390/galaxies8010006 , archivePrefix =. 2005.01759 , primaryClass =
2005 doi
- [60]
- [61]
- [62]
Reviewed May 22, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.