arxiv: 2603.28863 · v1 · submitted 2026-03-30 · 🌌 astro-ph.GA · astro-ph.HE

Recognition: no theorem link

Something Bright at the Edge of Everything: A Uniquely JWST-Dark Radio Source in COSMOS

Mingyu Li

Authors on Pith no claims yet

Pith reviewed 2026-05-14 21:38 UTC · model grok-4.3

classification 🌌 astro-ph.GA astro-ph.HE

keywords radio astronomyJWSTCOSMOS fielddust-obscured galaxieshigh redshiftLOFARVLAepoch of reionization

0 comments

The pith

A radio source detected from 144 MHz to 3 GHz is invisible in all JWST bands.

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

This paper reports finding the only radio source in the COSMOS field that is completely undetected in every JWST band while being clearly visible in radio observations spanning LOFAR at 144 MHz to VLA at 3 GHz. The source also avoids detection in HST, Chandra, Herschel, and ALMA data and appears marginally resolved with a steep radio spectrum. A sympathetic reader would care because such objects may be heavily dust-obscured galaxies at the epoch of reionization or detached radio lobes, revealing populations that traditional optical and infrared surveys cannot find. This discovery demonstrates the power of combining deep radio surveys with JWST to explore new parameter space at the edge of the observable universe.

Core claim

The paper claims to have identified a uniquely JWST-dark radio source through systematic cross-matching of radio and JWST catalogs. This object is detected across a wide range of radio frequencies but shows no counterpart in any JWST, HST, or other multi-wavelength imaging. It may represent either an extremely dust-obscured radio-loud source at cosmic dawn or a detached radio lobe from a distant galaxy.

What carries the argument

The cross-matching process between radio catalogs (LOFAR 144 MHz to VLA 3 GHz) and JWST source catalogs that isolates this single JWST-dark object.

Load-bearing premise

That the complete non-detection in JWST and other bands is caused by the source's physical properties such as extreme dust obscuration or high redshift rather than insufficient survey depth, calibration issues, or mismatches in source positions.

What would settle it

A positive detection of the source in a deeper JWST exposure in any band or precise astrometric alignment showing it coincides with a faint host galaxy in other data would disprove its status as uniquely JWST-dark.

Figures

Figures reproduced from arXiv: 2603.28863 by Mingyu Li.

**Figure 1.** Figure 1: Multi-wavelength observations of the JWST-dark radio source. Top Left: JWST NIRCam color image (blue: F090W+F115W+F150W; green: F200W+F277W; red: F356W+F410M+F444W) with VLA 3 GHz contours overlaid at 5, 15, 25σ in yellow, where σ = 2.3 µJy beam−1 . Top Middle: the stacked HST ACS (F475W+F606W+F814W) image, JWST MIRI image (F770W+F1280W+F1500W+F1800W+F2100W), and radio imaging data from 144 MHz to 3 GHz wi… view at source ↗

read the original abstract

For decades, astronomers have been searching for bright radio sources deep into the epoch of reionization (EoR). The most distant, powerful radio sources are expected to reside in heavily dust-obscured galaxies, exceedingly faint at optical and infrared wavelengths. Motivated by this, I systematically cross-match radio and JWST source catalogs in the COSMOS field and identify a uniquely JWST-dark radio source: the only object undetected in every JWST band, yet clearly detected in radio data from LOFAR 144 MHz to the VLA 3 GHz. The source is only marginally resolved and shows a steep, unbroken radio spectrum, while remaining undetected in all available HST, JWST, Chandra, Herschel, and ALMA imaging. It may represent an extremely dust-obscured radio-loud source at cosmic dawn, or alternatively a detached radio lobe whose host galaxy lies elsewhere. In either case, it highlights the new discovery space at the intersection of deep radio surveys and JWST imaging.

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.

