pith. sign in

arxiv: 2412.01893 · v2 · submitted 2024-12-02 · 🌌 astro-ph.IM · astro-ph.GA

Performance of Photometric Template Fitting for Ultra-High Redshift Galaxies

Thorbj{\o}rn Clausen , Charles L. Steinhardt , Arden Shao , Gaurav Senthil Kumar This is my paper

Pith reviewed 2026-05-23 08:20 UTC · model grok-4.3

classification 🌌 astro-ph.IM astro-ph.GA
keywords photometric redshiftshigh-redshift galaxiesJWSTspectral energy distributiontemplate fittingultra-high redshiftSED modeling
0
0 comments X

The pith

Photometric templates can estimate redshifts for z > 10 galaxies reliably with deep observations but cannot yet predict their full spectral energy distributions.

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

The paper tests several high-redshift template sets developed for JWST-discovered galaxies at z > 10. It compares how well each set recovers photometric redshifts against how well it reproduces the full spectral energy distribution. The templates that perform best for redshifts are not the same ones that best match the overall SED shape. Some templates achieve adequate redshift accuracy when data depth and filter coverage are sufficient, yet every template set tested fails to predict the full SED reliably, and a few galaxies remain poorly fit by all of them.

Core claim

The central claim is that although photometric redshifts can be reliable when given a high enough observational depth and adequate filters, models are not yet able to produce robust astrophysical properties for these ultra-high-redshift galaxies.

What carries the argument

Direct comparison of multiple high-redshift template sets on the same JWST sample, separating performance on photometric redshift recovery from performance on full SED reproduction.

If this is right

  • Photometric redshifts remain usable for z > 10 galaxies when observations reach sufficient depth and include adequate filters.
  • No current template set reliably recovers the full spectral energy distribution of these galaxies.
  • The templates optimal for redshift estimation differ from those optimal for SED fitting.
  • A subset of individual objects resist fitting by every template set examined.

Where Pith is reading between the lines

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

  • Deriving physical quantities such as stellar mass or star-formation rate from these galaxies will require template improvements beyond current redshift-focused models.
  • The mismatch between redshift success and SED failure points to systematic differences between observed spectra and model predictions that are not yet captured.
  • Adding more medium- or narrow-band filters could further separate which template features drive the remaining fit failures.

Load-bearing premise

The tested high-redshift template sets are representative of those used in the literature for z > 10 galaxies and that differences in fit quality primarily reflect model limitations rather than sample selection or data quality issues.

What would settle it

Discovery of a single template set that produces accurate full SED fits for the majority of a large z > 10 sample would falsify the claim that current models cannot produce robust astrophysical properties.

Figures

Figures reproduced from arXiv: 2412.01893 by Arden Shao, Charles L. Steinhardt, Gaurav Senthil Kumar, Thorbj{\o}rn Clausen.

Figure 1
Figure 1. Figure 1: Filters covered in the surveys used and the re [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Objects in the SED subsample. Selected by the cri￾teria above from JADES Data Release 2. Cutouts are selected from the DJA archive and are produced with the GRIZLI utility (Brammer 2023b). Redshifts, catalog NIRCam ID and coordi￾nates are noted. F150W, F277W and F444W was used to make an RGB cutout with equal weighing. The MSA shutter place￾ment is overlaid in orange, with primary source slits marked in ma… view at source ↗
Figure 3
Figure 3. Figure 3: JADES, CEERS, and UNCOVER template set redshift performance mosaics [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Best fit L23 reconstruction for NIRCam￾ID=113585, with zspec = 5.9 and zphot = 5.9. z is in this fit constrained to zspec. The template is fit both photometry (in blue) and a concatenated photometric/spectroscopic dataset (orange). JADES 113585 appears to be poorly represented in template sets optimized for high-redshift galaxies, likely due to emission lines and Lyman break physics [PITH_FULL_IMAGE:figur… view at source ↗
Figure 5
Figure 5. Figure 5: Best fit BlSFH reconstruction for NIRCam￾ID=110996, with zspec = 7.3 and zphot = 7.7. z is in this fit constrained to zspec. The template is fit both to photometry (in blue) and a concatenated photometry/spectroscopy dataset (orange). The template produces significantly stronger emission lines than in the observed SED of JADES 110996, since there are no BlSFH templates with minimal emission lines. the full… view at source ↗
read the original abstract

The James Webb Space Telescope (JWST) has enabled the discovery of a significant population of galaxies at z > 10. Our understanding of the astrophysical properties of these galaxies relies on fitting templates developed using models predicting the differences between these first galaxies and lower-redshift counterparts. In this work, tests are performed on several of these high-redshift template sets in order to determine how successful they are at predicting both photometric redshifts and full spectral energy distributions (SEDs). Our work shows that the best templates for photometric redshift estimation differ from the best templates for predicting the full SED. Overall, some templates perform adequately at photometric redshift estimation, while all are generally poor predictors of the full SED. A few objects in particular are poorly fit by all the template sets tested. We conclude that although photometric redshifts can be reliable when given a high enough observational depth and adequate filters, models are not yet able to produce robust astrophysical properties for these ultra-high-redshift galaxies.

