REVIEW 3 major objections 4 minor 8 cited by
A parametric strong-lensing model of MACS J1931.8-2635 predicts that the z=5.13 supernova SN Eos appears as five images, making it the most distant multiply imaged supernova known.
Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →
T0 review · deepseek-v4-flash
2026-08-02 23:20 UTC pith:5ICDX7OH
load-bearing objection Solid, useful lens-model paper with a real testable prediction; the five-image SN Eos claim is conditional on a single-halo parametric family that already leaves one host image ~2″ off. the 3 major comments →
VENUS: Strong-lensing model of MACS J1931.8-2635 -- revealing the farthest multiply imaged supernova
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
On the paper's terms, the central claim is that a single dark-matter halo plus a galaxy population scaled by a standard luminosity-velocity relation, with a free brightest cluster galaxy and no external shear, reproduces 51 lensed images across 19 systems with an r.m.s. error of 0.44 arcsec, and from that model the delay-potential differences predict exactly five images of SN Eos. The two observed images, detected simultaneously in JWST/NIRCam data, are reproduced to well within 0.1 arcsec, with magnifications of about 25 and 30 and a predicted time delay below 5 days; the other three images are predicted with observer-frame time delays of about 3.5, 3.5, and 54 years and magnifications of r
What carries the argument
The central mechanism is the parametric lens model: a single dark-matter halo of the Pseudo-Isothermal Elliptical Mass Distribution (PIEMD) shape, plus cluster galaxies modeled as dual Pseudo-Isothermal Elliptical (dPIE) profiles, all scaled by a fixed luminosity-velocity relation with a core radius of 0.2 kpc. The Fermat-potential difference between image positions, computed from this mass distribution, yields the predicted magnifications and time delays; those predictions are what turn two observed point sources into a five-image geometry and a transient confirmation.
Load-bearing premise
The load-bearing premise is that the cluster's mass is well described by a single smooth dark-matter halo plus galaxy-scale halos scaled by a fixed luminosity-velocity relation (Faber-Jackson), with the brightest cluster galaxy's weight capped at 20 and no external shear; all predicted supernova positions, magnifications, and time delays follow from the Fermat-potential differences of that particular mass model, and the model's own ~2 arcsec reproduction failure for host imag
What would settle it
Measure the time delay between the two observed images 101.1 and 101.2 from multi-epoch photometry; a delay exceeding about 5 days would contradict the predicted delay-potential difference. Alternatively, take a deep image at the predicted position of image p101.3 or p101.4 about 3.5 years after the observed pair: if no point source appears while the host-galaxy images are still visible, the five-image geometry is wrong. A point source appearing now at p101.5 would also falsify the transient interpretation, since that image should not arrive for another ~54 years.
If this is right
- If the predicted delays hold, monitoring the positions of p101.3 and p101.4 should reveal the same supernova reappearing in about 3.5 years, and p101.5 in about 54 years, turning SN Eos into a multi-epoch cosmic clock.
- A precise measurement of the small delay between the two observed images would directly link the Fermat-potential differences to the Hubble constant through the time-delay distance.
- The reproduction of 19 systems with a single dark-matter halo, compared with earlier two-halo models, supports a simpler mass description for this relaxed cool-core cluster and refines its Einstein radius and enclosed mass estimates.
- The triply imaged spiral galaxy Charybdis, with magnification about 5.5 and source-plane resolution near 90-100 pc, offers a detailed view of spiral structure at z~3.65.
- The confirmed absence of the three earlier images demonstrates that lens-model-driven searches can identify transients before their reappearance, complementing traditional difference-imaging searches.
Where Pith is reading between the lines
- A concrete test: the predicted 3.5-year delays imply that the next appearances of SN Eos should occur about 3.5 years after the observed pair. If JWST or another telescope observes those positions then and finds no point source while the host galaxy remains, the model's delay-potential differences are wrong.
- The 54-year predicted image would have arrived roughly five decades before the observed pair; archival plates or older digital images of this field could in principle record a point source at that position, giving an immediate check on the transient interpretation.
- If independent modeling with a second dark-matter halo (as in the earlier best published model) shifts the predicted positions substantially, then the single-halo assumption becomes the main systematic; comparing such models would quantify the roughly 20% error the paper assigns to magnifications and delays.
