{"id":"ca942f67-e60e-4763-9441-83b2dd02aec8","arxiv_id":"2412.03644","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A gravitationally lensed dusty galaxy at z=2.78 shows a nuclear double spiral and a possible bar, resolved at about 55 pc with ALMA.","lead":"Using long-baseline ALMA and gravitational lensing, the authors reconstruct the dusty centre of a z=2.78 star-forming galaxy at about 55 parsec resolution. They report a dual spiral arm pattern and a tentative nuclear bar in the central kiloparsec, suggesting that the same secular processes seen in nearby galaxies were already at work at cosmic noon.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The dual-arm spiral rests on a single lens-inversion pipeline; a smooth-source control is needed to rule out artefacts of the magnification-adapted grid and regularisation.","rationale":"Read in good faith: the paper is careful; it compares three regularisation types, checks total flux against LABOCA, performs 1000 noise realisations, and explicitly discloses the mass-sheet/group caveat. These are real independent supports. The strongest claim, however, is the existence of a nuclear spiral at z=2.78, and that claim rests entirely on the reconstructed source surface brightness obtained from one lens-inversion code. The robustness tests are all internal to that pipeline: they vary the smoothness penalty and the lens potential, but do not demonstrate that the adaptive Delaunay source grid plus regularisation cannot generate spiral-like coherence from a smooth source. This is the weakest link in the logical chain from data to conclusion. The reader's chosen weak point, the mass-sheet degeneracy, is disclosed and explicitly preserves structure; it threatens the physical scale and the super-Eddington SFR density but not the morphological central claim. I therefore partially agree with the reader: the overall conditional verdict is appropriate, but the condition that should be added or checked is a pipeline null test and independent reconstruction, not only a group-environment check. No change to the verdict is recommended: the evidence is sufficient for a conditional publication, and the control test would either remove or confirm the remaining concern.","tokens_in":10314,"tokens_out":11841,"duration_ms":133163,"concrete_test":"Forward-model a smooth, non-spiral source (e.g., an exponential Sersic or a single central clump) through the best-fit lens model, sample ALMA visibilities at the same uv coverage, add realistic noise and the nominal plus-or-minus 10 percent calibration error, and run the identical pronto inversion with all three regularisation types. If the recovered source shows a two-arm morphology or a position-angle swing of 45 degrees or more at 3-sigma significance, the pipeline can imprint the claimed features and the central claim is not secure. If the control source is recovered as smooth, the observed spiral is unlikely to be a pure inversion artefact. As a secondary check, re-invert the real data with an independent code using a regular source grid to confirm the arms persist.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is morphological: the reconstructed source in Section 3 (Figs. 2-4) shows a dual-arm spiral and a possible bar. The robustness tests vary the regularisation type (gradient, curvature, area-weighted gradient) and the angular freedom of the lens model within the same pronto semi-linear inversion. They do not include a null test in which a known smooth source is passed through the pipeline, nor an independent lens-modelling code with a differently constructed source grid. This matters because the source is built on a Delaunay grid adapted to the lens magnification; near the tangential caustics the vertex density and the regularisation prior can create coherent, elongated structures that resemble arms. The reported 3-sigma mask includes noise, data artefacts and 'non-linear effects of lensing', but not this pipeline-level imprinting. Lens-model parameter uncertainties are deliberately excluded because the posterior is narrow; however, the mass-sheet/group-environment degeneracy acknowledged in Section 4 changes physical scale and SFR density without changing structure, so it is not the main threat to the existence of the spiral. The isophote position-angle swing used to claim a bar could equally be produced by two spiral arms or an off-centre clump, and no barred-spiral versus arm-only model comparison is made. Thus the least secure link is not the astrophysical interpretation but the faithfulness of the source reconstruction itself.