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

2 major / 1 minor

Summary. The manuscript evaluates multiple high-redshift template sets for photometric redshift estimation and full SED fitting on JWST-observed galaxies at z > 10. It reports that the best-performing templates for photo-z differ from those optimal for SED recovery, that some templates yield adequate photo-z but all perform poorly on full SEDs, and that a subset of objects are poorly fit by every tested set. The central conclusion is that photometric redshifts can be reliable given sufficient depth and filter coverage, yet current models remain unable to deliver robust astrophysical properties for ultra-high-redshift galaxies.

Significance. If the quantitative results and sample details hold, the work provides a useful empirical benchmark for the JWST high-z community by showing that template performance is task-dependent and that SED recovery remains a limiting factor. The direct comparison of pre-existing templates against observational data, without circular parameter reduction, is a methodological strength that supports the scoped claim.

major comments (2)
  1. [Abstract] Abstract: the statement that 'some templates perform adequately at photometric redshift estimation, while all are generally poor predictors of the full SED' is presented without any reported metrics (e.g., outlier fraction, normalized median absolute deviation, or reduced chi-squared thresholds), sample size, or error analysis, so it is impossible to assess whether the data support the central claim.
  2. [Methods] Methods/Results (inferred from abstract description): the paper does not specify the exact template libraries tested, the fitting code or priors employed, the selection criteria for the galaxy sample, or the filter set and depth cuts applied; without these details the claim that differences reflect model limitations rather than sample or data-quality effects cannot be evaluated.
minor comments (1)
  1. [Abstract] The abstract would benefit from a brief parenthetical note on the number of objects and redshift range examined.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive comments on the abstract and methods. We address each point below and have revised the manuscript to improve clarity and quantitative support for the claims.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the statement that 'some templates perform adequately at photometric redshift estimation, while all are generally poor predictors of the full SED' is presented without any reported metrics (e.g., outlier fraction, normalized median absolute deviation, or reduced chi-squared thresholds), sample size, or error analysis, so it is impossible to assess whether the data support the central claim.

    Authors: We agree that the abstract would benefit from explicit metrics. In the revised manuscript we have added the sample size, photometric redshift outlier fraction, NMAD, and representative reduced chi-squared values for the SED fits. These quantities are taken directly from the results presented in Sections 3 and 4 and now appear in the abstract to substantiate the central claim. revision: yes

  2. Referee: [Methods] Methods/Results (inferred from abstract description): the paper does not specify the exact template libraries tested, the fitting code or priors employed, the selection criteria for the galaxy sample, or the filter set and depth cuts applied; without these details the claim that differences reflect model limitations rather than sample or data-quality effects cannot be evaluated.

    Authors: The full manuscript already specifies these elements in the Methods section: the template libraries are listed and described, the fitting code and priors are detailed, the galaxy sample selection criteria are given, and the filter set with depth cuts is stated. Because the identical observational dataset is used for every template set, performance differences can be attributed to the models. To address the concern we have added a concise summary table and explicit section references in the revised introduction and results. revision: partial

Circularity Check

0 steps flagged

Empirical comparison with no circular derivation chain

full rationale

The paper performs direct empirical tests of pre-existing template sets on JWST data for photo-z accuracy and SED recovery. No equations, fitted parameters, or self-citations are used to derive results by construction; conclusions rest on reported performance differences between template sets. This matches the default case of a self-contained empirical study against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The work is an empirical test relying on existing templates and JWST observational data; no new free parameters, axioms, or invented entities are introduced based on the abstract.

pith-pipeline@v0.9.0 · 5715 in / 1187 out tokens · 29817 ms · 2026-05-23T08:20:41.910167+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

42 extracted references · 42 canonical work pages

  1. [1]

    J., Conselice, C

    Adams, N. J., Conselice, C. J., Ferreira, L., et al. 2022, Monthly No- tices of the Royal Astronomical Society, 518, 4755 Arrabal Haro, P., Dickinson, M., Finkelstein, S. L., et al. 2023, Nature, 622, 707

    work page 2022

  2. [2]

    B., Finkelstein, S

    Bagley, M. B., Finkelstein, S. L., Koekemoer, A. M., et al. 2023, The Astrophysical Journal Letters, 946, L12

    work page 2023

  3. [3]

