{"id":"3acdcb16-e84b-445f-b71e-47ba58ed8be2","arxiv_id":"1908.09685","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"New ALMA observations show that the debris disks 49 Cet and HD 32297 harbor large CO reservoirs, supporting the shielded secondary gas disk model.","lead":"The authors detected rare CO molecules in two young debris disks and found each contains more than 0.01 Earth masses of CO gas. They argue this gas can be explained by collisions of icy planetesimals if the CO is shielded from UV light by carbon atoms and by itself.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The secondary-gas explanation hinges on an unmeasured, extremely low viscosity (alpha ~ 1e-5) that the model treats as a free parameter rather than deriving from its own ionization calculation.","rationale":"The observational part of the paper, detection of 13CO and C18O, optically thick 12CO, and CO masses above 0.01 Earth masses, appears robust and is not the weak point. The modeling conclusion is more fragile. Other possible concerns, such as the unreleased code, the marginal HD 32297 fit, or isotope-ratio uncertainties, are secondary or would move the result toward still-higher CO masses. The only assumption whose failure would break the explanation is the low alpha: without it, carbon does not accumulate, shielding does not switch on, and the high CO masses require primordial gas. The reader identified this same assumption; the additional point here is that the paper's own code is said to compute the ionization fraction, so the low alpha should be derivable rather than imposed. That is why the CONDITIONAL verdict is retained unchanged.","tokens_in":27398,"tokens_out":11963,"duration_ms":133202,"concrete_test":"Run the upgraded 1D model at the adopted best-fit Mdot_CO values (4.6e-3 Me/Myr for 49 Cet, 3.6e-2 for HD 32297), compute the radial ionization fraction from the model's UV transfer, and evaluate the MRI/ambipolar alpha using the Kral & Latter (2016) prescription. If the predicted alpha exceeds ~1e-4 (or, for HD 32297, exceeds the ~2e-5 upper limit), the low-alpha assumption is not self-consistent and the secondary-gas explanation fails. A minimal alternative: rerun the HD 32297 grid with alpha fixed to 1e-4 and test whether the predicted C18O mass falls below the observed 3-sigma range.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the CO content of 49 Cet and HD 32297 is explainable by a shielded secondary gas disk rests on the viscosity being extremely low. The acceptable grids in Fig. 4 require alpha ~ 2e-5 to 9e-5 for 49 Cet and alpha <~ 2e-5 for HD 32297 (Sect. 5.3.2). This alpha is not directly measured; it is argued only from the expectation that heavy shielding lowers the ionization fraction and thus MRI-driven transport (Sect. 5.3.2, citing Kral & Latter 2016). The new code is stated to compute the ionization fraction at every radial location, yet the reported model runs keep alpha as an independent grid parameter instead of deriving it from that computed ionization state. If the true alpha is only a few times higher, carbon viscously spreads before accumulating enough to shield CO, and the observed 13CO/C18O masses cannot be reproduced by a purely secondary disk. The paper's own Sect. 5.3.5 concedes that primordial H2 cannot be ruled out, so the secondary-only interpretation stands or falls with the low-alpha assumption. This is an input assumption, not a prediction of the model.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents ALMA 7m-array Band 6 observations of 12CO, 13CO, and C18O toward the debris disks around the young A-type stars 49 Cet and HD 32297. The measured 13CO/12CO and C18O/13CO line ratios imply that the 12CO emission is optically thick in both disks, and the optically thin isotopologue fluxes lead to inferred total CO masses of about 0.011 M_earth (49 Cet) and 0.074 M_earth (HD 32297), i.e. above the 0.01 M_earth threshold that defines the previously identified CO-rich debris disk class. The authors argue that such high CO masses require strong shielding, and they adapt the Kral et al. (2018) secondary gas disk model, now including radial viscous evolution of 12CO, 13CO, C18O, and atomic carbon plus isotope-selective photodissociation, to see whether the observed isotopologue masses can be reproduced without primordial H2. A grid over CO production rate and viscosity parameter alpha yields an acceptable region for 49 Cet that also matches the observed CI line