REVIEW 3 major objections 5 minor 20 references
C/O ratios in self-gravitating protoplanetary discs with dust evolution
T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read This paper shows that the carbon-to-oxygen ratio of the gas and ices in self-gravitating protoplanetary discs evolves with dust dynamics, producing a narrow C/O fingerprint for gravitational instability and a wide one for streaming…
desk verdict Solid modeling study of C/O in self-gravitating discs, but the GI/SI dichotomy in Figure 10 is softer than the abstract claims because it compares different phase quantities to observations. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The central object is the coupled hydrodynamics–dust–ice model FEOSAD, treated as a thin disc with self-gravity, temperature-dependent viscosity, dust growth and drift, and a phase-transition network for four volatiles. The mechanism carrying the argument is the combination of radial drift of icy grains, which transports oxygen-rich ices inward and concentrates them at snowlines, and the criterion-based identification of GI ($Q\le1$) versus SI regions (the Li–Youdin threshold). The paper then compares the $C/O$ distributions in those regions with observed exoplanet populations and cometary compositions.
What would settle it
Observe the $C/O$ ratios of a dozen directly imaged giant planets at separations beyond 10 au: if their $C/O$ ratios scatter as widely as those of transiting hot Jupiters instead of clustering near 0.3–0.6, the gravitational-instability fingerprint proposed here would be falsified.
Extended reading notes
Core claim
Using the FEOSAD thin-disc hydrodynamics code with self-gravity, two-temperature dust, and a four-species volatile model (H$_2$O, CO$_2$, CO, CH$_4$) that allows freeze-out, thermal desorption, and photodesorption, the authors simulate disc formation and evolution to 0.5 Myr for two core masses. They show that dust drift and the feedback of icy mantles on fragmentation create rings, spirals, and multiple snowlines, and that the total $C/O$ ratio anticorrelates with the dust-to-gas ratio because oxygen-rich ices dominate in dust-rich regions. Weighting all disc regions where the Toomre criterion ($Q \le 1$) is met, they find that the $C/O$ distribution for gravitational instability has a narrow peak near 0.5, while the regions meeting the Li–Youdin criterion for streaming instability give a broad gas-phase $C/O$ range of roughly 0.2–1.4 and low ice-phase values. They argue this dichotomy is consistent with the observed separation between directly imaged exoplanets (narrow, lower $C/O$) and transiting hot Jupiters (wide $C/O$), and that the ice-phase $C/O\approx 0.2$–0.3 between the CO, CO$_2$, and CH$_4$ snowlines matches Solar System comets.
Load-bearing premise
The modelling restricts chemistry to freezing and desorption of four ices; if gas-phase or surface reactions (e.g. CO converting to CO$_2$, methanol forming, volatile trapping in mantles) materially alter the midplane composition, the assigned $C/O$ fingerprints would shift.
Editorial extensions
If this is right
- Planets formed by gravitational instability should have atmospheres with $C/O$ clustered near 0.3–0.6, while planets formed by streaming instability and migration can show $C/O$ anywhere from 0 to 1.4.
- Directly imaged exoplanets likely formed via GI in the outer disc, whereas transiting hot Jupiters' wide $C/O$ spread reflects migration through varied disc conditions.
- Comets with ice-phase $C/O$ 0.2–0.3 formed between the CO and CO$_2$ snowlines, whereas carbon-rich comets likely originated at the snowlines themselves.
- Dust-to-gas ratio and total $C/O$ are anticorrelated, so planetesimals assembled in dust-rich rings should start with low $C/O$ unless they later accrete carbon-rich gas.
Reading between the lines
- If gas-phase and surface reactions (CO-to-CO$_2$ conversion, methanol formation, volatile trapping) were added, the boundaries of the $C/O$ zones would likely shift, but the qualitative anticorrelation and the GI/SI separation may survive; rerunning the same model with a reduced chemical network is a testable extension.
- The narrow GI peak near 0.5 depends on excluding refractory carbon and on the assumed initial volatile abundances; including carbon-rich rock cores would raise the baseline $C/O$ and could move the GI peak closer to the observed directly imaged value.