- The same lens-model machinery can be applied to other cluster fields: any point source with two images near a critical curve and a short time delay is a candidate transient, and the predicted earlier images tell observers exactly when to look.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper presents a parametric strong-lensing model of the galaxy cluster MACS J1931.8-2635 using new JWST/NIRCam VENUS imaging, archival HST/CLASH data, and MUSE/NIRSpec spectroscopy. The model is constrained by 19 multiply-imaged systems (10 newly identified, 9 with spectroscopic redshifts) and achieves a reported image-plane r.m.s. of 0.44 arcsec. For the spectroscopically confirmed z=5.133 supernova SN Eos, the model predicts five images: the two observed images with magnification ~25-30 and a <5 day mutual delay, plus three earlier images with predicted delays of about 3.5, 3.5, and 54 years and magnifications of roughly 13, 13, and 2. The paper interprets the absence of detections at the predicted unobserved positions as confirmation of the transient nature of the source, and additionally presents a triply imaged grand-design spiral galaxy ('Charybdis') and a discussion of lensed SNe for cosmography.
Significance. If the model predictions are robust, SN Eos is the most distant multiply imaged supernova known and the first transient whose lensed nature was inferred largely from a lens-model-predicted five-image geometry prior to deep spectroscopic confirmation. The paper has real strengths: the model did not fit time delays or magnifications, so those are genuine outputs; the independent archival MUSE Ly-alpha detection near two predicted positions is supportive; and the lens model and maps are publicly released via Zenodo. However, the central quantitative predictions (delays and magnifications of p101.3-p101.5) rest on a single parametric mass-model family with asserted ~20% systematic errors and with a documented ~2 arcsec reproduction failure for image 100.4 in the same source system. These issues directly affect the strength of the transient-confirmation claim, so the manuscript needs additional model-family validation before the central claims can be accepted at face value.
major comments (3)
- [Table 3, §3.2] The systematic uncertainties on the predicted magnifications and time delays (~20% for both |mu| and tau, and ±5 days for the 101.1-101.2 delay) are asserted rather than derived from model-family variation. These quantities are Fermat-potential differences computed for a single-PIEMD DMH plus dPIE galaxy profiles with fixed Faber-Jackson scaling and no external shear. The prior Caminha et al. (2019) best model used two DMHs and achieved 0.34 arcsec r.m.s.; the present 0.44 arcsec with more systems does not rule out that family. I request at least one alternative-model run (e.g., two DMHs, external shear, or free scaling indices) and a report of how the Table 3 values shift. This is load-bearing because the headline predictions are the delays and magnifications of p101.3-p101.5.
- [§3.3, Table 4] Image 100.4, one of the five Eos-host images, is reproduced about 2 arcsec away from its MUSE position even though the SN images 101.1-101.2 are reproduced within 0.1 arcsec. Predicted p101.4 lies very close to 100.4; therefore a 2 arcsec local error in the potential at exactly the region of a predicted image means the statement that 'the absence of detections at the predicted positions ... confirms the transient nature' is only as strong as those predicted positions. Please provide a position-uncertainty map, or alternative-model predictions for p101.3-p101.5, so the non-detection argument can be evaluated against realistic model scatter.
- [Abstract, Table 3] The numbers for the predicted earlier images are inconsistent: the first abstract gives delays of 3.6±0.7 and 3.4±0.7 years and magnifications of 14.5±2.9 and 11.9±2.4, while the second abstract and Table 3 give 3.7±0.7, 3.5±0.7, 12.9±2.6, and 13.0±2.9. Please reconcile these values and ensure the abstract matches the table.
minor comments (4)
- [§3.2] The two observed SN images are included as constraints with sigma_pos=0.02 arcsec, giving them substantial weight. Their positions are therefore fitted, and the predicted <5 day delay between them, while not directly fitted, is conditional on the model being anchored to those positions. I suggest stating this explicitly when presenting the 'agreement with simultaneous observation' as a supporting point.
- [Table 4] The lens-model redshift for system 9 is listed as z_model = 19.03(+0.57,-4.86), which appears implausible and likely a typo. Please check the entry.
- [Figure 6 caption] The BCG weight w_Gal1 is capped at 20, but the posterior panel appears truncated at this boundary and the unconstrained value reaches 44. Please explain how the cap and the truncation affect the reported statistical uncertainties on other parameters.
- [§3.1, Data availability] AstroLensPy is described as 'available through private communication,' which limits reproducibility. The Zenodo lens-model release is welcome, but I encourage also releasing the software (or a frozen version) used to produce the reported numbers.
