{"id":"196090cd-b4a2-42b9-82b9-4f6fbeba2923","arxiv_id":"2412.07853","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"JWST MINDS spectra show that inner disk molecular compositions vary widely, with very low-mass star disks dominated by hydrocarbons indicative of C/O greater than 1.","lead":"This proceedings paper summarizes early JWST MIRI results from the MINDS program on the chemistry of the inner regions of planet-forming disks. It reports a wide diversity in molecular spectra and argues that disks around very low-mass stars are rich in hydrocarbons, suggesting carbon-rich gas with a carbon-to-oxygen ratio greater than one.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The volatile C/O>1 inference for VLMS disks is not yet uniquely established: oxygen (H2O) is hidden under the hydrocarbon forest, so the chemical-model threshold has not been tested against a retrieval that lets H2O vary freely.","rationale":"The reader's CONDITIONAL verdict already captures the main risk, and I would keep that verdict. The paper is a proceedings review, not an original measurement, and it is careful: it identifies the H2O retrieval problem, the co-location caveat, and the reliance on 'subm.' papers. The observed hydrocarbon-dominated spectra around VLMS disks are a strong dataset-level result, and the molecular assignments are supported by published work (Tabone et al. 2023; Arabhavi et al. 2024; Kanwar et al. 2024a). The central chemical inference, however, is not yet secured because the oxygen side of the budget is observationally hidden. The reader's weakest assumption focused on the completeness of the chemical models and slab-fit degeneracies; I agree with that but would narrow the concern to an oxygen-inventory degeneracy: H2O lines are expected beneath the hydrocarbon forest, and until a retrieval includes H2O as a free parameter, the claim that C/O>1 is required does not follow from the spectra alone. This is a reason to keep the verdict CONDITIONAL, not to accept or reject the paper. My read does not change the reader's verdict, so I mark it UNCHANGED.","tokens_in":7388,"tokens_out":6366,"duration_ms":66570,"concrete_test":"For one well-characterized VLMS disk (e.g., J16053215-1933159 or a strong Arabhavi et al. 2024 source), run a simultaneous Bayesian retrieval of the full 5-20 micron MIRI-MRS spectrum with H2O column density as a free parameter, alongside the published C2H2, CH4, C2H4, C2H6, C4H2, C6H6, CO2, and continuum parameters, using LTE slab models with a temperature gradient. Compute the Bayesian evidence for a C/O>1 model family versus a C/O<=1 model family that includes the same line list and H2O opacity. If the C/O<=1 fit is favored or comparable, or if the recovered N(H2O) is within a factor of a few of the stellar C/O value, then the volatile C/O>1 inference is not uniquely established; if the C/O>1 fit wins decisively with N(H2O) constrained low, the concern is retired.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central astrochemical claim is that the VLMS disk spectra imply volatile C/O>1 (Abstract; §3.2). This is a review of published results, not a new analysis, and the text honestly flags its own limits, but the claim depends on a chain: (i) the slab-model fits of the hydrocarbon bands are unique; (ii) the chemical models of Kanwar et al. (2024b) correctly map observed hydrocarbon columns to C/O; and (iii) the oxygen inventory is genuinely low. Link (iii) is explicitly unresolved: the paper states that 'Quantification of these models requires better constraints on the actual H2O emission which is hidden below the forest of hydrocarbon lines in VLMS disks.' The statement that 'H2O is weak, if detected at all' is an upper limit on visibility, not on column density. If H2O is present but blended into the pseudo-continuum, the retrieved hydrocarbon column could be biased, and the inferred carbon enhancement could vanish or be strongly reduced. Link (ii) is also external: the 'only if C/O>1' result comes from a single published chemical model, and the paper does not show that alternative C/O<1 pathways, such as photochemical C2H2 production or a vertically stratified dry upper layer, are excluded. The claim is appropriately hedged as 'suggesting', but the headline 'volatile C/O>1' would be unsupported if a retrieval that includes H2O as a free parameter fits the full MIRI spectrum with C/O<1.