flux and inner CO radius, while for HD 32297 the C18O mass can be matched only with very low alpha and the agreement is admitted to be marginal. The paper then discusses gas-dust coupling, concluding that small grains in HD 32297 are plausibly coupled to the gas, and argues that the two systems are new representatives of shielded debris disks, while noting that a primordial origin cannot be excluded.","tokens_in":27545,"tokens_out":6252,"duration_ms":69691,"significance":"If the modeling claim holds, the paper significantly enlarges the sample of CO-rich debris disks with measured isotopologues from four to six and provides the first such measurements for 49 Cet and HD 32297. The observational analysis is careful: line flux ratios, optical depth arguments, and mass estimates from optically thin isotopologues are standard and the systematic uncertainties are discussed. The upgraded model is a genuine step forward in adding radial evolution and isotope-selective photodissociation to secondary gas disk modeling, and it makes testable predictions for CI emission and for the radial extent of CO versus C0. The independent constraints used for 49 Cet (CI line flux and inner CO radius) are valuable because they are not simply the mass being fitted. However, the central explanation is partly constructed by construction: the model's production rate and alpha are varied to match the very masses that are then said to be explained, and for HD 32297 the surviving parameter region depends on an unmeasured and extremely low viscosity.","major_comments":[{"comment":"The central demonstration that the shielded secondary gas disk model explains the observed CO content is partly circular: the model grid is scored against the same 13CO (49 Cet) and C18O (HD 32297) masses that the free parameters Mdot_CO and alpha are varied to reproduce. The solid black contours therefore trace the fitted relation rather than a prediction. The independent constraints (CI line flux and inner CO radius for 49 Cet; inner radius alone for HD 32297) are what rescue the exercise, but for HD 32297 the acceptable region is essentially defined by the requirement alpha <~ 2e-5 and the authors themselves call the agreement marginal. Please separate the fitted constraints from the predictive checks and state explicitly how much parameter space survives if the isotopologue mass contours are omitted.","section":"Sect. 5.3.2, Fig. 4"},{"comment":"The extremely low viscosity, alpha ~ 1e-5 to 9e-5, is an input assumption rather than a model output. Although the text states that the new code computes the ionization fraction at every radial location, alpha is treated as an independent grid parameter and is not derived from that ionization state. For HD 32297 the required alpha is below about 2e-5, and the supporting argument from ambipolar diffusion and low ionization is only qualitative. Please add a sensitivity analysis showing whether any Mdot_CO can reproduce the observed masses for, e.g., alpha = 1e-4 or 1e-3, or alternatively derive alpha from the computed ionization balance. Without such a test, the secondary-only interpretation rests on an untested assumption about the disk's transport properties.","section":"Sect. 5.3.1 and 5.3.2"},{"comment":"The model predicts that isotope-selective photodissociation lowers the 13CO/12CO and C18O/12CO ratios below the ISM values used in Section 4.2, and as a result the simulated 12CO masses are about an order of magnitude higher than the observationally inferred masses quoted in Table 1 and Figure 2. This is acknowledged in Section 5.3.3, but the abstract and Section 6 should not present the Section 4.2 masses as the definitive CO content without noting that the adopted model implies substantially larger 12CO masses. The discrepancy does not weaken the 'optically thick' conclusion, but it affects the quantitative comparison with protoplanetary disks and the shielding factors derived in Section 5.2.","section":"Sect. 5.3.3 and Sect. 4.2"},{"comment":"The abstract's and Section 6's framing that the secondary gas disk model can explain the observed CO levels is too strong given the paper's own admission in Section 5.3.5 that primordial H2 cannot be ruled out. The observations and model demonstrate consistency with a secondary origin, but they do not uniquely establish it, because the same CO masses and shielding factors can in principle be produced by leftover hydrogen. Please temper the language to 'consistent with' rather than 'can explain' in the abstract, or add an explicit statement that the two scenarios are currently degenerate and identify the specific observations that would break the degeneracy.","section":"Sect. 5.3.5 and Sect. 6"}],"minor_comments":[{"comment":"There are several typographical issues, e.g. 