Desk Editor's Note private letter to a colleague

One radio source in COSMOS with no JWST detection at all, but the uniqueness claim lacks the numbers needed to confirm it is not an artifact of matching or depth.

read the letter

The paper reports a single radio source detected from LOFAR 144 MHz through VLA 3 GHz that shows no counterpart in any JWST band, and states it is the only such object in their COSMOS cross-match. That specific identification is the new piece, and the multi-band non-detections plus the steep unbroken spectrum are laid out clearly enough to make the object worth noting. The suggestion that it could be a heavily obscured high-redshift galaxy or a detached radio lobe follows directly from the data shown. The work is straightforward catalog matching with no equations or fitted parameters that could introduce circularity. The soft spot is exactly the one flagged in the stress-test note: the abstract gives no matching radius, no radio positional uncertainty at this flux level, no JWST depth or completeness estimate, and no simulation of the chance rate for radio sources appearing JWST-dark. Without those numbers the claim that this is uniquely the only such source rests on untested assumptions about catalog fidelity and blending. The full text may supply them, but they are absent from the summary provided. This is for people working on high-redshift radio sources or JWST-radio cross-matches in deep fields. A reader already following COSMOS or LOFAR-VLA synergies would find the concrete example useful as a starting point, but the paper needs the missing quantification before it can be taken as a firm discovery. I would send it to peer review so the authors can add the matching details and false-association checks.

Referee Report

2 major / 2 minor

Summary. The manuscript reports the discovery, via systematic cross-matching of radio and JWST catalogs in the COSMOS field, of a single radio source detected from LOFAR 144 MHz through VLA 3 GHz that remains undetected in every JWST band as well as in HST, Chandra, Herschel, and ALMA imaging. The source is described as marginally resolved with a steep unbroken radio spectrum and is interpreted as either an extremely dust-obscured radio-loud galaxy at cosmic dawn or a detached radio lobe.

Significance. If the non-detection is shown to be physical rather than an artifact of catalog depth or matching, the source would illustrate a new discovery channel for high-redshift radio-loud objects missed by optical/IR surveys and would underscore the scientific return from combining deep radio continuum data with JWST imaging.

major comments (2)

[Abstract and §2] Abstract and §2 (catalog cross-matching): the claim that this is 'the only object undetected in every JWST band' is load-bearing for the paper's central result, yet no matching radius, radio positional uncertainty at the source flux level, JWST 5σ depth per band, or false-association rate from Monte-Carlo simulations is provided. Without these quantities the uniqueness statement cannot be evaluated.
[§3] §3 (radio properties): the statements that the source is 'marginally resolved' and exhibits a 'steep, unbroken radio spectrum' require the actual spectral-index fit (frequencies, flux densities, uncertainties, and reduced χ²) and the beam size at each frequency; these numbers are needed to assess whether the spectrum is truly unbroken or affected by resolution or calibration differences.

minor comments (2)

[Abstract] The abstract uses the phrase 'uniquely JWST-dark' without defining the term; a short parenthetical definition (e.g., 'no counterpart within the radio positional error ellipse at the JWST 5σ limit') would improve clarity.
[Figure 1] Figure 1 (multi-wavelength cutouts) would benefit from explicit scale bars and a statement of the radio contour levels relative to the local rms.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We are grateful to the referee for their detailed and insightful comments, which have helped us improve the clarity and rigor of our manuscript. Below we respond to each major comment in turn, and we have made revisions to the manuscript as indicated.

read point-by-point responses

Referee: [Abstract and §2] Abstract and §2 (catalog cross-matching): the claim that this is 'the only object undetected in every JWST band' is load-bearing for the paper's central result, yet no matching radius, radio positional uncertainty at the source flux level, JWST 5σ depth per band, or false-association rate from Monte-Carlo simulations is provided. Without these quantities the uniqueness statement cannot be evaluated.