- This claim implies that atmospheric $C/O$ measurements, combined with orbital separation and age, could identify individual planets formed by disc fragmentation, a prediction testable with current or near-future JWST spectroscopy of wide-orbit planets.
- The ice-phase match to comets suggests that most Solar System comets formed between snowlines, while carbon-rich comets like C/2016 R2 require formation at a carbon-rich snowline; linking specific comets to specific zones is a testable prediction.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper uses the thin-disc hydrodynamic code FEOSAD, including self-gravity, thermal balance, dust growth and drift, and turbulent diffusion, to follow the phase transitions (freeze-out, thermal desorption, photodesorption) of H2O, CO2, CH4, and CO in two self-gravitating protoplanetary disc models over 0.5 Myr. The authors map the resulting C/O ratio in the gas, in the ice, and in total, and connect these distributions to potential planet formation sites. They identify regions where gravitational instability (GI) and streaming instability (SI) criteria are satisfied, and claim that GI regions have a narrow C/O range (≈0.3–0.6, total) matching directly imaged exoplanets, while SI regions have a wide C/O range (≈0–1.4, gas phase) matching transiting hot Jupiters. They also report an anticorrelation between dust-to-gas ratio and total C/O, and an ice-phase C/O of ≈0.2–0.3 between the CO2, CH4, and CO snowlines consistent with Solar System comets.
Significance. If the GI/SI C/O dichotomy survives scrutiny, it would provide a useful observational diagnostic for distinguishing planet formation mechanisms from atmospheric compositions. The paper's strengths include a global 2D model with self-consistent dust dynamics and phase transitions, explicit consideration of disc substructures and multiple snowline geometries, two different core masses, and quantitative comparisons with exoplanet population data. The authors are also transparent about the model's simplifications, such as the absence of gas-phase and surface chemistry and the thin-disc approximation. However, the central claim is currently built on an inconsistent definition of the observable C/O between the two formation channels, and on a simplified chemical scheme whose quantitative ranges are not shown to be robust. These issues need to be addressed before the main diagnostic can be accepted.
major comments (3)
- [Section 4, Figure 10] The comparison between GI and SI regions and observed exoplanet C/O ratios uses different C/O definitions: for GI the total volatile C/O is plotted, while for SI the gas-phase and ice-phase C/O are plotted separately. Observed atmospheric C/O is a single gas-phase quantity, so this is not a like-for-like comparison. A GI planet's atmospheric C/O equals the local total C/O only if all accreted solids are vaporized and mixed into the envelope; otherwise it is closer to the gas-phase value. Similarly, an SI-formed planet's atmospheric C/O depends on its solid-to-gas accretion ratio, which is not specified. The apparent narrow GI peak and broad SI gas-phase tail could therefore arise in part from the phase choice rather than from a genuine formation-mechanism difference. I request that Figure 10 be recomputed using a common observable (e.g., gas-phase C/O for both mechanisms, or a simple envelope-formation model that mixes gas and solids with an explicit accretion ratio), and that the conclusions be re-evaluated accordingly.
- [Section 2.3 and Discussion] The chemical model includes only freeze-out, thermal desorption, and photodesorption of four species, with no gas-phase or surface reactions. The authors correctly note in the Discussion that CO-to-CO2 conversion, methanol formation, and volatile trapping in ice mantles can substantially change C/O ratios. Since the paper makes a quantitative prediction (GI total C/O ≈0.3–0.6, SI gas-phase C/O ≈0–1.4), it is important to quantify how sensitive these ranges are to the missing chemistry. A simple post-processing test, such as converting a fraction of CO ice to CO2 in the relevant regions or adopting a methanol abundance, would show whether the GI/SI dichotomy survives. Without such a test, the claimed C/O fingerprints are statements about the simplified chemistry, not robust predictions of the physical disc.