'obser v ations' in the title and section headings and 'neccessarily' in Section 5.2; these should be corrected in the final version.","section":"Throughout"},{"comment":"The caption contains run-together notation such as 'F12CO(2-1) = 1.05Jykms-1' and 'i=84º'; please add spaces and consistent units for readability.","section":"Fig. 4 caption"},{"comment":"The phrase 'CO mass estimates are from Fig. 2a' appears to reference a panel that does not exist in Figure 2; please correct the cross-reference.","section":"Sect. 5.2"},{"comment":"The upgraded model is described as being presented in detail in a forthcoming paper, and no code or data products are made public. Since several central results depend on these numerical simulations, a public code release or a detailed appendix would substantially improve reproducibility.","section":"Sect. 5.3.1"},{"comment":"The sentence 'Since with the adopted temperatures the fractional population is close to its maximum in LTE, these deviations can typically lead to higher gas masses' is slightly ambiguous; it would be clearer to state explicitly that lower x2 values correspond to higher inferred masses.","section":"Sect. 4.2"}],"recommendation":"major_revision","confidential_remarks":"The observational data and the line-ratio analysis are solid and publishable. The main weakness is the inferential loop in Section 5.3: the model is tuned to the isotopologue masses, and the very low viscosity required for HD 32297 is not independently measured or derived from the model's own ionization calculation. The stress-test concern about alpha is valid and should be addressed with explicit sensitivity tests. This is a case where the manuscript's scope can accommodate the fix, so I recommend major revision rather than rejection. The paper would also benefit from a more careful distinction between a consistency check and a unique explanation, especially in the abstract, because the authors themselves concede that primordial H2 remains a viable alternative."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper gives you the first 13CO and C18O detections in 49 Cet and HD 32297, and those data robustly push both disks into the high-CO, 'shielded debris disk' class. The observational part is the real contribution; the secondary-gas model is plausible but the load-bearing assumption is an extremely low viscosity that is not measured.\n\nThe new observations are carefully done. Line flux ratios clearly imply optically thick 12CO; masses derived from the optically thin isotopologues are ~1.1e-2 and ~7.4e-2 M⊕ for 49 Cet and HD 32297. The authors check against earlier SMA/JCMT/ALMA data and the agreement is good. They also flag the isotope-selective photodissociation uncertainty honestly. These results expand the known shielded-debris-disk sample from four to six and are the strongest part of the paper.\n\nThe model section is where I wince. They fit Mdot_CO and alpha to the observed 13CO/C18O masses. The grid search is constrained for 49 Cet by the CI line flux and inner CO radius, which gives some independent leverage. But for HD 32297 the match is marginal, as the authors admit, and it needs alpha ≲ 2e-5. That low alpha is argued from the expectation that shielding lowers the ionization fraction, but it is not measured, and the model treats it as a free grid parameter even though the code computes ionization fractions. So the stress-test note is on target: the secondary-only explanation stands or falls with that alpha. The paper itself concedes in Sect. 5.3.5 that primordial H2 cannot be ruled out. That is the right level of caution, but it means the central model is a demonstration of viability, not a unique explanation.