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents long-baseline ALMA Band 7 (350 GHz) observations of the gravitationally lensed dusty star-forming galaxy SPT0538-50 at z=2.78, reconstructed in the source plane with the semi-linear pixellated lens-modelling code pronto. The authors use three source-regularisation choices (gradient, curvature, area-weighted gradient) and additional angular freedom in the lens potential, and report a rest-frame ~230 micron map with a surface-brightness-weighted effective resolution of ~55 pc. They find that the central kiloparsec has an exponential profile with effective radius 0.6±0.1 kpc and Sérsic index 1.2±0.2, a dual spiral arm morphology with ~100 pc clumps at 3–9 sigma, a possible nuclear bar within ~300 pc, and a super-Eddington central star-formation-rate surface density of about 2000 M_sun yr^-1 kpc^-2. They interpret these features as evidence that secular dynamical processes can funnel gas into the nucleus, and argue that gravitational lensing combined with long-baseline ALMA is the only way to reach these physical scales at cosmic noon.","tokens_in":10631,"tokens_out":7918,"duration_ms":73100,"significance":"If the reconstruction is faithful, this is a rare and important result: a compact, rotationally supported nuclear disc with spiral structure and a possible bar in a DSFG at z=2.78, at physical scales that are otherwise inaccessible. The paper has real strengths: the morphology is stable across three regularisation types; the reconstructed total flux agrees with an independent APEX/LABOCA measurement; the significance maps come from 1000 noise realisations; and the authors honestly discuss the mass-sheet degeneracy and the uniform dust-temperature assumption. However, the central claim is morphological, so its support depends on the absence of pipeline-induced structure. The evidence does not yet include a smooth-source control or an independent lens-modelling code, and the bar identification rests on isophote position-angle swings that have not been compared with an arm-only model. These gaps are fixable within the manuscript's scope and should be addressed before the result is stated as established.","major_comments":[{"comment":"The dual-arm spiral and associated clumps are the central result, but they come from a single inversion pipeline (pronto) whose Delaunay source grid is adapted to the lens magnification. No control experiment is shown in which a known smooth source is passed through the same pipeline and reconstructed, to check that the magnification-adapted grid and the regularisation prior do not imprint coherent elongated structures near the tangential caustics. The 3-sigma mask accounts for noise, data artefacts and non-linear lensing effects, but not for this possible pipeline-level imprinting. The reference to Galan et al. (2024) concerns mock optical data and does not cover the present interferometric case. I request a mock-injection test or an independent source-grid/code reconstruction before the spiral-arm morphology is asserted.","section":"Section 3, Figs. 2-4"},{"comment":"The 'potential nuclear bar' is inferred from a swing of the isophote position angle (from about 25 deg at 0.2 kpc to -20 deg at 0.6 kpc) together with ellipticity peaks in Fig. 4. These signatures are also expected for a two-arm spiral or an off-centre clump, and no quantitative comparison is made between a barred-spiral model and an arm-only model (for example, via the Fourier m=2 phase, bar/interarm contrast, or a model-selection statistic). Since the bar appears in the abstract and conclusions, the evidence should be strengthened, or the conclusion should be explicitly limited to 'tentative' and the bar claim removed from the abstract's definitive list.","section":"Section 3, Fig. 4"},{"comment":"The source-plane uncertainties and significance maps are generated from 1000 noise realisations at the maximum a posteriori lens model, and the text states that the posterior is narrow. However, the lens model parameters, including the multipole coefficients and the regularisation hyper-parameter, were determined jointly with the source surface brightness, so their covariance contributes to the total uncertainty. Neglecting these terms may understate the uncertainties on clump significance and on the isophote parameters used for the bar claim. Please report or bound the effect of lens-model parameter variations, or justify quantitatively why the narrow posterior implies a negligible contribution to the source-plane maps.","section":"Sections 2 and 3"},{"comment":"The mass-sheet/group-environment degeneracy is acknowledged, and the authors correctly note that it would not change the morphology; however, it directly changes the physical scale (including the quoted ~55 pc resolution and the 0.6 kpc effective radius) and the super-Eddington surface-density interpretation. Because the lens environment of SPT0538-50 is unknown, this degeneracy is a live source of error in the astrophysical conclusions, not merely a caveat. At minimum, please estimate the range of magnification and physical-scale shifts allowed by plausible group contributions, or state more explicitly how the main conclusions would move under such a transform.","section":"Section 4"}],"minor_comments":[{"comment":"Please clarify how the Delaunay source-grid vertex density is set relative to the magnification map and whether the same grid is used for all three regularisation types; this affects the interpretation of the median vertex spacing of ~50 pc quoted in Section 3.","section":"Section 2, second paragraph"},{"comment":"The definition of the 