Circularity Check
No significant circularity: SN Eos magnifications and time delays are unconstrained model outputs, with independent archival MUSE and HST checks.
full rationale
The derivation chain is not circular. Section 3.1 states that the chi-square minimization uses only image positions and that "constraints such as parity, relative magnification and time delays for multiple images... are not used here." SN Eos images 101.1-101.2 are used as positional constraints with sigma_pos=0.02 arcsec (Sec. 3.2), so the model is anchored to their positions, but the claimed predictions - the three additional images p101.3-p101.5, their positions, magnifications, and time delays in Table 3 - are computed from the best-fit lens potential at source-plane solutions that were not fitted, and are therefore genuine outputs rather than fitted inputs. The five-image prediction is not identical to the observed constraints: the model predicts unseen images (p101.5, and host image p100.5 outside the MUSE field of view). Independent checks are quoted: archival VLT/MUSE Lyman-alpha at 754.4 nm detected "near the model-predicted position of the two additional images," HST 2012 non-detection versus 2024 detection, and NIRSpec z=5.133 confirmation. The parametric family (single PIEMD DMH, dPIE galaxies with fixed Faber-Jackson slopes, no external shear) is a stated modeling assumption, not an input that encodes the SN Eos time delays or magnifications. The paper itself flags limitations: image 100.4 is reproduced about 2 arcsec away from its MUSE position (Sec. 3.3), the BCG weight is capped at 20 to avoid an unphysical w=44 solution (Fig. 6 caption), and the 20% systematic errors on mu and tau are asserted rather than derived (Table 3 note). These are accuracy and robustness concerns, not circular reductions. Self-citations (Zitrin et al. 2015; Furtak et al. 2023c) supply the parametrization and code provenance; the lens equations are given in Appendix A and the model is checked against the independent Caminha et al. (2019) model, so no load-bearing claim rests on an unverified self-citation.
Axiom & Free-Parameter Ledger
free parameters (10)
- DMH velocity dispersion sigma_DMH =
1172 km/s
- DMH core radius a_DMH =
75.6 kpc
- DMH ellipticity e_DMH and position angle =
e=0.551, PA=88.07 deg
- DMH center offsets =
Delta_alpha=1.33", Delta_delta=-0.17"
- BCG core radius, ellipticity, luminosity weight =
a=11.4 kpc, e=0.393, w_BCG=20 (capped)
- L* galaxy velocity dispersion sigma_star and cut radius s_star =
sigma_star=194.3 km/s, s_star=17.1 kpc
- Redshifts of 10 non-spectroscopic systems =
z_model in Table 4 (e.g., 2.45, 3.65, 19.03)
- Positional-error weights sigma_pos =
0.2" (all systems) vs 0.02" (SN Eos); rescaled to 0.5" for chi2_nu
- Systematic-error envelope for |mu| and tau =
~20%, plus +/-5 days for the 101.1-101.2 pair
- Fixed scaling priors =
lambda=1/4, beta=1/2, alpha=1/2; a_star=0.2 kpc; M_star=-22.6
axioms (8)
- standard math Thin-lens, single-plane lens equation and Fermat-potential time-delay formalism apply.
- domain assumption One PIEMD dark-matter halo + dPIE galaxies with no external shear describes the cluster potential; a second DMH is not needed.
- domain assumption Cluster-member galaxies trace mass via the Faber-Jackson relation with fixed slopes; the 107-galaxy red-sequence catalog is adequate.
- domain assumption The 10 newly identified multiple-image systems are genuine associations of the same background source.
- domain assumption eazy photometric redshifts for non-spectroscopic systems are reliable priors.
- domain assumption SN Eos is a point-like transient at z=5.133 and the two JWST point sources are its images.
- domain assumption The absence of detections at the three predicted positions is a real non-detection.
- domain assumption Flat LambdaCDM with H0=70, Omega_m=0.3, Omega_Lambda=0.7.
