Pith. sign in

REVIEW 4 major objections 5 minor 25 references

COSMIC-L: A Photometric Catalog of Observed Stars in the Large MagellanIc Cloud

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read COSMIC-L presents a photometric catalog of 57,997,665 stars in the Large Magellanic Cloud, complete to about 21st magnitude with detections to about 22nd magnitude in the g, r, and i bands.

desk verdict A potentially useful 58M-source LMC catalog, but the headline 'deepest' claim rests on an SNR extrapolation that is biased at the faint end and there is no comparison with existing surveys. read the letter →

arxiv 2506.16896 v1 pith:EWNYOIFO submitted 2025-06-20 astro-ph.GA astro-ph.IMastro-ph.SR

A. Franco , A.A. Nucita , F. De Paolis. F. Strafella This is my paper · ORCID
classification astro-ph.GAastro-ph.IMastro-ph.SR
keywords catalogsphotometryLargeMagellanicCloudDECamPSFstellarpopulationsCloudssignal-to-noisedepth
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper presents COSMIC-L, a photometric catalog of 57,997,665 stars in the direction of the Large Magellanic Cloud, built from 23 DECam pointings in the SDSS g, r, and i bands. The authors' central claim is that the catalog is complete to about 21st magnitude and reaches detections near 22nd magnitude in all three bands, making it the deepest ground-based photometric catalog of LMC stars currently available. Roughly 18.7 million sources have reliable photometry in all three bands, enough to build color-magnitude and color-color diagrams that resolve the LMC main sequence, giant branch, red clump, and a foreground population of Milky Way K- and M-type stars. A catalog this deep matters because the LMC hosts standard candles such as RR Lyrae stars and Cepheids, so precise multiband photometry feeds the extragalactic distance scale and searches for microlensing transients.

What carries the argument

The analysis is carried by a PSF-photometry pipeline: SExtractor detects sources and PSFEx builds point-spread-function models, which are then used to measure calibrated magnitudes for every source, with the ATLAS All-Sky Stellar Reference Catalog supplying the photometric zero points through a linear fit per field and band. Depth estimates rest on two auxiliary fits: a linear fit to the rising side of the magnitude histogram, used to define the completeness magnitude where counts fall 5% below the line, and an exponential decay fit to signal-to-noise versus magnitude, extrapolated to the signal-to-noise threshold of 5 to define the limiting magnitude. The color-magnitude and color-color diagrams, built from the 18.7 million three-band sources, are what allow the authors to separate LMC populations from foreground Galactic stars.

What would settle it

Run injection-recovery simulations on the same DECam images: plant artificial stars of known brightness in the g, r, and i frames, re-run the detection pipeline, and measure the magnitude at which the recovery fraction drops to 50%. If that 50% recovery magnitude is significantly brighter than the claimed limiting magnitudes (22.97, 21.64, 21.71 in g, r, i), the extrapolated signal-to-noise depth is not real and the catalog is not as deep as claimed.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that a uniformly processed, PSF-based photometric reduction of 23 public DECam fields yields a catalog of 57,997,665 sources in the LMC region, with 18,676,294 measured in all three gri bands. Photometric calibration against the ATLAS reference catalog gives mean uncertainties of 0.03 mag in g, 0.033 in r, and 0.035 in i. The authors estimate completeness magnitudes of 21.38 (g), 20.82 (r), and 20.79 (i) from the point where the magnitude histogram deviates 5% from a linear fit, and limiting magnitudes of 22.97, 21.64, and 21.71 by extrapolating an exponential fit of signal-to-noise versus magnitude to a signal-to-noise ratio of 5. The catalog is therefore presented as the deepest current ground-based photometric catalog of the Large Magellanic Cloud.

Load-bearing premise

The quoted limiting magnitudes assume that the exponential fit to signal-to-noise versus magnitude remains valid when extrapolated to the signal-to-noise threshold of 5, and that the noise does not flatten or turn over at fainter magnitudes; if the fit is wrong, the claimed depth and the 'deepest ground-based catalog' claim would be unsupported.

