{"id":"c5597c4c-5b23-4eb6-add1-0d4ca9fed77a","arxiv_id":"2508.18374","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Simulated galaxies with on-the-fly dust evolution appear brighter-centered and more bulge-dominated in mock JWST images than with a fixed dust-to-metal ratio, lowering the measured late-type fraction at high redshift.","lead":"A galaxy-simulation study finds that how dust is modeled changes how simulated galaxies appear in JWST-like images: galaxies with time-evolving dust look brighter at their centers and less disk-like than the same galaxies with a simple fixed amount of dust. Many simulations use the simpler recipe, so the finding warns that shape comparisons between simulations and JWST may be biased by this choice.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"OTF dust subgrid fidelity is the key uncertainty: without sensitivity tests of dust destruction/growth rates, the DTM cavity and LTG-fraction shift may be model artifacts.","rationale":"The reader's conditional verdict is appropriate. The paper's central, novel assertion is that a physically motivated dust evolution model (OTF) produces a lower LTG fraction than a fixed DTM at high z, via a DTM cavity in starbursting bulges. This is a strong claim because it implies morphological classification depends on the dust model in a specific way, and that standard fixed-DTM approaches are biased. The strength of the claim rests on the OTF dust model being realistic. The NewCluster simulation implements a subgrid recipe for dust formation and destruction; the DTM cavity is the outcome of that recipe. If the recipe is calibrated to local observations and extrapolated to high-z starbursts, uncertainties in destruction efficiency and accretion timescale could easily change the cavity depth. The abstract does not report sensitivity tests, and the corrupt full text prevents verification. Thus the most load-bearing concern is not the internal differential comparison (which is well controlled) but the external interpretation that OTF is more physical. A concrete sensitivity study varying the key dust parameters would settle it. If the morphology shift survives such variations, the conclusion is robust; if not, the paper overstates the failure of fixed DTM models. No formal verification or code release is apparent from the abstract. Hence I concur with the CONDITIONAL verdict and suggest it remain.","tokens_in":18817,"tokens_out":6870,"duration_ms":77325,"concrete_test":"Run NewCluster (or post-processing dust re-simulation) with dust destruction efficiency and ISM accretion timescale each varied by factors of 0.5 and 2, holding all other physics fixed; regenerate SKIRT mock images and G-M20 classifications at z≈4–6. If the OTF-versus-fixed DTM difference in LTG fraction and central-surface-brightness remains within the sample's statistical errors across all variants, the DTM cavity and morphological shift are robust. If any variant flips or erases the effect, the central claim is parameter-dependent and must be qualified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central claim – that OTF dust models produce brighter bulges and lower late-type fractions at high redshift via a 'DTM cavity' – is only as credible as the subgrid dust physics in NewCluster. The cavity is the difference between dust destruction and dust growth in starbursting bulges; if those rates are mis-calibrated, the cavity (and hence the morphology shift) is an artifact. The abstract does not report any calibration or sensitivity analysis for these parameters, and the supplied full text is corrupted beyond reliable reading, so this cannot be checked. Furthermore, the conclusion that fixed DTM models 'fail' depends on the OTF model being the more physical one; absent robustness tests to the dust recipes, the failure could be model-dependent. The internal OTF-vs-fixed comparison is sound as a differential statement, but the interpretation as a physical failure of fixed DTM models requires showing the OTF result is not sensitive to the dust model's free parameters.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper asks whether the choice of dust model changes the apparent morphology of high-redshift galaxies in mock JWST images. Using the NewCluster simulation, the authors post-process the same galaxies with an on-the-fly (OTF) dust model and with a fixed dust-to-metal (DTM) ratio, run SKIRT radiative transfer, and apply the G-M20 morphological classification. They report that OTF models produce brighter centers and more pronounced bulges, lowering the late-type galaxy (LTG) fraction especially at high redshift, and attribute this to a 'DTM cavity'—a localized depression in the dust-to-metal ratio in starbursting bulges. They conclude that fixed-DTM models fail to capture key morphological features. The abstract is clear and the differential design is a strength, but the supplied full text is almost entirely garbled and unreadable, including an embedded reference to an unrelated arXiv paper. Consequently, the derivations, figures, tables, and quantitative results could not be independently checked.","tokens_in":18942,"tokens_out":4954,"duration_ms":62442,"significance":"If the result holds, it is significant: it would demonstrate that dust-evolution subgrid physics changes morphological classifications in a way that matters for JWST-based high-z studies, and it would identify a systematic bias in fixed-DTM mock observations. The internal differential comparison (same galaxies, two dust treatments) is well posed, and the use of SKIRT with a JWST comparison sample is appropriate. However, the interpretation that fixed-DTM models 'fail' depends on the OTF dust recipes faithfully representing dust in starbursting bulges. Because the supplied text is unreadable, the calibration and sensitivity of those recipes cannot be evaluated; this is the main barrier to accepting the paper's central claim.","major_comments":[{"comment":"The supplied full text is largely illegible mojibake; equations, figures, and