'surface-brightness-weighted mean FWHM of 55 pc' should be more explicit: state whether this is computed from the magnification-corrected beam in Fig. 2 and whether the three regularisation types give individually consistent values.","section":"Section 3, first paragraph"},{"comment":"The statement that the Sérsic fit is poor (Fig. 3) would benefit from a quantitative goodness-of-fit statistic and from reporting the priors, chain length, and convergence of the emcee fit.","section":"Section 3, Sérsic fit"},{"comment":"The phrase 'contours of signal-to-noise ratio in steps of 3-sigma' is ambiguous; specify whether these contours come from the noise-realisation significance maps or from the lensed image, and define the sigma used.","section":"Section 3, Fig. 2 caption"},{"comment":"There is a typographical double parenthesis after the citation 'Erwin et al. 2015))' that should be corrected.","section":"Section 4, first paragraph"},{"comment":"The statement that smoothing the source to the native 30 mas resolution erases any evidence of the spiral is expected given the resolution loss; it should not be used as a robustness argument without a forward-modelled low-resolution mock source.","section":"Section 4, third paragraph"}],"recommendation":"major_revision","confidential_remarks":"To the editor: This is a well-written, honest letter with a potentially important result. The gaps identified in the major comments—smooth-source control, bar versus arm-only comparison, and propagation of lens-model uncertainties—are fixable with analysis already in hand or modest additional computation, and do not, in my view, require new observations. I would support acceptance after a major revision that adds the control and either verifies or tempers the bar claim. There are no concerns about citation practices or novelty overlap."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this is the first attempt to resolve the central kiloparsec of a dusty star-forming galaxy at z~2.8 in dust continuum, and it reports a dual spiral arm morphology with a tentative nuclear bar at ~55 pc resolution. If real, that is a genuine first—previous spiral/bar claims at these redshifts were an order of magnitude larger in scale. The paper is careful with its claims. The source reconstruction is tested across three regularization types (gradient, curvature, area-weighted), the total reconstructed flux agrees with an independent APEX/LABOCA measurement, and the significance maps come from 1000 noise realizations. The authors also disclose the main caveats: possible mass-sheet degeneracy from a group environment, uniform dust temperature assumption, and neglect of lens model parameter uncertainties.\n\nThe genuine novelty is physical scale, not method. The lens modelling is inherited from Stacey et al. (2024) and the isophote analysis is standard; the resolution is what enables a sub-kpc view of secular structure in a cosmic-noon DSFG.\n\nThe soft spots are proportionate. The stress-test concern is fair: all reconstructions come from the same pronto pipeline, and there is no null test in which a known smooth source is passed through to see whether the magnification-adapted grid and regularization can produce coherent arm-like features. That is a real gap for a morphological claim. It does not kill the paper—the arms are also visible in the lensed image and the clumps survive noise realizations—but 'spiral' is a stronger word than the evidence strictly supports. The bar is softer still: a 45-degree position-angle swing in isophotes is suggestive, but no barred-versus-arm-only model comparison is made. The authors label it tentative, which is honest. The mass-sheet/group issue would rescale size and SFR density but not structure, so it is not the main threat. The super-Eddington SFR density depends on uniform dust temperature, which the authors acknowledge.\n\nBottom line: this is a solid, well-scoped letter that deserves a serious referee. The main requests should be a smooth-source control and an explicit bar/arm model comparison, not rejection. I would cite it as a candidate nuclear spiral, with the caveat that the morphology is pipeline-dependent until tested further.","headline":"First sub-kpc look at a DSFG nucleus: a careful lensing reconstruction with a tentative spiral and bar, but the morphology rests on one pipeline and wants a smooth-source null test.","tokens_in":11195,"tokens_out":3080,"would_cite":true,"duration_ms":29946,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Long-baseline ALMA observations and pixellated lens modelling resolve the central kiloparsec of the z=2.78 dusty star-forming galaxy SPT0538-50 into a dual spiral arm morphology with a tentative nuclear bar, implying a compact…","keywords":["high-redshift galaxies","galaxy structure","dusty star-forming galaxies","gravitational lensing","nuclear spiral","galactic bars","star formation rate density","ALMA continuum"],"falsifier":"A concrete test would be resolved molecular-gas kinematics, for example CO or [CII] line emission at matched sub-kiloparsec resolution: if the reconstructed spiral