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This IAU symposium proceedings paper summarizes results from the JWST MINDS MIRI-MRS GTO program on the chemical composition of warm inner disk gas. It reports a large spectral diversity among T Tauri disks, with some dominated by water lines and others by CO2, and highlights that disks around very low-mass stars (<0.3 Msun) show hydrocarbon-dominated spectra including C2H2, C4H2, C6H6, and newly detected C2H6. The paper interprets these VLMS spectra as suggesting volatile C/O>1 in the emitting layers, discusses mechanisms (soot line, hydrocarbon-rich ice drift, oxygen depletion in dust traps), and connects the observations to scenarios of pebble drift and dust trapping. It also reports water detection in inner regions of transitional disks, concluding that dust traps are 'leaky.' The paper is explicitly a review of already published and submitted work, with figures and models taken from the cited team papers.","tokens_in":7732,"tokens_out":5731,"duration_ms":82530,"significance":"If the C/O>1 inference holds, it has direct consequences for the compositions of planets forming around the most common stars in the Galaxy, making this a timely and important synthesis. The paper's strengths are its concise summary of a major GTO program, its explicit caveats about molecule co-location and about the hidden H2O emission, and its reliance on published, checkable analyses (Grant et al., Temmink et al., Tabone et al., Kanwar et al.). It also articulates falsifiable predictions: better constraints on H2O emission will test the carbon-rich interpretation. As a review, it does not present new data or derivations, and its significance is primarily as a status report and interpretative synthesis for the community.","major_comments":[{"comment":"The statement 'Chemical modeling shows that such high abundances of hydrocarbon molecules can only be achieved if C/O >1 in the emitting layers (Kanwar et al. 2024b)' is stronger than the evidence presented in this review. The paper immediately acknowledges that the H2O emission is hidden below the hydrocarbon forest and that quantification requires better H2O constraints, meaning the oxygen inventory is not directly measured. As written, 'only if' asserts a uniqueness that neither the slab-model fits nor the cited model establish within this paper. Recommend rewording to something like 'are reproduced by chemical models with C/O >1 in the emitting layers' and explicitly noting that this is a model-dependent interpretation pending direct H2O measurements. This is load-bearing because the abstract's headline 'suggesting volatile C/O>1' rests on this inference.","section":"§3.2"}],"minor_comments":[{"comment":"The phrase 'H2O is weak, if detected at all' refers to line visibility in the observed spectra; since the paper later notes H2O may be hidden below the hydrocarbon forest, please clarify that this is not a column-density upper limit.","section":"§3.2"},{"comment":"The adverb 'clearly' in 'dust traps are clearly “leaky”' is too strong given that the supporting systematic study (Gasman et al., subm.) is still in submission and not available for scrutiny; suggest 'appear to be leaky' or 'are inferred to be leaky in the submitted analysis.'","section":"§3.1"},{"comment":"There is a typo in the second paragraph: 'with respected to hydrogen' should be 'with respect to hydrogen.'","section":"§1"},{"comment":"Several cited works are marked 'subm.' (Vlasblom et al., Gasman et al., Arabhavi et al.) but have no reference list entries; please add full citations or a footnote with author lists and availability dates.","section":"References"},{"comment":"The caption notes that the spectra are normalized to peak emission; it would be helpful to state explicitly in the text that this normalization suppresses absolute line-strength differences between objects, which is relevant for comparing, e.g., GW Lup with DR Tau.","section":"Figure 2"}],"recommendation":"minor_revision","confidential_remarks":"This is a concise proceedings summary of a major JWST GTO program, and the scientific content is sound as a review. The stress-test concern about the C/O>1 inference partially lands: the paper itself acknowledges the hidden H2O and the reliance on external models, but the body-text sentence 'only if C/O>1' overstates the uniqueness and should be aligned with the more hedged abstract. The submitted-work citations need full entries for a published proceedings. The heavy self-citation is appropriate for a program-status paper. I recommend minor revision rather than major revision because the central claim is defensible and the needed fixes are local wording and reference updates."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a review, not a new result, but it is a good review. The MINDS team's synthesis of their own JWST papers is clearly written, the data are traceable to peer-reviewed work, and the paper is unusually honest about where its own central claim is soft.\n\nWhat I found new: the 'leaky dust traps' argument. The observation that water shows up in the inner disks of PDS 70 and SY Cha despite large cavities, and that all gapped disks in the sample seem to have inner-disk water, is a real cross-paper synthesis that had not been stated this sharply before. The paper also earns credit for flagging that the various molecules are not necessarily co-located, and for stating directly that H2O is hidden below the hydrocarbon forest in VLMS disks.