- [Section 3.5 and Figure 9] The GI C/O distribution in Figure 10 is derived from all regions with Q_Toomre ≤ 1, but the actual GI clumps are under-resolved: the authors state that the clump lifetime is too short for differentiation because of insufficient numerical resolution, and that focused higher-resolution studies are needed. If GI planets form preferentially inside clumps, the material C/O in those clumps may differ from the surrounding Q ≤ 1 regions. The paper should either provide a resolution/convergence test demonstrating that the clump C/O is adequately captured, or justify why the disc-averaged total C/O in GI-unstable regions is representative of the material that ends up in a GI-formed planet.
minor comments (5)
- [Introduction, Section 1] There are several typos and grammatical slips: "the disc matter can be roughly divided into three component" should be "three components"; "annual structures" in Section 3.1 should be "annular structures"; "circumcise" in Section 3.1 should be "circumscribe"; "protorstar" in Section 3.4 should be "protostar"; "he most important" in Section 2.3 should be "the most important"; "drown dust" in Section 4 should be "grown dust"; "boader" in Section 3.3 should be "border"; "V ariations" in Section 3.3 should be "Variations"; "preset" in Section 3.3 should be "present"; "two-dimentional" in Section 3.5 should be "two-dimensional"; and "F ormation" in the References should be "Formation".
- [Figure 3 caption] The caption of Figure 3 reads "Same as Figure 3 but for model M2" but it should refer to Figure 2; please correct the cross-reference.
- [Section 3.3, Figure 4 caption] In the caption of Figure 4, "By the age of 490 yr" in the text should read "490 kyr"; the time unit is inconsistent with the rest of the paper.
- [Section 3.1] The sentence "The most prominent dust rings are located in the vicinity of the water snowline: the ring outside the primary snowline at 5 – 8 au (depending on the time) and the ring at 1 – 2 au, inside the primary snowline, which at later times also contains water ice and additional snowlines" is somewhat confusing because the second ring is first described as inside the primary snowline and then as containing water ice. Please rephrase for clarity.
- [Section 4, Figure 10] The observed exoplanet C/O data from Hoch et al. (2023) are plotted at arbitrary y-axis positions; the figure would be clearer if the data points were shown in a separate panel or with a legend explaining the arbitrary offset.
Circularity Check
No significant circularity: simulated C/O values are emergent model outputs compared with independent exoplanet and cometary datasets.
full rationale
The paper's central C/O predictions are not fitted to the exoplanet data they are compared with. The model starts from protostellar ice abundances (Karin I. Öberg et al. 2011) and evolves gas, dust, and four volatiles through freeze-out, sublimation, advection, and dust dynamics; the resulting gas/ice/total C/O distributions are emergent outputs. The GI and SI regions are identified a priori using dynamical criteria (Toomre Q ≤ 1 and the Li & Youdin 2021 streaming-instability threshold), and the C/O values in those regions are then post-processed, not tuned to match Hoch et al. (2023). The comparison to observed exoplanet C/O uses published, independent data. Self-citations to Molyarova et al. (2021) and Vorobyov et al. (2024) supply the model machinery and the SI post-processing method; they are not uniqueness theorems and do not by themselves force the C/O dichotomy. The one weak citation, 'Molyarova et al., in prep', is used for the snowline accumulation mechanism, but the same mechanism is also attributed to the published Molyarova et al. (2021), and the C/O conclusions do not logically reduce to that citation. The choice to plot total C/O for GI and gas/ice C/O separately for SI is an interpretive assumption about how each formation mechanism incorporates matter; this affects the strength of the comparison but is not a reduction of the output to the input by construction. The dust-to-gas vs total C/O anticorrelation is explicitly acknowledged as a logical consequence of low-C/O ices being attached to dust, not presented as an independent empirical law. No equation reduces to another by definition, and no fitted parameter is renamed as a prediction.