Editorial extensions

If this is right

  • COSMIC-L provides a deep, homogeneous gri photometric base for studying the LMC's stellar populations, star formation history, and red clump structure.
  • The catalog's standard-candle stars (RR Lyrae stars and Cepheids) can be used to refine extragalactic distance estimates and, in turn, the local calibration of the Hubble constant.
  • The detection of nearly 57 million sources, including sources to about 25th magnitude, offers a dense target list for transient searches such as microlensing events.
  • The identification of 76,267 foreground K- and M-type stars with distances mostly within 5 kpc provides a sample for mapping the Milky Way's thin and thick disks.
  • Because the catalog is built from public DECam data using a reproducible pipeline, the same strategy can be applied to other Magellanic Cloud fields or neighboring dwarf galaxies.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension the paper does not pursue is a time-series analysis: the 23 fields were observed between February 2018 and January 2020, so the multi-epoch exposures could be stacked or differenced to yield variability and proper-motion information rather than a single-epoch catalog.
  • The completeness and limiting magnitudes could be checked directly with injection-recovery simulations on the same images; the paper explicitly notes it uses indirect methods in the absence of such simulations.
  • Cross-matching COSMIC-L with space-based photometry in the same region would test whether the exponential signal-to-noise extrapolation holds at the faint end, where crowding and noise may depart from the fit.
  • If the faint-end depths hold, the catalog becomes a resource for identifying low-luminosity stellar populations and rare evolved stars that shallower ground-based surveys have missed.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The manuscript presents COSMIC-L, a photometric catalog of 57,997,665 sources detected by the DECam in the g, r, and i bands over 23 fields covering the Large Magellanic Cloud. The authors reduce public DECam images with a pipeline based on SExtractor and PSFEx, calibrate instrumental magnitudes against the ATLAS reference catalog with a per-field linear fit, and release positions, magnitudes, and uncertainties. They report completeness magnitudes of 21.38 (g), 20.82 (r), and 20.79 (i) and limiting magnitudes of 22.97 (g), 21.64 (r), and 21.71 (i), derived from a linear fit to the bright side of the magnitude histograms and from an exponential fit to the signal-to-noise ratio versus magnitude relation extrapolated to S/N = 5. The paper also presents color-magnitude and color-color diagrams and identifies a foreground population of K- and M-type stars using Gaia EDR3 distances. The central claim is that COSMIC-L is the deepest currently available ground-based photometric catalog of LMC stars.

Significance. If the depth claims are reliable, COSMIC-L would be a valuable community resource: it is a large, homogeneous, publicly derived catalog with per-field photometric calibration, and the color-magnitude products demonstrate immediate scientific utility for studying LMC stellar populations and foreground contamination. The authors correctly follow a published pipeline from Franco et al. (2025) and make the data products publicly available. However, the headline depth claims rest on two indirect estimation procedures that are not validated against injection-recovery simulations or external catalogs, and no comparison is made with existing DECam-based LMC surveys such as SMASH. The significance of the catalog is therefore conditional on strengthening the depth analysis.