\n\nThe soft spot is exactly the one you'd suspect. The headline claim that VLMS disks have volatile C/O > 1 rests on chemical models (Kanwar et al. 2024b) and on slab-model fits, and the oxygen inventory is not directly constrained because H2O is blended. The stress-test note is right that 'H2O is weak, if detected at all' is an upper limit on visibility, not on column density. But the paper itself says this. It explicitly says quantification requires better constraints on H2O. So the claim is hedged appropriately. The stronger version—that C/O > 1 is uniquely established—would be unjustified, but that's not what the paper claims. Still, I'd want the companion papers (Arabhavi et al., submitted, and Kanwar et al. 2024b) before citing the C/O > 1 result as settled. The 'only if C/O > 1' from one chemical model is a single-chain inference; alternative pathways like photochemical C2H2 production or a stratified dry upper layer are not excluded.\n\nNothing here is circular. The paper is a derivative review of the team's own published work, but the cited analyses are independent and checkable. No equations reduce to input assumptions.\n\nWho's it for: anyone who wants a one-stop summary of the MINDS inner-disk results without reading six papers. It's a proceedings contribution, so don't expect new measurements. I'd send it to review rather than desk reject, because the synthesis is useful and the C/O>1 claim, even hedged, is the kind of thing a referee should sanity-check. The hedges and caveats are there; a referee would mostly verify that the summary matches the cited papers.\n\nRecommendation: engage with it as a review, cite it for the synthesis, and check the companion papers before leaning on the C/O>1 inference.","headline":"A clearly written proceedings review that adds no new data but offers a useful synthesis; the headline C/O>1 claim for VLMS disks is real but the paper itself flags its own soft underbelly.","tokens_in":8279,"tokens_out":2485,"would_cite":true,"duration_ms":23951,"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":"JWST observations of planet-forming disks around very low-mass stars are dominated by hydrocarbon emission, implying carbon-rich gas with $C/O > 1$ in the inner disk.","keywords":["protoplanetary disks","planet formation","inner disk chemistry","JWST MIRI spectroscopy","very low-mass stars","C/O ratio","hydrocarbons","mid-infrared spectroscopy"],"falsifier":"A high-resolution mid-infrared spectrum of a very low-mass star disk such as J1605 that cleanly separates H2O lines from the hydrocarbon pseudo-continuum would settle the matter: if H2O column densities come out comparable to C2H2, the C/O > 1 inference collapses.","tokens_in":1747,"feed_emoji":"🪐","tokens_out":5119,"duration_ms":88564,"temperature":0.7,"pith_summary":"This review of the JWST MINDS program synthesizes infrared spectra of the innermost few au of planet-forming disks. It reports a stark split: disks around solar-mass T Tauri stars are generally oxygen-rich, with H$_2$O and CO$_2$ dominant, while disks around very low-mass stars ($<0.3\\,M_\\odot$, late-M type like Trappist-1) are dominated by C$_2$H$_2$, larger hydrocarbons, and even benzene. The paper argues that such hydrocarbon-rich spectra require a volatile carbon-to-oxygen ratio $C/O > 1$ in the emitting gas. If correct, rocky planets forming in these disks would accrete from carbon-rich gas, with consequences for their bulk composition. The data also show that dust traps are leaky and that radial drift of icy pebbles is not the whole story.","feed_headline":"JWST shows planet-forming disks of tiny stars are carbon-rich","feed_subtitle":"Mid-infrared spectra of late-M star disks are dominated by C2H2 and benzene, implying C/O > 1.","key_machinery":"The central observational object is the MIRI medium-resolution spectrum (5--28 $\\mu$m) of the inner disk, fitted with LTE slab models that assign a single temperature, column density, and emitting area to each molecule. The chemical diagnostics are the relative band strengths of H$_2$O, CO$_2$, C$_2$H$_2$, HCN, and benzene; the C$_2$H$_2$ $\\nu_4+\\nu_5$ and $\\nu_5$ bands and the pseudo-continuum of CH$_4$/C$_2$H$_4$ lines are used to infer very high hydrocarbon column densities. The interpretive machinery is the $C/O$ ratio as the controlling variable: thermochemical models show that large hydrocarbon abundances require $C/O > 1$, and the paper links that condition to scenarios of icy-pebble drift, dust trapping, and hydrocarbon-grain destruction at a warm 'soot line'.","core_discovery":"The paper's central claim is that inner-disk gas chemistry is set by stellar mass and disk structure, not by a single universal process. JWST/MIRI spectra from the MINDS program show a clear dichotomy: T Tauri disks typically have strong H$_2$O and CO$_2$ emission with $C/O < 1$, whereas very low-mass star disks show a forest of hydrocarbon emission (C$_2$H$_2$, CH$_4$, C$_2$H$_4$, C$_2$H$_6$, C$_4$H$_2$, C$_6$H$_6$) with weak or absent H$_2$O. Modeling of these spectra implies $C/O > 1$ in the emitting layers, meaning either enhanced carbon delivery or oxygen depletion. The paper treats this as evidence that planet-forming material in the inner regions