Assumptions & free parameters
free parameters (5)
- Initial volatile ice abundances =
H2O 100, CO2 29, CO 29, CH4 5 relative to H2O; baseline C/O = 0.34
- Turbulent viscosity parameters =
alpha_MRI = 1e-3, alpha_dz = 1e-5 (or 1e-1 above 1300 K), Sigma_MRI = 100 g cm^-2
- Fragmentation velocities =
vfrag = 0.5 m/s for bare grains, 5 m/s for icy grains
- Interstellar UV field strength =
Genv = 5.5 G0
- Initial ice-to-rock mass ratio =
0.08
assumptions (6)
- domain assumption Thin-disc, vertically integrated approximation for gas and dust dynamics
- domain assumption Gas and dust temperatures are equal and set by the midplane temperature
- domain assumption Chemistry is restricted to adsorption and thermal/photodesorption of four volatile species
- domain assumption Initial state has all ices on small grains and no gas-phase volatiles
- domain assumption No late Bondi-Hoyle infall from the ambient medium
- domain assumption GI and SI criteria identify actual planet-forming regions
Cite this review
Pith. "Pith review of C/O ratios in self-gravitating protoplanetary discs with dust evolution." pith.science (2026). https://pith.science/paper/HMCNA27M
@misc{pith2026241205099,
author = {Pith},
title = {Pith review of: C/O ratios in self-gravitating protoplanetary discs with dust evolution},
year = {2026},
howpublished = {\url{https://pith.science/paper/HMCNA27M}},
note = {Machine review of arXiv:2412.05099}
}
abstract
Elemental abundances, particularly the C/O ratio, are seen as a way to connect the composition of planetary atmospheres with planet formation scenario and the disc chemical environment. We model the chemical composition of gas and ices in a self-gravitating disc on timescales of 0.5\,Myr since its formation to study the evolution of C/O ratio due to dust dynamics and growth, and phase transitions of the volatile species. We use the thin-disc hydrodynamic code FEOSAD, which includes disc self-gravity, thermal balance, dust evolution and turbulent diffusion, and treats dust as a dynamically different and evolving component interacting with the gas. It also describes freeze-out, sublimation and advection of four volatile species: H$_2$O, CO$_2$, CH$_4$ and CO. We demonstrate the effect of gas and dust substructures on the distribution of volatiles and C/O ratios, including the formation of multiple snowlines of one species, and point out the anticorrelation between dust-to-gas ratio and total C/O ratio emerging due to the contribution of oxygen-rich ice mantles. We identify time and spatial locations where two distinct trigger mechanisms for planet formation are operating and differentiate them by C/O ratio range: wide range of the C/O ratios of $0-1.4$ for streaming instability, and a much narrower range $0.3-0.6$ for gravitational instability (with the initial value of 0.34). This conclusion is corroborated by observations, showing that transiting exoplanets, which possibly experienced migration through a variety of disc conditions, have significantly larger spread of C/O in comparison with directly imaged exoplanets likely formed in gravitationally unstable outer disk regions. We show that the ice-phase C/O$\approx0.2-0.3$ between the CO, CO$_2$ and CH$_4$ snowlines corresponds to the composition of the Solar system comets, that represent primordial planetesimals.
Figures
Figures from the paper (8 more)
Reference graph
Works this paper leans on
-
[10]
Transport of CO in Protoplanetary Disks: Consequences of Pebble Formation, Settling, and Radial Drift. ApJ 864, no. 1 (September): 78. https : / / doi . org / 10 . 3847 / 1538 - 4357 / aad69b. arXiv: 1808 . 01840 [astro-ph.EP]. Lambrechts, M., and A. Johansen. 2012. Rapid growth of gas-giant cores by pebble accretion.A&A 544 (August): A32. https://doi.org...
arXiv 2012
-
[15]
Revealing dust segregation in protoplanetary discs with the help of multifrequency spectral index maps. MNRAS 486, no. 3 (July): 3907–
-
[45]
arXiv: 2311.12210 [astro-ph.SR]
https://doi.org/10.3847/1538-3881/ad0f 93. arXiv: 2311.12210 [astro-ph.SR]. Schneider, Aaron David, and Bertram Bitsch. 2021. How drifting and evapo- rating pebbles shape giant planets. I. Heavy element content and atmo- spheric C/O. A&A 654 (October): A71. https://doi.org/10.1051/0004- 6361/202039640. arXiv: 2105.13267 [astro-ph.EP]. Seager, S., L. J. Ri...