    E., et al

    Bezanson, R., Labbe, I., Whitaker, K. E., et al. 2022, The JWST UN- COVER Treasury survey: Ultradeep NIRSpec and NIRCam Ob- serVations before the Epoch of Reionization

    work page 2022

  4. [4]

    J., Curtis-Lake, E., et al

    Boyett, K., Bunker, A. J., Curtis-Lake, E., et al. 2024, Monthly No- tices of the Royal Astronomical Society, stae2430

    work page 2024

  5. [5]

    B., van Dokkum, P

    Brammer, G. B., van Dokkum, P. G., & Coppi, P. 2008, The Astro- physical Journal, 686, 1503

    work page 2008

  6. [6]

    J., Cameron, A

    Bunker, A. J., Cameron, A. J., Curtis-Lake, E., et al. 2024, JADES NIRSpec Initial Data Release for the Hubble Ultra Deep Field: Redshifts and Line Fluxes of Distant Galaxies from the Deepest JWST Cycle 1 NIRSpec Multi-Object Spectroscopy

    work page 2024

  7. [7]

    C., McLure, R

    Carnall, A. C., McLure, R. J., Dunlop, J. S., et al. 2023, Nature, 619, 716–719

    work page 2023

  8. [8]

    2022, The Astrophysical Journal Letters, 938, L15 D’Eugenio, F., Cameron, A

    Castellano, M., Fontana, A., Treu, T., et al. 2022, The Astrophysical Journal Letters, 938, L15 D’Eugenio, F., Cameron, A. J., Scholtz, J., et al. 2024, JADES Data Release 3 – NIRSpec/MSA spectroscopy for 4,000 galaxies in the GOODS fields

    work page 2022

  9. [9]

    T., McLeod, D

    Donnan, C. T., McLeod, D. J., Dunlop, J. S., et al. 2022, Monthly Notices of the Royal Astronomical Society, 518, 6011

    work page 2022

  10. [10]

    2023, Data from The Cosmic Evolution Early Release Science Survey (CEERS)

    Finkelstein, Bagley, Y. 2023, Data from The Cosmic Evolution Early Release Science Survey (CEERS)

    work page 2023

  11. [11]

    A., Dickinson, M., et al

    Fujimoto, S., Haro, P. A., Dickinson, M., et al. 2023, The Astrophys- ical Journal Letters, 949, L25

    work page 2023

  12. [12]

    J., Zitrin, A., Weaver, J

    Furtak, L. J., Zitrin, A., Weaver, J. R., et al. 2023, Monthly Notices of the Royal Astronomical Society, 523, 4568 Geman,S.,Bienenstock,E.,&Doursat,R.1992,NeuralComputation, 4, 1

    work page 2023

  13. [13]

    C., Giavalisco, M., et al

    Guo, Y., Ferguson, H. C., Giavalisco, M., et al. 2013, The Astrophys- ical Journal Supplement Series, 207, 24

    work page 2013

  14. [14]

    N., Johnson, B

    Hainline, K. N., Johnson, B. D., Robertson, B., et al. 2024, The As- trophysical Journal, 964, 71

    work page 2024

  15. [15]

    A., Dickinson, M., Finkelstein, S

    Haro, P. A., Dickinson, M., Finkelstein, S. L., et al. 2023, The Astro- physical Journal Letters, 951, L22

    work page 2023

  16. [16]

    Hayward, C. C. & Smith, D. J. B. 2014, Monthly Notices of the Royal Astronomical Society, 446, 1512

    work page 2014

  17. [17]

    2010, A&A, 523, A31

    Hildebrandt, H., Arnouts, S., Capak, P., et al. 2010, A&A, 523, A31

    work page 2010

  18. [18]

    L., Masters, D., & Salvato, M

    Hovis-Afflerbach, B., Steinhardt, C. L., Masters, D., & Salvato, M. 2021, The Astrophysical Journal, 908, 148

    work page 2021

  19. [19]

    S., Steinhardt, C

    Jermyn, A. S., Steinhardt, C. L., & Tout, C. A. 2018, Monthly Notices of the Royal Astronomical Society, 480, 4265

    work page 2018

  20. [20]

    L., Hutchison, T

    Larson, R. L., Hutchison, T. A., Bagley, M., et al. 2023, The Astro- physical Journal, 958, 141

    work page 2023

  21. [21]

    M., Koekemoer, A., Coe, D., et al

    Lotz, J. M., Koekemoer, A., Coe, D., et al. 2017, The Astrophysical Journal, 837, 97

    work page 2017

  22. [22]