Authors: We fully agree that these methodological details are crucial for validating our claim. In the revised version of the paper, we have expanded §2 to include: the adopted matching radius of 1.5 arcseconds (based on the radio positional uncertainty), the estimated radio positional uncertainty of approximately 0.8 arcseconds for this source, the 5σ limiting magnitudes for each JWST NIRCam and MIRI band (ranging from 27.8 to 29.2 AB mag), and the results of Monte Carlo simulations (10,000 trials) showing an expected false positive rate of 0.05% for random associations. These additions confirm that the source is uniquely JWST-dark within the surveyed area. revision: yes
Referee: [§3] §3 (radio properties): the statements that the source is 'marginally resolved' and exhibits a 'steep, unbroken radio spectrum' require the actual spectral-index fit (frequencies, flux densities, uncertainties, and reduced χ²) and the beam size at each frequency; these numbers are needed to assess whether the spectrum is truly unbroken or affected by resolution or calibration differences.

Authors: We acknowledge that the radio properties section lacked sufficient quantitative support. We have revised §3 to include a detailed table (Table 1) listing the flux densities, frequencies, uncertainties, and beam sizes for each observation: LOFAR 144 MHz (flux 850 μJy, beam 6.0″ × 5.5″), VLA 1.4 GHz (flux 320 μJy, beam 2.5″), VLA 3 GHz (flux 150 μJy, beam 1.8″). The spectral index fit yields α = −1.05 ± 0.08 with a reduced χ² of 1.1, indicating an unbroken steep spectrum. We also note that the source appears marginally resolved only in the LOFAR data, consistent with the interpretation. revision: yes

Circularity Check

0 steps flagged

No circularity: direct observational identification without derivations or fitted parameters

full rationale

The paper reports an observational result obtained by cross-matching existing radio (LOFAR, VLA) and JWST source catalogs in the COSMOS field. It identifies one source detected across multiple radio frequencies but undetected in every JWST, HST, Chandra, Herschel, and ALMA band. No equations, model fits, parameter estimations, or derivations appear in the provided text. The uniqueness claim rests on catalog non-detections rather than any self-referential construction, self-citation chain, or renaming of prior results. The derivation chain is therefore empty; the output is independent of any fitted inputs or ansatzes internal to the paper.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The claim relies on standard domain assumptions about radio source detection and the physical meaning of non-detections in infrared bands; no free parameters or new entities are introduced.

axioms (2)

domain assumption Radio continuum emission can be detected independently of optical or infrared emission from the same object
Invoked implicitly when treating the radio detection as real despite complete non-detection at other wavelengths.
domain assumption Non-detection in JWST bands implies either extreme dust obscuration or very high redshift
Used to interpret the source as potentially at cosmic dawn.

pith-pipeline@v0.9.0 · 5473 in / 1361 out tokens · 40473 ms · 2026-05-14T21:38:48.659265+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages

[1]

2024, A&A, 685, A1, doi: 10.1051/0004-6361/202348407 Ba˜ nados, E., Momjian, E., Connor, T., et al

Adscheid, S., Magnelli, B., Liu, D., et al. 2024, A&A, 685, A1, doi: 10.1051/0004-6361/202348407 Ba˜ nados, E., Momjian, E., Connor, T., et al. 2025, NatAs, 9, 293, doi: 10.1038/s41550-024-02431-4

work page doi:10.1051/0004-6361/202348407 2024
[2]

2023, grizli, v.1.9.11, Zenodo, doi:10.5281/zenodo.8370018

Brammer, G. 2023, grizli, 1.9.11 Zenodo, doi: 10.5281/zenodo.8370018

work page doi:10.5281/zenodo.8370018 2023
[3]

L., Gnedin, N

Carilli, C. L., Gnedin, N. Y., & Owen, F. 2002, ApJ, 577, 22, doi: 10.1086/342179

work page doi:10.1086/342179 2002
[4]

M., Kartaltepe, J

Casey, C. M., Kartaltepe, J. S., Drakos, N. E., et al. 2023, ApJ, 954, 31, doi: 10.3847/1538-4357/acc2bc

work page doi:10.3847/1538-4357/acc2bc 2023
[5]