arXiv 2021
-
[76]
arXiv: 1301.0318 [astro-ph.EP]
https://doi.org/10.1088/0004-637X/769/1/76. arXiv: 1301.0318 [astro-ph.EP]. Banzatti, Andrea, Ilaria Pascucci, Arthur D. Bosman, Paola Pinilla, Colette Salyk, Gregory J. Herczeg, Klaus M. Pontoppidan, et al. 2020. Hints for Icy Pebble Migration Feeding an Oxygen-rich Chemistry in the Inner Planet-forming Region of Disks. ApJ 903, no. 2 (November):
arXiv 2020
-
[95]
Establishing Dust Rings and Forming Planets Within Them
https://doi.org/10.3847/1538- 4357/ac8cf e. arXiv: 2206.01219 [astro-ph.EP]. Lee, Jeong-Eun, Edwin A. Bergin, and Hideko Nomura. 2010. The Solar Nebula on Fire: A Solution to the Carbon Deficit in the Inner Solar System. ApJ 710, no. 1 (February): L21–L25. https://doi.org/10.1088/ 2041-8205/710/1/L21. arXiv: 1001.0818 [astro-ph.GA]. Lee, Jeong-Eun, Seokho...
work page Pith review arXiv 2010
-
[101]
Hydrocarbon emission rings in protoplanetary disks induced by dust evolution
https://doi.org/10.3847/0004-637X/831/1/101. arXiv: 1609.06337 [astro-ph.EP]. Binney, James, and Scott Tremaine. 1987. Galactic dynamics. Birnstiel, Tilman. 2023. Dust growth and evolution in protoplanetary disks. arXiv e-prints (December): arXiv:2312.13287. https://doi.org/10.48550/ arXiv.2312.13287. arXiv: 2312.13287 [astro-ph.EP]. Bisschop, S. E., H. J...
work page Pith review arXiv 1987
-
[124]
https://doi.org/10.3847/1538-4357/abbc1a. arXiv: 2009.13525 [astro-ph.EP]. Benneke, Björn, Heather A. Knutson, Joshua Lothringer, Ian J. M. Crossfield, Julianne I. Moses, Caroline Morley, Laura Kreidberg, et al. 2019. A sub-Neptune exoplanet with a low-metallicity methane-depleted atmo- sphere and Mie-scattering clouds. Nature Astronomy 3 (July): 813–821....
work page Pith review arXiv 2009
-
[130]
arXiv: 1710.02993 [astro-ph.EP]
https://doi.org/10.3847/1538-4357/aa9227. arXiv: 1710.02993 [astro-ph.EP]. Molyarova, Tamara, Eduard I. V orobyov, Vitaly Akimkin, Aleksandr Skliarevskii, Dmitri Wiebe, and Manuel Güdel. 2021. Gravitoviscous Protoplanetary Disks with a Dust Component. V. The Dynamic Model for Freeze-out and Sublimation of V olatiles.ApJ 910, no. 2 (April): 153. https://do...
arXiv 2021
Show all 20 references
-
[684]
Akimkin, Vitaly, Eduard V orobyov, Yaroslav Pavlyuchenkov, and Olga Stoy- anovskaya
https://doi.org/10.1086/177644. Akimkin, Vitaly, Eduard V orobyov, Yaroslav Pavlyuchenkov, and Olga Stoy- anovskaya. 2020. Gravitoviscous protoplanetary discs with a dust compo- nent - IV. Disc outer edges, spectral indices, and opacity gaps. MNRAS 499, no. 4 (December): 5578–...
2020 arXiv
-
[1142]
org / 10
https : / / doi . org / 10 . 1088 / 0004 - 637X / 713 / 2 / 1134. arXiv: 1003.1759 [astro-ph.SR]. Zhu, Zhaohuan, Yan-Fei Jiang, and James M. Stone. 2020. Global 3D radiation magnetohydrodynamic simulations for FU Ori’s accretion disc and observational signatures of magnetic fi...
2020 arXiv
-
[1521]
arXiv: 1610.05139 [astro-ph.GA]
https://doi.org/10.1126/science.aaf 8296. arXiv: 1610.05139 [astro-ph.GA]. Piso, Ana-Maria A., Karin I. Öberg, Tilman Birnstiel, and Ruth A. Murray- Clay. 2015. C/O and Snowline Locations in Protoplanetary Disks: The Effect of Radial Drift and Viscous Gas Accretion. ApJ 815, n...