of late-M star disks is carbon-rich.","pith_inferences":["If $C/O > 1$ in very low-mass star disks is confirmed, the Trappist-1 planets, which likely formed in such a disk, could have accreted carbon-rich gas as solids; their present-day compositions may preserve a record of that chemistry.","The hydrocarbon pseudo-continuum may hide weak H$_2$O lines; higher-resolution or line-selective analyses could test whether the apparent oxygen depletion is real or partly a degeneracy in the slab-model fits.","A direct comparison between inner-disk $C/O$ values inferred from JWST and the atmospheric $C/O$ ratios of exoplanets around M dwarfs could test the disk--planet composition link; such a comparison is not made in this paper.","The 'soot line' mechanism predicts a sharp temperature boundary; spatially resolved JWST observations across the inner disk could look for a radial transition in hydrocarbon abundance."],"forward_implications":["The inner disks of very low-mass stars are carbon-rich, so planets forming there accrete gas with $C/O > 1$, altering their atmospheric and possibly bulk composition.","Terrestrial planets forming in such environments may end up carbon-poor if most solid carbon is removed from the disk in the form of carbon-rich gas.","The diversity between water-rich and hydrocarbon-rich disks reflects stellar mass and disk evolution, not just radial drift of icy pebbles.","Dust traps are leaky: even gapped disks like PDS 70 and SY Cha deliver water and other volatiles to the inner disk.","ALMA alone cannot determine the composition of the planet-forming zone; JWST mid-IR spectroscopy is required to probe the innermost few au."],"supporting_citations":[{"why":"First JWST detection of the hydrocarbon-rich disk J1605, including the broad C2H2 bumps and first detections of C4H2 and benzene.","marker":"Tabone et al. 2023"},{"why":"Provides subsequent VLMS disk examples with rich hydrocarbon spectra, supporting the generality of the carbon-rich finding.","marker":"Arabhavi et al. 2024"},{"why":"Identifies the pseudo-continuum from CH4 and C2H4 and reports the first C2H6 detection in a disk.","marker":"Kanwar et al. 2024a"},{"why":"Chemical modeling that shows hydrocarbon abundances this high require C/O > 1 in the emitting layers.","marker":"Kanwar et al. 2024b"},{"why":"Spitzer spectra gave early hints of bright C2H2 emission from very low-mass star disks.","marker":"Pascucci et al. 2013"},{"why":"Proposes the 'soot line' mechanism for carbon enhancement via destruction of hydrocarbon grains at roughly 500 K.","marker":"Li et al. 2021"},{"why":"Provides the T Tauri disk spectra with strong CO2 and isotopologs that anchor the oxygen-rich side of the dichotomy.","marker":"Grant et al. 2023"},{"why":"Detection of water in the inner disk of the gapped PDS 70 disk, demonstrating that dust traps are leaky.","marker":"Perotti et al. 2023"}],"fun_headline_variants":["JWST finds carbon-rich gas in planet-forming zones of tiny stars","Tiny stars' planet-forming disks are carbon-rich, JWST shows","JWST reveals carbon-rich planet-building gas around tiny stars","Hydrocarbon-rich inner disks found around very low-mass stars","Low-mass star disks show carbon-rich gas, JWST MINDS finds"],"cache_read_input_tokens":10368,"weakest_assumption_plain":"The load-bearing premise is that the chemical models used to interpret the spectra capture all relevant reactions, so the inferred carbon enhancement is real and not an artifact of missing chemistry or of the simplified temperature and column-density fits.","fun_headline_variants_meta":{"raw":{"variants":["JWST finds carbon-rich gas in planet-forming zones of tiny stars","Tiny stars' planet-forming disks are carbon-rich, JWST shows","JWST reveals carbon-rich planet-building gas around tiny stars","Hydrocarbon-rich inner disks found around very low-mass stars","Low-mass star disks show carbon-rich gas, JWST MINDS finds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00092,"raw_usage":{"total_tokens":3927,"prompt_tokens":904,"completion_tokens":3023,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":520,"completion_tokens_details":{"reasoning_tokens":2934}},"tokens_in":520,"tokens_out":3023,"duration_ms":20971,"temperature":1.0,"reasoning_tokens":2934,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:28:43.002131+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-resolution mid-infrared spectrum of a very low-mass star disk such as J1605 that cleanly separates H2O lines from the hydrocarbon pseudo-continuum would settle the matter: if H2O column densities come out comparable to C2H2, the C/O > 1 inference collapses.","supporting_citations":[{"cited_title":"S., & Bruderer, S","cited_arxiv_id":null,"evidence_quote":"Spitzer spectra gave early hints of bright C2H2 emission from very low-mass star disks."},{"cited_title":"L., van Dishoeck, E","cited_arxiv_id":null,"evidence_quote":"Provides the T Tauri disk spectra with strong CO2 and isotopologs that anchor the oxygen-rich side of the dichotomy."}],"review_version":1}