tables cannot be read. An unrelated identifier, 'arXiv:2508.18373v1 [cond-mat.str-el]', appears embedded mid-manuscript, indicating contamination. This is not a trivial typo: it prevents verification of every quantitative claim in the paper. A clean manuscript must be provided before any scientific assessment can be made.","section":"Full text (passim)"},{"comment":"The statement that 'fixed DTM models fail to capture key morphological features' is stronger than the evidence described. The paper compares one OTF implementation against one fixed-DTM prescription. Without sensitivity tests varying the dust formation/destruction efficiencies and timescales, the failure could be specific to NewCluster's subgrid parameters. The abstract reports no such tests. Please either add robustness tests or soften the claim to state that fixed-DTM models can differ substantially from OTF models.","section":"Abstract (last sentence)"},{"comment":"The DTM cavity is introduced as the physical mechanism, but the abstract provides no quantitative information (depth, radial extent, redshift dependence) and no direct comparison of the simulated cavity to observational constraints on dust-to-metal ratios in bulges. The causal chain 'intense bulge starburst -> localized DTM depression -> central brightening -> lower LTG fraction' is asserted. If the full text contains the demonstration, it must be made legible; if not, this is a gap in the argument.","section":"Abstract ('DTM cavity')"},{"comment":"The claim of a lower LTG fraction at high redshift relative to JWST requires the comparison sample to be matched in stellar mass, redshift, rest-frame wavelength, PSF, and noise. None of these selection/matching criteria are visible in the abstract, and the full text is unreadable. Please state the sample selection and matching explicitly, as the morphology differences could be affected by sample mismatch.","section":"Abstract, JWST comparison"}],"minor_comments":[{"comment":"The phrase 'OTF galaxy models exhibit' should be rephrased to 'galaxies in the OTF models exhibit' or similar, for clarity.","section":"Abstract"},{"comment":"Define 'DTM cavity' at its first use, e.g., 'a depression in the dust-to-metal ratio (hereafter the DTM cavity)'.","section":"Abstract"},{"comment":"Remove the embedded 'arXiv:2508.18373v1 [cond-mat.str-el]' and any other extraneous or corrupted material.","section":"Full text"},{"comment":"Specify the fixed DTM ratio value and whether it is a constant or metallicity-dependent; the abstract does not indicate which assumption is used.","section":"Methods (fixed DTM model)"}],"recommendation":"major_revision","confidential_remarks":"The full text appears corrupted at the source, not merely in the OCR extraction; the embedded cond-mat arXiv identifier is a red flag. I recommend that the editor obtain a clean PDF or TeX source before sending the paper to referees. The authors should also be asked to provide sensitivity tests of the OTF dust parameters, because the central 'failure' claim depends on those parameters."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, what to know: this is a well-controlled simulation comparison showing that the fixed dust-to-metal ratio commonly used in high-z simulations biases morphology toward late-type disks. The OTF dust model produces brighter bulges and a lower late-type fraction, especially at high z. That's a useful result for anyone comparing mock JWST images to observed morphologies.\n\nThe new bit is the application: prior work has compared dust models, but not specifically through the lens of JWST morphology classifications with G-M20, and the DTM cavity—the localized depression in DTM ratio caused by bulge starbursts—is a clearly named mechanism. The design is the paper's biggest strength: the same NewCluster galaxies are imaged under both dust treatments with SKIRT, so the morphology difference isolates the dust treatment rather than cosmic variance or resolution differences.\n\nThe soft spots are real but not fatal. The interpretation that fixed DTM models 'fail' assumes the OTF subgrid recipes for dust formation and destruction are calibrated properly. The abstract doesn't report sensitivity tests for those rates, and if they're off, the DTM cavity could be a model artifact. That's worth asking for in revision—not a reason to reject. Also, the JWST comparison relies on sample and resolution matching, which the abstract doesn't detail. Minor: the DTM cavity is a post hoc description of the OTF dust field rather than a prediction, so it shouldn't be oversold as independent confirmation.\n\nI can't check the body because the version I received is corrupted; the abstract is coherent and the methodology is standard in the field. The internal control makes the differential claim robust, but the physical verdict rests on the subgrid dust model's fidelity. That's a common issue in this literature, and a referee can handle it.\n\nWho is this for? People working on high-z galaxy morphology, JWST comparisons, or dust in simulations. It doesn't change the field, but it's a good cautionary result. I'd engage with it if I were doing morphology comparisons. Recommend a serious referee. The paper deserves referee time, with a request for robustness tests and sample-matching details.","headline":"Well-controlled OTF-vs-fixed dust comparison shows dust treatment biases morphological classifications at high z, but the physical verdict rests on subgrid recipes we can't audit in the abstract.","tokens_in":19582,"tokens_out":3234,"would_cite":true,"duration_ms":36334,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A fixed dust ratio biases how simulated galaxies appear at high redshift","keywords":["dust-to-metal ratio","on-the-fly dust evolution","galaxy morphology","cosmological simulation","radiative transfer","high-redshift galaxies","mock JWST images","G-M20 classification"],"falsifier":"Resolved dust-to-metal ratio or attenuation maps of high-redshift star-forming bulges, from ALMA dust continuum or JWST NIRCam/MIRI