2015 arXiv
-
[2016]
ApJ 832, no
The Imprint of Exoplanet Formation History on Observable Present-day Spectra of Hot Jupiters. ApJ 832, no. 1 (November): 41. https://doi.org/10.3847/0004- 637X/832/1/41. arXiv: 1609.03019 [astro-ph.EP]. Moses, J. I., N. Madhusudhan, C. Visscher, and R. S. Freedman. 2013. Chem-...
-
[2017]
MNRAS 469, no
Chemical enrichment of giant planets and discs due to pebble drift. MNRAS 469, no. 4 (August): 3994–4011. https://doi.org/10.1093/ mnras/stx1103. arXiv: 1705.03305 [astro-ph.EP]. Bosman, A. D., A. J. Cridland, and Y. Miguel. 2019. Jupiter formed as a pebble pile around the N2 ...
-
[2018]
A&A 611 (April): A80
Efficiency of radial transport of ices in protoplanetary disks probed with infrared observations: the case of CO 2. A&A 611 (April): A80. https://doi.org/10.1051/0004-6361/201732056. arXiv: 1712.03989 [astro-ph.EP]. Brown, Paul D., and S. B. Charnley. 1990. Chemical models of ...
-
[2019]
Nature Astronomy 3 (February): 314–319
The ice composition in the disk around V883 Ori revealed by its stellar outburst. Nature Astronomy 3 (February): 314–319. https://doi. org/10.1038/s41550-018-0680-0. arXiv: 1809.00353 [astro-ph.SR]. Leemker, M., A. S. Booth, E. F. van Dishoeck, L. W ölfer, and B. Dent. 2024. C...
-
[2020]
Nature Astronomy 4 (April): 867–871
The carbon monoxide-rich interstellar comet 2I/Borisov. Nature Astronomy 4 (April): 867–871. https://doi.org/10.1038/s41550-020- 1095-2. arXiv: 2004.08972 [astro-ph.EP]. Booth, R. A., and J. D. Ilee. 2019. Planet-forming material in a protoplanetary disc: the interplay between...
2004 arXiv
-
[2022]
PSJ 3, no
A Survey of CO, CO2, and H2O in Comets and Centaurs. PSJ 3, no. 11 (November): 247. https://doi.org/10.3847/PSJ/ac960d. arXiv: 2209.09985 [astro-ph.EP]. Hartmann, L., and S. J. Kenyon. 1985. On the nature of FU Orionis objects. ApJ 299 (December): 462–478. https://doi.org/10.1...
1985 arXiv
-
[2024]
3 (May): 2731–2748
Ices on pebbles in protoplanetary discs.MNRAS 530, no. 3 (May): 2731–2748. https://doi.org/10.1093/mnras/stae597. arXiv: 2403.02895 [astro-ph.EP]. Turrini, D., E. Schisano, S. Fonte, S. Molinari, R. Politi, D. Fedele, O. Panić, M. Kama, Q. Changeat, and G. Tinetti. 2021. Traci...
-
[3900]
arXiv: 2207.13002 [astro-ph.EP]
https://doi.org/10.1093/mnras/stac3275. arXiv: 2207.13002 [astro-ph.EP]. Kadam, Kundan, Eduard V orobyov, and Shantanu Basu. 2022. Primordial dusty rings and episodic outbursts in protoplanetary discs.MNRAS 516, no. 3 (November): 4448–4468. https://doi.org/10.1093/mnras/stac24...
2022 arXiv
-
[3914]
arXiv: 1904.05251 [astro-ph.IM]
https://doi.org/10.1093/mnras/stz1046. arXiv: 1904.05251 [astro-ph.IM]. Pelkonen, Veli-Matti, Paolo Padoan, Mika Juvela, Troels Haugbølle, and Åke Nordlund. 2024. Origin and Evolution of Angular Momentum of Class II Disks. arXiv e-prints (May): arXiv:2405.06520. https://doi.or...
1904 arXiv
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.