    V., Swain, M

    Mugnai, L. V., Swain, M. R., Estrela, R., & Roudier, G. M. 2024, Monthly Notices of the Royal Astronomical Society, 531, 35–51

    work page 2024

  23. [23]

    2013, The Astrophysical Journal, 775, 11

    Pirzkal, N., Rothberg, B., Ryan, R., et al. 2013, The Astrophysical Journal, 775, 11

    work page 2013

  24. [24]

    J., Robertson, B., Tacchella, S., et al

    Rieke, M. J., Robertson, B., Tacchella, S., et al. 2023, The Astrophys- ical Journal Supplement Series, 269, 16

    work page 2023

  25. [25]

    Romanowsky, A. J. & Fall, S. M. 2012, The Astrophysical Journal Supplement Series, 203, 17

    work page 2012

  26. [26]

    Schreiber, N. M. F., van Dokkum, P. G., Franx, M., et al. 2004, The Astrophysical Journal, 616, 40 Article number, page 13 A&A proofs: manuscript no. main

    work page 2004

  27. [27]

    Smith, D. J. B. & Hayward, C. C. 2015, Monthly Notices of the Royal Astronomical Society, 453, 1597

    work page 2015

  28. [28]

    L., Hensley, H., et al

    Sneppen, A., Steinhardt, C. L., Hensley, H., et al. 2022, The Astro- physical Journal, 931, 57

    work page 2022

  29. [29]

    Steinhardt, C. L. 2016, Monthly Notices of the Royal Astronomical Society, 461, 3432

    work page 2016

  30. [30]

    L., Kokorev, V., Rusakov, V., Garcia, E., & Sneppen, A

    Steinhardt, C. L., Kokorev, V., Rusakov, V., Garcia, E., & Sneppen, A. 2023, The Astrophysical Journal Letters, 951, L40

    work page 2023

  31. [31]

    L., Sneppen, A., Clausen, T., et al

    Steinhardt, C. L., Sneppen, A., Clausen, T., et al. 2024, The Astro- physical Journal, 967, 172 STScI. 2024, NIRCam Imaging Calibration Status

    work page 2024

  32. [32]

    R., Haro, P

    Trump, J. R., Haro, P. A., Simons, R. C., et al. 2023, The Astrophys- ical Journal, 945, 35

    work page 2023

  33. [33]

    2024, Like a candle in the wind: The embers of once aflame, now smouldering galaxies at 5 < z < 8

    Trussler, J., Conselice, C., Adams, N., et al. 2024, Like a candle in the wind: The embers of once aflame, now smouldering galaxies at 5 < z < 8

    work page 2024

  34. [34]

    Turner, J. C. in prep. von Luxburg, U. & Schoelkopf, B. 2008, Statistical Learning Theory:

    work page 2008

  35. [35]

    L., Jespersen, C

    Vujeva, L., Steinhardt, C. L., Jespersen, C. K., et al. 2024, The As- trophysical Journal, 974, 23

    work page 2024

  36. [36]

    2023, The Astrophysical Journal Supplement Series, 270, 12

    Wang, B., Leja, J., Labbé, I., et al. 2023, The Astrophysical Journal Supplement Series, 270, 12

    work page 2023

  37. [37]

    R., Cutler, S

    Weaver, J. R., Cutler, S. E., Pan, R., et al. 2023, The Astrophysical Journal Supplement Series, 270, 7

    work page 2023

  38. [38]

    C., Tacchella, S., Maseda, M

    Williams, C. C., Tacchella, S., Maseda, M. V., et al. 2023, The Astro- physical Journal Supplement Series, 268, 64

    work page 2023

  39. [39]

    2022, The As- trophysical Journal Letters, 942, L9

    Yan, H., Ma, Z., Ling, C., Cheng, C., & Huang, J.-S. 2022, The As- trophysical Journal Letters, 942, L9

    work page 2022

  40. [40]

    B., et al

    Yang, G., Papovich, C., Bagley, M. B., et al. 2021, The Astrophysical Journal, 908, 144

    work page 2021

  41. [41]

    B., et al

    Yang, G., Papovich, C., Bagley, M. B., et al. 2023, The Astrophysical Journal Letters, 956, L12

    work page 2023

  42. [42]

    A., Buat, V., Casey, C

    Zavala, J. A., Buat, V., Casey, C. M., et al. 2023, The Astrophysical Journal Letters, 943, L9 Article number, page 14 Thorbjørn Clausen et al.: Performance of Photometric Template Fitting for Ultra-High Redshift Galaxies Appendix A: Best SED fits Fig. A.1. Best fitting template fit for every object in SED sample Article number, page 15 A&A proofs: manusc...

    work page 2023