major comments (4)
  1. [Section 3, Eq. (1), Figure 3, Table 3] The limiting magnitudes in Table 3 are obtained by fitting an exponential decay to the measured SNR-magnitude relation and extrapolating to S/N = 5. This is problematic because the catalog is detection-thresholded: at a given faint magnitude, only sources whose noise fluctuations pushed them above the detection limit enter the sample, so the mean measured SNR of cataloged sources is biased high relative to the true SNR of all sources at that magnitude. The exponential fit to this selected sample therefore likely overestimates the faint-end SNR and yields limiting magnitudes that are too faint. The authors explicitly note that this is done 'in the absence of injection-recovery simulations,' but they do not quantify the resulting bias or provide any external validation. I request an injection-recovery analysis, or a cross-check against a deeper external catalog, and a statement of the propagated uncertainty on the quoted limiting magnitudes.
  2. [Section 3, Figure 5, Table 3] The completeness magnitudes are derived from a linear fit to the rising side of the magnitude histogram with an ad hoc 5% deviation criterion. No uncertainties are quoted for the fitted slopes, intercepts, or the resulting 5% crossing magnitudes, and the choice of the fitting range is not justified. Because the completeness values are a central component of the 'complete to about 21st magnitude' claim, the authors should quote uncertainties, justify the fitting range and criterion, and ideally validate the completeness estimate with injected sources or a comparison to a deeper catalog.
  3. [Section 4 (Conclusions)] The statement that COSMIC-L is 'the deepest ground-based photometric catalog of the LMC stars currently available' is not supported by the analysis as presented. No comparison is made with existing LMC surveys such as SMASH, which also uses DECam, nor with other deep ground-based catalogs. The depth comparison should be made with a consistent limiting-magnitude definition, and the claim should either be substantiated with such a comparison or softened to a description of the present catalog's measured depth.
  4. [Section 3, Figure 3 and Figure 4] The paper states that 'numerous sources are detected down to magnitudes as faint as m 25,' yet the fitted limiting magnitudes are about 22 in all bands. This discrepancy is not discussed. If sources at m 25 are real, the exponential SNR fit is clearly not describing the faint end of the catalog, which further undermines the extrapolated limiting magnitudes; if they are spurious, the catalog's faint-end contamination should be quantified. Either way, this point needs to be addressed.
minor comments (5)
  1. [Abstract] The abstract describes the LMC as containing 'approximately 30 billion stars,' while the catalog contains about 58 million sources; the wording should be clarified to distinguish the total stellar content of the galaxy from the number of cataloged sources.
  2. [Section 2, Figure 1] Figure 1 applies vertical shifts of 0.2 and 1.0 to the r- and i-band distributions for readability, but the shifts are only described in the caption; the same information should be indicated in the figure itself or in a clearer legend.
  3. [Section 3, Table 3] The table gives completeness and limiting magnitudes without uncertainties; since both quantities come from fits, the fit uncertainties should be reported in the table or in the text.
  4. [Section 3] The catalog access information is incomplete: the paper states that the data are publicly available from NOIRLab and describes the catalog columns, but it does not state where the derived COSMIC-L catalog itself can be downloaded, nor does it provide a DOI or a persistent archive link.
  5. [References] The paper cites OGLE and other surveys but does not cite SMASH, the most relevant DECam-based LMC survey for a depth comparison. Adding SMASH and related LMC photometric catalogs to the discussion would strengthen the context for the 'deepest' claim.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the catalog is a data product, its depth estimates are descriptive fits to its own statistics, and the cited prior pipeline work is not load-bearing.

full rationale

The paper presents a photometric catalog; its core product is the catalog itself, not a derived prediction. The photometric calibration uses the external ATLAS All-Sky Stellar Reference Catalog (Tonry et al. 2018) via a linear fit m_ATLAS = C1 m_DECam + C0, and the final magnitudes are calibrated outputs, not re-fitted predictions. Completeness and limiting magnitudes are descriptive characterizations of the catalog's own detection statistics: completeness is read from a linear fit to the rising side of the magnitude histogram at a 5% deviation, and limiting magnitudes are read from an exponential fit to the SNR-magnitude relation at S/N = 5 (Eq. 1, Figure 3). Because these values are fit-derived descriptions of the delivered catalog rather than independent predictions, no quantity is being 'predicted' from itself. The paper explicitly flags the indirectness: 'we estimate the completeness and limiting magnitudes by adopting a widely used method in the absence of injection-recovery simulations' and 'Although based on indirect methods,' which is a limitation on the depth claim, not a circular reduction. The main self-citations (Franco et al. 2023, 2024, 2025) concern prior work on LMC/SMC photometry and the analysis pipeline; the pipeline citation is appropriate because the same reduction strategy is applied, and the catalog's calibration and source extraction are performed on the new DECam data with external tools (SExtractor, PSFEx) and an external reference catalog. No uniqueness theorem, ansatz, or fitted input is imported from the authors' prior work to force the conclusions. The absence of a comparison with other LMC surveys such as SMASH weakens the 'deepest ground-based catalog' claim as a support matter, but does not make the derivation circular.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central catalog depends on standard photometric calibration and detection assumptions. No new physical entities are introduced. The main free parameters are the calibration coefficients, which are standard but essential, and the SNR and histogram fit parameters that define the depth estimates.