observations, could settle it: if the centers of starbursting galaxies show no depression in dust-to-metal ratio relative to their disks, the DTM cavity predicted by the on-the-fly model is not present in real galaxies. A simpler internal check is to rerun the mock images with the dust opacity law or destruction efficiency varied and see whether the late-type fraction shift disappears.","tokens_in":18638,"feed_emoji":"🔭","tokens_out":3819,"duration_ms":43562,"temperature":0.7,"pith_summary":"The paper asks whether the common shortcut of giving every gas cell a fixed dust-to-metal ratio distorts how galaxies look when rendered as telescope images. Using the NewCluster cosmological simulation, which tracks dust formation and destruction on the fly, the authors make mock JWST images of the same galaxies under two dust treatments. They find that with on-the-fly dust, galaxy centers are brighter and bulges more pronounced, so fewer galaxies classify as late-type disks, especially at high redshift. The cause is a localized depression in the dust-to-metal ratio, called the DTM cavity, produced by intense bulge starbursts. If correct, fixed dust-ratio models systematically bias morphological comparisons against high-redshift observations.","feed_headline":"Dust modeling choice flips simulated galaxy types","feed_subtitle":"On-the-fly dust evolution brightens galaxy centers and cuts late-type fractions in mock JWST images.","key_machinery":"The DTM cavity: a localized depression in the dust-to-metal ratio in galaxy centers caused by intense bulge starbursts, which in the on-the-fly dust model reduces central dust attenuation and brightens the bulge relative to fixed-ratio models. The comparison is carried out with radiative transfer code SKIRT to make mock JWST images, followed by G-M20 morphology classification.","core_discovery":"The central claim is that adopting a fixed dust-to-metal ratio in post-processing changes the apparent morphology of simulated galaxies enough to alter their classification. In NewCluster, the physically evolved dust distribution leaves central regions less dusty relative to metals after bulge starbursts, so stellar light at the center is less attenuated; the same galaxies rendered with a fixed DTM show fainter centers and look more like disks. The G-M20 morphology test then assigns a lower late-type galaxy fraction to the on-the-fly sample, with the difference strongest at high redshift. The paper interprets this as evidence that fixed DTM models fail to capture a key morphological feature.","pith_inferences":["A natural next step is to test the DTM cavity directly with resolved dust or attenuation observations of high-redshift bulges; a clean detection would support the mechanism, while a flat dust-to-metal ratio in starburst centers would challenge it.","Re-deriving dust-to-metal ratios from metal maps in older fixed-ratio simulations may recover some of the missing central brightening without rerunning the hydrodynamics.","If the effect is general, it could also affect observables that depend on central attenuation, such as half-light radii, color gradients, or SED-derived stellar masses, though the paper does not test these.","The bias may be redshift-dependent in a predictable way, allowing observers to use morphology samples to constrain dust evolution timescales."],"forward_implications":["Mock JWST images from simulations that assume a fixed dust-to-metal ratio will systematically classify high-redshift galaxies as later types than they are.","The late-type fraction measured in simulations depends on dust modeling choices, not only on star formation and feedback physics.","Dust evolution needs to be included when rendering simulated galaxies, not only in the hydrodynamic run, to compare morphology with observations.","The DTM cavity predicts that bulge regions of high-redshift starbursting galaxies should show measurably lower dust-to-metal ratios than their disks.","At higher redshift, where starbursts are more frequent, the morphology bias from fixed dust ratios should grow."],"supporting_citations":[],"fun_headline_variants":["On-the-fly dust brightens galaxy centers in simulations","Fixed dust ratio makes galaxies appear more disk-like","Dust model choice flips simulated galaxy type fractions","Simulated galaxy morphology depends on dust evolution"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The on-the-fly dust model's subgrid recipes for dust formation, destruction, and grain opacity accurately represent real dust in starbursting bulges; if these recipes destroy or fail to re-form dust too aggressively, the DTM cavity and the resulting central brightening are artifacts of the model rather than a real feature of galaxies.","fun_headline_variants_meta":{"raw":{"variants":["On-the-fly dust brightens galaxy centers in simulations","Fixed dust ratio makes galaxies appear more disk-like","Dust model choice flips simulated galaxy type fractions","Simulated galaxy morphology depends on dust evolution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000257,"raw_usage":{"total_tokens":1414,"prompt_tokens":741,"completion_tokens":673,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":485,"completion_tokens_details":{"reasoning_tokens":624}},"tokens_in":485,"tokens_out":673,"duration_ms":7466,"temperature":1.0,"reasoning_tokens":624,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T16:28:30.181598+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Resolved dust-to-metal ratio or attenuation maps of high-redshift star-forming bulges, from ALMA dust continuum or JWST NIRCam/MIRI observations, could settle it: if the centers of starbursting galaxies show no depression in dust-to-metal ratio relative to their disks, the DTM cavity predicted by the on-the-fly model is not present in real galaxies. A simpler internal check is to rerun the mock images with the dust opacity law or destruction efficiency varied and see whether the late-type fraction shift disappears.","supporting_citations":[],"review_version":1}