arms and bar do not trace coherent, rotating gas motions with non-circular streaming along the arms, the morphological features would be artefacts or transient tidal debris rather than a secular nuclear disc. Imaging the lens environment to establish whether the lens is in a group would also decide whether the mass-sheet degeneracy rescales the reported physical numbers.","tokens_in":10157,"feed_emoji":"🌌","tokens_out":8408,"duration_ms":77948,"temperature":0.7,"pith_summary":"The paper tries to establish that the innermost kiloparsec of the distant, dusty, star-forming galaxy SPT0538-50 at redshift 2.78 is not a chaotic merger pile-up but a structured, rotating nuclear disc. Using long-baseline ALMA continuum data and a pixellated gravitational lens reconstruction, the authors resolve the rest-frame 230 $\\mu$m dust emission at a mean effective resolution of about 55 pc. They find a central exponential profile, a dual spiral arm morphology, clumps of about 100 pc, and tentative evidence for a bar in the central 300 pc. The implied star formation rate density near the centre (roughly $2000\\ M_\\odot\\,\\mathrm{yr}^{-1}\\,\\mathrm{kpc}^{-2}$) is super-Eddington compared with local ultraluminous infrared galaxies. If right, secular processes such as spiral arms and bars were already channelling gas into the centres of massive galaxies at cosmic noon, feeding nuclear starbursts and supermassive black hole growth.","feed_headline":"Nuclear spiral found in a galaxy core at z=2.78","feed_subtitle":"Gravitational lensing plus long-baseline ALMA resolves 55-pc dust structure, hinting at how gas feeds starbursts and black holes.","key_machinery":"The load-bearing tool is pixellated gravitational lens modelling: a semi-linear inversion that solves for lens parameters and source surface brightness simultaneously, with the source placed on a Delaunay grid adapted to the lens magnification. Long-baseline ALMA observations at 350 GHz (rest-frame about 230 $\\mu$m) provide the data, and gravitational magnification boosts the effective resolution to a mean of about 55 pc. The paper tests robustness by repeating the inversion with three regularisation types (gradient, curvature, and area-weighted gradient), then uses isophote fitting with Fourier harmonics to identify the bar's signature in ellipticity and position-angle swing.","core_discovery":"The central claim, stated on the paper's own terms, is that the central kiloparsec of SPT0538-50 exhibits a dual spiral arm morphology and a potential nuclear bar that could facilitate gas inflow, feeding the nuclear starburst and supermassive black hole. The authors reconstruct the dust emission at an effective resolution of about 55 pc (varying from below 50 to 80 pc across the source) and show the same morphology emerges under gradient, curvature, and area-weighted gradient source regularisation. A two-dimensional Sérsic fit gives an effective radius of $0.6\\pm0.1$ kpc and a Sérsic index of $1.2\\pm0.2$, consistent with a compact exponential disc, but the residuals and isophote fits reveal the spiral arms and a position-angle swing of roughly 45 degrees that is a typical signature of a bar. The central region is resolved into two clumps rather than a point source, and the inferred star formation rate density reaches about $2000\\ M_\\odot\\,\\mathrm{yr}^{-1}\\,\\mathrm{kpc}^{-2}$, which the authors call super-Eddington relative to the limit for local ultraluminous infrared galaxies.","pith_inferences":["Not in the paper: if nuclear bars are common in dusty star-forming galaxies at cosmic noon, the usual picture in which gas-rich mergers dominate angular-momentum transport would need to share the stage with internal secular evolution.","Not in the paper: the super-Eddington surface density estimate assumes uniform dust temperature; a radially decreasing temperature gradient would raise the central value, while AGN heating would lower the inferred star formation rate, so multi-frequency dust observations could distinguish these cases.","Not in the paper: the same lensing technique applied to rest-frame optical or infrared data could test whether these nuclear structures sit inside larger stellar bars or discs, linking nuclear spirals to the large-scale morphology seen by other high-redshift surveys."],"forward_implications":["If the nuclear spiral and bar are real, compact rotationally supported discs existed at z=2.8, with effective radii near 0.6 kpc, 5 to 10 times smaller than typical stellar discs of dusty star-forming galaxies.","Secular dynamical processes could supply gas to the central 100 pc at rates consistent with the apparent super-Eddington starburst, reducing the need for major mergers to explain compact spheroid formation.","Strong lensing combined with long-baseline ALMA can reach roughly 55 pc in dusty star-forming galaxies, and smoothing to the native 30 mas resolution erases the spiral, so lensing is currently the only feasible route to these scales.","Applying the same approach to a sample of lensed dusty star-forming galaxies could determine how common nuclear discs, spirals, and bars