free parameters (3)
  • Photometric calibration coefficients C0 and C1 = Reported in Figure 1 per field and band
    Fitted by linear regression of DECam instrumental magnitudes against ATLAS reference magnitudes; these coefficients directly set all catalog magnitudes.
  • Exponential SNR fit amplitude and decay constant = Not reported numerically, fits shown in Figure 3
    The limiting magnitude is the magnitude where the fitted exponential reaches S/N=5; the fit parameters are free and determine the claimed depth.
  • Linear completeness fit slope and intercept = Not reported numerically, fits shown in Figure 5
    The completeness magnitude is defined as the point where the histogram deviates from the extrapolated linear fit; the fit parameters are free but only mildly affect the claim.
assumptions (5)
  • domain assumption The ATLAS All-Sky Stellar Reference Catalog provides accurate magnitudes suitable as a photometric reference.
    Used in Section 2 to calibrate instrumental magnitudes; errors in ATLAS zero points propagate directly into COSMIC-L.
  • standard math The magnitude error to S/N relation S/N = 2.5/(ln(10) sigma_m) holds for all detected sources.
    Used in Section 3, Eq. (1) to convert photometric uncertainties to signal-to-noise ratios.
  • domain assumption The exponential decay function fitted to the SNR-magnitude distribution remains valid beyond the observed range and can be extrapolated to S/N=5.
    Used in Section 3, Figure 3 to derive limiting magnitudes; this is the load-bearing assumption behind the depth claim.
  • domain assumption The histogram completeness method, linear fit and 5% deviation, estimates the true completeness even without injection-recovery simulations.
    Stated in Section 3 as 'a widely used method in the absence of injection-recovery simulations'; the validity is assumed.
  • domain assumption Gaia EDR3 distances corrected with Bailer-Jones priors correctly classify foreground stars.
    Used in Section 3, Figure 7 to identify 76,267 foreground K/M stars; depends on Gaia astrometry and the prior used in the distance correction.

how reviews work

0 comments
Cite this review

Pith. "Pith review of COSMIC-L: A Photometric Catalog of Observed Stars in the Large MagellanIc Cloud." pith.science (2026). https://pith.science/paper/EWNYOIFO

@misc{pith2026250616896,
  author       = {Pith},
  title        = {Pith review of: COSMIC-L: A Photometric Catalog of Observed Stars in the Large MagellanIc Cloud},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/EWNYOIFO}},
  note         = {Machine review of arXiv:2506.16896}
}
abstract

The Magellanic Clouds are two nearby dwarf irregular galaxies whose study can help us in understanding galaxy and stellar evolution. In particular, the Large Magellanic Cloud, the larger one, contains approximately 30 billion stars at various evolutionary stages. In this work, we present an SDSS $gri$-bands photometric analysis based on multiple images acquired with DECam, the Dark Energy Camera, installed on the Blanco telescope at Cerro Tololo Inter-American Observatory (Chile). We performed a full image analysis and photometric calibration, resulting in a photometric catalog named COSMIC-L, consisting of 57,997,665 stars, of which 18,676,294 contain estimates for all three $gri$ magnitudes, resulting in a completeness magnitude of $\simeq 21$ and a limiting magnitude of $\simeq 22$ in all three bands.