\n\nWho should read this? Anyone working on debris disk gas, volatile delivery, or gas-dust coupling. The CO mass and isotopologue data will be citable regardless of which model wins. I would not desk-reject this; it deserves a serious referee, mainly to push for the code release or at least a full description of the upgraded model, and for the low-alpha assumption to be flagged as a caveat rather than a result. The observational core is solid and the authors are candid about the model's limits.","headline":"Solid new isotopologue detections that double the shielded-debris-disk sample, wrapped in a plausible but viscosity-sensitive model that the authors themselves hedge.","tokens_in":28239,"tokens_out":2930,"would_cite":true,"duration_ms":31305,"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":"Carbon shielding explains CO-rich debris disks around two young stars","keywords":["debris disks","CO gas","secondary gas","carbon shielding","photodissociation","ALMA","49 Cet","HD 32297"],"falsifier":"Measure the non-thermal line width of the CO (2-1) emission in HD 32297 with sufficient spectral resolution to constrain $\\alpha$: if the implied viscosity exceeds about $10^{-4}$, the carbon shield cannot build up and the observed CO mass would not be sustainable under the secondary-only scenario.","tokens_in":27113,"feed_emoji":"🔭","tokens_out":10347,"duration_ms":92533,"temperature":0.7,"pith_summary":"This paper reports ALMA observations of two gas-rich debris disks, 49 Cet and HD 32297, and finds that each holds more than one hundredth of an Earth mass of carbon monoxide. Such large CO masses ought to be destroyed by ultraviolet light within about a century, so something must be shielding the molecules. The authors argue that a purely secondary gas disk—CO and other volatiles released from colliding icy planetesimals—can explain the observed gas, as long as carbon atoms produced by the CO destruction accumulate and block the UV, with CO shielding itself as well. If this is right, the most CO-rich debris disks need no leftover primordial hydrogen gas from the planet-forming phase, and they form a distinct class of shielded debris disks.","feed_headline":"Carbon shielding explains CO-rich debris disks around two young stars","feed_subtitle":"New ALMA data show carbon shielding from comet collisions can preserve CO, without leftover hydrogen.","key_machinery":"The central mechanism is the shielded secondary gas disk: CO and CO$_2$ released from colliding icy planetesimals are photodissociated into carbon and oxygen, and the neutral carbon atoms, if they accumulate faster than they spread, become optically thick to ultraviolet radiation and shield the remaining CO; CO molecules also self-shield against dissociation. The paper's specific new tool is a one-dimensional viscous evolution code that tracks $^{12}$CO, $^{13}$CO, C$^{18}$O, and C$^0$ together at all radii, treating isotope-selective photodissociation for the first time in a debris disk model, which allows the rare-isotope line fluxes to be fit directly.","core_discovery":"Using the ALMA 7-m array, the team detected $^{13}$CO toward 49 Cet and HD 32297 and C$^{18}$O toward HD 32297, showing that the $^{12}$CO emission is optically thick and implying total CO masses of about $0.011\\,M_\\oplus$ and $0.074\\,M_\\oplus$ respectively. They then adapted the shielded secondary gas disk model of Kral et al. (2018) into a one-dimensional viscous model that follows $^{12}$CO, $^{13}$CO, C$^{18}$O, and neutral carbon at every radius, including isotope-selective photodissociation. The paper's central claim is that this model reproduces the observed isotopologue masses with CO production rates of roughly $0.005$–$0.03\\,M_\\oplus$/Myr and a low viscosity parameter ($\\alpha \\lesssim 10^{-4}$), because neutral carbon and CO self-shielding extend the CO lifetime by factors of several hundred to a few thousand. In the case of 49 Cet, the predicted CI line flux also matches the observed value, adding a third independent constraint. The authors thus conclude that the high CO content of both disks can be understood without primordial hydrogen gas, though they do not rule it out.","pith_inferences":["If the low viscosities ($\\alpha \\approx 10^{-5}$–$10^{-4}$) are real, similar carbon-shielding may operate in other moderately gas-rich debris disks, and the absence of CO in dust-rich disks like HR 4796 could reflect either lower production rates or higher viscosity rather than a lack of icy planetesimals.","A direct measurement of turbulent line broadening in the CO lines of HD 32297 with ALMA at higher spectral resolution could test the assumed viscosity, since the model's $\\alpha$ range corresponds to subsonic, but measurable, non-thermal velocity dispersions.","The same mechanism might protect other volatile molecules (e.g., CO$_2$ or H$_2$O fragments) from photodissociation, implying that shielded debris disks