2016, ApJ, 819, 62, doi: 10.3847/0004-637X/819/1/62

Civano, F., Marchesi, S., Comastri, A., et al. 2016, ApJ, 819, 62, doi: 10.3847/0004-637X/819/1/62

work page doi:10.3847/0004-637x/819/1/62 2016
[6]

P., Lyu, J., et al

Endsley, R., Stark, D. P., Lyu, J., et al. 2023, MNRAS, 520, 4609, doi: 10.1093/mnras/stad266

work page doi:10.1093/mnras/stad266 2023
[7]

2025, A&A, 697, A46, doi: 10.1051/0004-6361/202452461

Gentile, F., Talia, M., Enia, A., et al. 2025, A&A, 697, A46, doi: 10.1051/0004-6361/202452461

work page doi:10.1051/0004-6361/202452461 2025
[8]

J., Duncan, K

Gloudemans, A. J., Duncan, K. J., R¨ ottgering, H. J. A., et al. 2021, A&A, 656, A137, doi: 10.1051/0004-6361/202141722

work page doi:10.1051/0004-6361/202141722 2021
[9]

V., Taylor, R., Sekhar, S., et al

Lal, D. V., Taylor, R., Sekhar, S., et al. 2025, ApJ, 991, 9, doi: 10.3847/1538-4357/adf6dc

work page doi:10.3847/1538-4357/adf6dc 2025
[10]

S., Casey, C

Long, A. S., Casey, C. M., McKinney, J., et al. 2026, ApJ, 999, 47, doi: 10.3847/1538-4357/ae2fda

work page doi:10.3847/1538-4357/ae2fda 2026
[11]

2008, A&ARv, 15, 67, doi: 10.1007/s00159-007-0008-z

Miley, G., & De Breuck, C. 2008, A&ARv, 15, 67, doi: 10.1007/s00159-007-0008-z

work page doi:10.1007/s00159-007-0008-z 2008
[12]

arXiv e-prints , keywords =

Muzzin, A., Suess, K. A., Marchesini, D., et al. 2025, arXiv, arXiv:2507.19706, doi: 10.48550/arXiv.2507.19706

work page doi:10.48550/arxiv.2507.19706 2025
[13]

M., Foley , R

Oliver, S. J., Bock, J., Altieri, B., et al. 2012, MNRAS, 424, 1614, doi: 10.1111/j.1365-2966.2012.20912.x

work page doi:10.1111/j.1365-2966.2012.20912.x 2012
[14]

L., et al

Schinnerer, E., Smolˇ ci´ c, V., Carilli, C. L., et al. 2007, ApJS, 172, 46, doi: 10.1086/516587

work page doi:10.1086/516587 2007
[15]

W., Hardcastle, M

Shimwell, T. W., Hardcastle, M. J., Tasse, C., et al. 2026, A&A, 707, A198, doi: 10.1051/0004-6361/202557749

work page doi:10.1051/0004-6361/202557749 2026
[16]

B., Paquereau, L., et al

Shuntov, M., Akins, H. B., Paquereau, L., et al. 2025, A&A, 704, A339, doi: 10.1051/0004-6361/202555799 Smolˇ ci´ c, V., Novak, M., Bondi, M., et al. 2017, A&A, 602, A1, doi: 10.1051/0004-6361/201628704

work page doi:10.1051/0004-6361/202555799 2025
[17]

2025, arXiv, arXiv:2506.06418, doi: 10.48550/arXiv.2506.06418 van Breugel, W., De Breuck, C., Stanford, S

Sun, F., Yang, J., Wang, F., et al. 2025, arXiv, arXiv:2506.06418, doi: 10.48550/arXiv.2506.06418 van Breugel, W., De Breuck, C., Stanford, S. A., et al. 1999, ApJL, 518, L61, doi: 10.1086/312080

work page doi:10.48550/arxiv.2506.06418 2025
[18]

A., Faisst, A

Zavala, J. A., Faisst, A. L., Aravena, M., et al. 2026, ApJL, 998, L36, doi: 10.3847/2041-8213/ae382a

work page doi:10.3847/2041-8213/ae382a 2026