Figures

Figures reproduced from arXiv: 2506.16896 by the authors.

Figure 1
Figure 1. Slope (left panel, named C1) and Y-intercept (right panel, named C) distribution obtained after the photometric calibration fit for all three photometric bands, specifically g -band in green, r -band in red and i-band in orange. The two values have been provided by the photometric calibration fit, i.e. mAT L AS = C1 mDEC am +C0. On the x-axis the investigated DECam fields are reported labeled as F*, where the asteri… view at source ↗
Figure 2
Figure 2. Photometric uncertainties as a function of magnitude in the g , r , and i bands (top to bottom) for all sources in the COSMIC￾L catalog. The red horizontal line indicates the mean photometric error in each band. Photometry is based on PSF fitting. 3. THE COSMIC-L PHOTOMETRIC CATALOG The final catalog, named COSMIC-L, includes 57,997,665 sources, of which 18,676,294 exhibit reliable photometry in all three g r i band… view at source ↗
Figure 3
Figure 3. Signal-to-noise ratio (SNR) as a function of magnitude in the g (top), r (middle), and i (bottom) bands. The data are fitted with a decaying exponential function, shown as a yellow line. The adopted SNR threshold (S/N = 5) is marked by the red horizontal dashed line, while the corresponding limiting magnitude is indicated by the green vertical dashed line. Each panel includes an upper-right sub-panel that zooms into… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Spatial distribution of the completeness (left panel) and limiting (right panel) magnitudes across the LMC region, based on the COSMIC-L catalog. The color bars indicate the magnitude values for each of the 23 DECam fields, labeled according to the field numbering conv…
Figure 5
Figure 5. Figure 5: Magnitude distributions of the detected sources in the three photometric bands used in this study: g (green), r (red), and i (yellow). The legend in the top-left corner indicates the number of detections per band. The upper-right sub-panel shows a zoom-in around the hi…
Figure 6
Figure 6. Figure 6: Left panel: color–magnitude diagram (i vs. g − r ) for LMC sources, highlighting the main sequence (left) and the giant branch (right). A dense population with g − r > 1.2 corresponds to foreground K-type and M-type stars in the Milky Way. Right panel: color–color diag…
Figure 7
Figure 7. Figure 7: Distance distribution of the 76,267 stars with g −r > 1.2, identified in [PITH_FULL_IMAGE:figures/full_fig_p006_7.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 21 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter doi edition editor eprint howpublished institution journal key month number organization pages publisher school series title misctitle type volume year version url label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts ...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION format.url url empty "" new.block "" url * "" * if FUNCTION format.eprint eprint empty "" archivePrefix empty "" archivePrefix "arXiv" = new.block " " eprint * " " * new.block " " eprint * " " * if if if FUNCTION format.doi doi empty "" " " doi * " " * if FUNCTION format.pid doi empty eprint empty ur...

  3. [3]

    (" "! _ u q '((U W e

    thebibliography [1] 20pt to REFERENCES 6pt =0pt -12pt 10pt plus 3pt =0pt =0pt =1pt plus 1pt =0pt =0pt -12pt =13pt plus 1pt =20pt =13pt plus 1pt \@M =10000 =-1.0em =0pt =0pt 0pt =0pt =1.0em @enumiv\@empty 10000 10000 `\.\@m \@noitemerr \@latex@warning Empty `thebibliography' environment \@ifnextchar \@reference \@latexerr Missing key on reference command E...