could be identifiable by their full volatile inventory, not just CO."],"forward_implications":["The two disks join four previously known CO-rich debris disks as shielded debris disks, a class that can be powered entirely by second-generation gas from planetesimal collisions.","The required CO production rates translate to CO+CO$_2$ ice mass fractions of about 5% for 49 Cet and 0.7% for HD 32297, comparable to the ice content of solar-system comets.","In HD 32297 the derived gas surface density gives a Stokes number $St \\lesssim 1$ for the smallest bound grains, meaning gas drag should shape the dust distribution; this can explain the warm small grains observed in that disk.","The model predicts that neutral carbon spreads farther than CO, so resolved CI maps should show a more extended gas halo; this is a testable signature of the shielding mechanism."],"supporting_citations":[{"why":"Supplies the shielded secondary gas disk model in which neutral carbon and CO self-shielding extend the CO lifetime; the paper adapts this model to a 1-D viscous code.","marker":"Kral et al. 2018"},{"why":"Provides the photodissociation timescales and isotope-selective shielding treatment used to compute CO and isotopologue survival.","marker":"Visser et al. 2009"},{"why":"Provides the resolved $^{12}$CO map and disk geometry for 49 Cet, including the inner CO radius used as a model constraint.","marker":"Hughes et al. 2017"},{"why":"Gives the ALMA observations of HD 32297 that locate the CO with the dust belt and supply the disk inclination and belt radii.","marker":"MacGregor et al. 2018"},{"why":"Measures the CI line flux toward 49 Cet, used as an independent constraint on the model's carbon content.","marker":"Higuchi et al. 2017"},{"why":"Provides the cometary ice mass fractions that set the required CO production rates and make the model physically plausible.","marker":"Mumma & Charnley 2011"},{"why":"Established the very high CO masses in young debris disks and the hybrid-disk hypothesis that the secondary model aims to replace.","marker":"Kóspál et al. 2013"}],"fun_headline_variants":["Carbon shield keeps CO alive in two debris disks","CO-rich debris disks decoded: shielding is key","ALMA reveals carbon protection for disk CO","Shielded debris disks: no hydrogen needed for CO"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The model requires the gas to spread very slowly, with a viscosity parameter $\\alpha$ between about $10^{-5}$ and $10^{-4}$, a value inferred from the low ionization in shielded gas but not directly measured; if the true viscosity were much higher, carbon would disperse and the shielding would fail.","fun_headline_variants_meta":{"raw":{"variants":["Carbon shield keeps CO alive in two debris disks","CO-rich debris disks decoded: shielding is key","ALMA reveals carbon protection for disk CO","Shielded debris disks: no hydrogen needed for CO"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000149,"raw_usage":{"total_tokens":1228,"prompt_tokens":1015,"completion_tokens":213,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":631,"completion_tokens_details":{"reasoning_tokens":153}},"tokens_in":631,"tokens_out":213,"duration_ms":3019,"temperature":1.0,"reasoning_tokens":153,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:04:25.496632+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the non-thermal line width of the CO (2-1) emission in HD 32297 with sufficient spectral resolution to constrain $\\alpha$: if the implied viscosity exceeds about $10^{-4}$, the carbon shield cannot build up and the observed CO mass would not be sustainable under the secondary-only scenario.","supporting_citations":[{"cited_title":"Imaging [CI] around HD 131835: reinterpreting young debris discs with protoplanetary disc levels of CO gas as shielded secondary discs","cited_arxiv_id":"1811.08439","evidence_quote":"Supplies the shielded secondary gas disk model in which neutral carbon and CO self-shielding extend the CO lifetime; the paper adapts this model to a 1-D viscous code."},{"cited_title":"A., Weinberger, A","cited_arxiv_id":null,"evidence_quote":"Gives the ALMA observations of HD 32297 that locate the CO with the dust belt and supply the disk inclination and belt radii."},{"cited_title":"E., Sato, A., Tsukagoshi, T., et al","cited_arxiv_id":null,"evidence_quote":"Measures the CI line flux toward 49 Cet, used as an independent constraint on the model's carbon content."}],"review_version":1}