read the original abstract
We present a parametric strong-lensing model for the galaxy cluster MACS J1931.8-2635 ($z_l = 0.35$), accompanying the detection of the spectroscopically confirmed SN Eos at $z = 5.13$ (Coulter et al. 2026). We identify 10 new multiple-image systems in recent VENUS JWST/NIRCam imaging, so that the model is constrained with a total of 19 robust multiple-image systems -- nine of which also have a spectroscopic redshift. For the point-like source corresponding to SN Eos, our model predicts a total of five images, with the observed radial image pair having a similar magnification of $\mu \simeq 25 - 30$ and a small time delay of $< 5$ days, in agreement with their simultaneous observation. According to the model, the other three predicted images arrived earlier, with time delays of $3.6 \pm 0.7$, $3.4 \pm 0.7$ and $53.9 \pm 10.8$ years prior to the two observed images, and with magnifications of $14.5 \pm 2.9$, $11.9 \pm 2.4$ and $2.2 \pm 0.4$, respectively. The absence of detections at the predicted positions, where the host galaxy's images are also visible, confirms the transient nature of the source. SN Eos and its host galaxy are studied in separate articles, and we here focus on the lens model. The final model reaches a very good $r.m.s.$ distance between model and observations of $0.44''$. We present the lens-modeling results, including newly identified systems such as a triply imaged, grand-design spiral galaxy candidate at $z \simeq 3.65_{-0.09}^{+0.04}$, and discuss the potential of using high-redshift lensed SNe for cosmography.
Figures
Forward citations
Cited by 8 Pith papers
-
A massive barred spiral galaxy at z = 5.102 discovered by JWST
Discovery and characterization of the highest-redshift barred spiral galaxy candidate at z=5.102, with bar length ~4.5 kpc, stellar mass 10^10.45 solar masses, SFR 144 solar masses per year, and evidence for AGN and i...
-
Other red dots: A possible GLIMPSE of normal AGB stars at Cosmic Noon through extreme lensing
Four faint red point sources near critical curves in JWST images of Abell S1063 are interpreted as extremely magnified AGB stars and a yellow supergiant at cosmic noon.
-
VENUS: an ultra-faint galaxy hosting the metal-poor type II supernova at $z=5.13$ Witnessing the initial metal enrichment with extremely frequent core-collapse supernovae?
The z=5.13 supernova SN Eos exploded in an ultra-faint, very metal-poor dwarf galaxy, suggesting core-collapse supernovae were much more frequent in early low-metallicity environments.
-
Other red dots: A possible GLIMPSE of normal AGB stars at Cosmic Noon through extreme lensing
Detection of extremely magnified individual AGB stars and a yellow supergiant at z~1-4 in JWST lensing observations of Abell S1063.
-
On the correlation between globular clusters and the distribution of dark matter in galaxy clusters: the case of Abell 2744
Bright globular clusters in Abell 2744 trace the three main mass clumps and match weak-lensing mass maps more closely than other galactic components, using a Poisson point process model.
-
High-Redshift Gravitational Lens Discoveries in JWST NIRCam Using AnomalyMatch
AnomalyMatch applied to JWST data identified 58 gravitational lenses, 37 new, graded by experts and spanning redshifts to zphot ~2.1.
-
Strong Gravitational Lensing with the James Webb Space Telescope
A review summarizing recent advances in strong gravitational lensing applications and near-future prospects with the James Webb Space Telescope.
-
Strong Gravitational Lensing with the James Webb Space Telescope
Strong gravitational lensing paired with JWST enables magnified high-resolution views of distant sources and improved constraints on dark matter.
Reference graph
Works this paper leans on
-
[1]
2025, arXiv e-prints, arXiv:2512.08054
Abdurro’uf, Coe, D., Resseguier, T., et al. 2025, arXiv e-prints, arXiv:2512.08054. https://arxiv.org/abs/2512.08054 Aguena, M., Aiola, S., Allam, S., et al. 2026, The Open Journal of Astrophysics, 9, 55863, doi: 10.33232/001c.155863 Allingham, J. F. V., Jauzac, M., Lagattuta, D. J., et al. 2023, MNRAS, 522, 1118, doi: 10.1093/mnras/stad917 Allingham, J. ...
arXiv 2025
-
[3]
For dark-matter halos (DMH), we generally prefer the un-truncated Pseudo-Isothermal Elliptical Matter Distribution (PIEMD, C. R. Keeton 2001). We give its 3D mass density: ρ(r) = ρ0 1 + (r/a)2 = σ2 0 2πG 1 a2 +r 2 , where: ρ0 = σ2 0 2πGa2 ,(A2) is the 3D-density normalization,athe core radius,σ 0 the 1D velocity dispersion andGthe constant of Newton. In t...
arXiv 2001
-
[2026]
‘c’ stands for candidate where identification was more ambiguous, or if image was not used as constraint
Note—Columns: (i) Arc Id. ‘c’ stands for candidate where identification was more ambiguous, or if image was not used as constraint. Systems 1-7 were already included in G. B. Caminha et al. (2019), and systems 2, 3, 4, 8 and 9 were identified in A. Zitrin et al. (2015). Systems 10 to 18, as well as the Eos systems (100,
2019
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.