are in this population.","Resolved gas kinematics at matched resolution would test whether the spiral arms and bar trace genuine non-circular gas flows rather than transient tidal debris."],"supporting_citations":[{"why":"Provides the ALMA data reduction, the ellipsoidal power-law plus shear and multipole lens model, and the posterior lens parameters that this paper re-inverts with three regularisation schemes.","marker":"Stacey et al. (2024)"},{"why":"Introduces the semi-linear pixellated lens-inversion method used to reconstruct the source surface brightness from interferometric visibilities.","marker":"Powell et al. (2021)"},{"why":"Generalises the method to include additional lens-model freedom and shows how foreground and mass-sheet effects appear, underpinning the robustness discussion.","marker":"Powell et al. (2022)"},{"why":"Gives the unresolved 350 GHz flux density, the SED-based obscured star formation rate of 700 plus or minus 200 solar masses per year, and a magnification estimate that the new measurement is compared with.","marker":"Reuter et al. (2020)"},{"why":"Identifies SPT0538-50 as a lensed dusty star-forming galaxy at z=2.78 with a z=0.414 foreground elliptical and provides the molecular gas mass and the lack of a clear AGN signature in the SED.","marker":"Bothwell et al. (2013)"},{"why":"Defines the Eddington limit for local ultraluminous infrared galaxies against which the central star formation rate density is called super-Eddington.","marker":"Barcos-Muñoz et al. (2017)"},{"why":"Motivates testing whether different regularisation choices bias lens models and source reconstructions, the robustness check that is central to this work.","marker":"Galan et al. (2024)"},{"why":"Simulations predict that secular inner-kiloparsec gas inflow feeds nuclear starbursts and black hole growth, the physical process the spiral and bar would realise.","marker":"Anglés-Alcázar et al. (2021)"},{"why":"Provides a high-redshift comparison where isophote ellipticity and position-angle behaviour are used to identify bars, the same diagnostic applied here.","marker":"Tsukui et al. (2024)"}],"fun_headline_variants":["Zooming into a galaxy's heart reveals nuclear spiral","55-pc view uncovers spiral arms in distant galaxy core","Galaxy core at z=2.78 shows spiral arms and a hint of a bar","Super-Eddington starburst in a 55-pc nuclear spiral","Lensed ALMA sees 55-pc spiral arms feeding a distant starburst"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the foreground lens can be treated as an isolated mass distribution; if it actually belongs to a galaxy group, the inferred physical size and star formation rate of the source would be rescaled, although the spiral morphology itself would survive.","fun_headline_variants_meta":{"raw":{"variants":["Zooming into a galaxy's heart reveals nuclear spiral","55-pc view uncovers spiral arms in distant galaxy core","Galaxy core at z=2.78 shows spiral arms and a hint of a bar","Super-Eddington starburst in a 55-pc nuclear spiral","Lensed ALMA sees 55-pc spiral arms feeding a distant starburst"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000553,"raw_usage":{"total_tokens":2665,"prompt_tokens":1004,"completion_tokens":1661,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":620,"completion_tokens_details":{"reasoning_tokens":1563}},"tokens_in":620,"tokens_out":1661,"duration_ms":12258,"temperature":1.0,"reasoning_tokens":1563,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T22:14:37.590575+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete test would be resolved molecular-gas kinematics, for example CO or [CII] line emission at matched sub-kiloparsec resolution: if the reconstructed spiral arms and bar do not trace coherent, rotating gas motions with non-circular streaming along the arms, the morphological features would be artefacts or transient tidal debris rather than a secular nuclear disc. Imaging the lens environment to establish whether the lens is in a group would also decide whether the mass-sheet degeneracy rescales the reported physical numbers.","supporting_citations":[{"cited_title":"R., Powell, D","cited_arxiv_id":null,"evidence_quote":"Provides the ALMA data reduction, the ellipsoidal power-law plus shear and multipole lens model, and the posterior lens parameters that this paper re-inverts with three regularisation schemes."},{"cited_title":"P., et al","cited_arxiv_id":null,"evidence_quote":"Introduces the semi-linear pixellated lens-inversion method used to reconstruct the source surface brightness from interferometric visibilities."},{"cited_title":"M., Vegetti, S., McKean, J","cited_arxiv_id":null,"evidence_quote":"Generalises the method to include additional lens-model freedom and shows how foreground and mass-sheet effects appear, underpinning the robustness discussion."},{"cited_title":"S., Aguirre, J","cited_arxiv_id":null,"evidence_quote":"Identifies SPT0538-50 as a lensed dusty star-forming galaxy at z=2.78 with a z=0.414 foreground elliptical and provides the molecular gas mass and the lack of a clear AGN signature in the SED."}],"review_version":1}