    2021

  4. [4]

    Bailer-Jones, C. A. L., Rybizki, J., Fouesneau, M., Demleitner, M. et al. 2021, AJ, 161, 147

    work page 2021

  5. [5]

    and Arnouts, S

    Bertin, E. and Arnouts, S. 1996, Astron. Astrophys. Suppl. Ser. 117, 393

    work page 1996

  6. [6]

    2011, Astronomical Data Analysis Software and Systems XX

    Bertin, E. 2011, Astronomical Data Analysis Software and Systems XX. ASP Conference Proceedings, 442, 435

    work page 2011

  7. [7]

    & Scarpetta, G

    Calchi Novati, S., Mirzoyan, S., Jetzer, Ph. & Scarpetta, G. 2013, MNRAS, 435, 2, 1582

    work page 2013

  8. [8]

    R., Ivezic, Z., Schlegel, D., Finkbeiner, D

    Covey, K. R., Ivezic, Z., Schlegel, D., Finkbeiner, D. et al. 2007, Astron. J, 134, 2398

    work page 2007

Show all 25 references
  1. [9]

    T., Honscheid, K., Abbott, T

    Flaugher, B., Diehl, H. T., Honscheid, K., Abbott, T. M. C. et al. 2015, Astron. J., 150, 150

    2015

  2. [10]

    A., De Paolis, F., Strafella F

    Franco, A., Nucita A. A., De Paolis, F., Strafella F. et al. 2023, New Astronomy, 103, 102043

    2023

  3. [11]

    A., De Paolis, F

    Franco, A., Nucita A. A., De Paolis, F. & Strafella F. 2024, New Astronomy, 108, 102174

    2024

  4. [12]

    A., De Paolis, F

    Franco, A., Nucita A. A., De Paolis, F. & Strafella F. 2025, ApJS, 277, 33

    2025

  5. [13]

    Gaia Collaboration: Vallenari, A., Brown, A. G. A., Prusti, T., de Bruijne, J. H. J. et al. 2023, A&A, 674, A1, 22

    2023

  6. [14]

    L., Abbott, T., Annis, J

    Honscheid, K., DePoy, D. L., Abbott, T., Annis, J. et al. 2008, 34th International Conference on High Energy Physics, Philadelphia, arxiv:0810.3600

    2008 arXiv

  7. [15]

    D., Newberg, J., Rosner, R., Richards, G

    Lenz, D. D., Newberg, J., Rosner, R., Richards, G. T. et al. 1998, ApJS, 119, 121

    1998

  8. [16]

    K., Kapusta, M

    Mróz, P., Udalski, A., Szymański, M. K., Kapusta, M. et al. 2024, ApJS, 273, 4

    2024

  9. [17]

    K., Soszyński, I

    Mróz, P., Udalski, A., Szymański, M. K., Soszyński, I. et al. 2024, Nature, 632, 8026, 749

    2024

  10. [18]

    Sánchez, E. 2010, J. Phys.: Conf. Ser., 259, 012080

    2010

  11. [19]

    K., Skowron, D

    Soszyński, I., Udalski, A., Szymański, M. K., Skowron, D. et al. 2015, Acta Astron., 65, 297

    2015

  12. [20]

    K., Poleski, R

    Soszyński, I., Udalski, A., Szymański, M. K., Poleski, R. et al. 2015, Acta Astron., 65, 329

    2015

  13. [21]

    K., Wyrzykowski, Ł

    Soszyński, I., Udalski, A., Szymański, M. K., Wyrzykowski, Ł. et al. 2016, Acta Astron., 66, 131

    2016

  14. [22]

    K., Wyrzykowski, Ł

    Soszyński, I., Udalski, A., Szymański, M. K., Wyrzykowski, Ł. et al. 2017, Acta Astron., 67, 103

    2017

  15. [23]

    K., Wyrzykowski, Ł

    Soszyński, I., Udalski, A., Szymański, M. K., Wyrzykowski, Ł. et al. 2018, Acta Astron., 68, 89

    2018

  16. [24]

    L., Denneau, L., Flewelling, H., Heinze, A

    Tonry, J. L., Denneau, L., Flewelling, H., Heinze, A. N. et al. 2018, ApJ, 867, 105

    2018

  17. [25]

    K., Pietrzyński, G

    Udalski, A., Soszyński, I., Szymański, M. K., Pietrzyński, G. et al. 2015, Acta Astron., 65, 341

    2015

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.