{"id":"2fdb798b-c55d-4e36-99b8-afe575441f4e","arxiv_id":"2412.09930","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Adding a hot atmosphere to protoplanetary disk models strengthens vertical shear instability turbulence by more than an order of magnitude and makes its signatures visible in synthetic observations.","lead":"Gas disks around young stars may be churned by the vertical shear instability, a type of swirling motion. New simulations show that a hotter layer above the disk makes this turbulence about ten times stronger, creating velocity patterns that telescopes like ALMA could detect.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quantitative α_Rφ≳10^-3 result assumes instantaneous cooling (locally isothermal EOS); finite cooling in real disks—acknowledged in §4.4—likely lowers it, so the abstract's number is an upper limit rather than a robust prediction.","rationale":"The paper's strongest result is a quantitative one: thermally stratified disks saturate at α_Rφ ∼ 10^-3, more than an order of magnitude above the isothermal case, and produce observable ring segments. The numerical implementation is careful: resolution and boundary-condition tests (Appendix A) support convergence of the saturated stress, and the synthetic-observation pipeline follows established tools. The weakest link is the physical assumption behind the thermodynamics: a locally isothermal EOS fixes the temperature and removes all cooling physics. This is known to be the most favorable limit for VSI growth. The paper acknowledges this in Section 4.4, but the abstract's headline number is not similarly qualified. If finite cooling is included, the volume-averaged α_Rφ is likely to fall, because the midplane becomes stable; however, the observable signatures could survive because they trace the surface layers where cooling is rapid. This is precisely the uncertainty that determines whether the paper's central claim is a real disk prediction or an idealized upper bound. Since the paper itself flags the caveat and the reader made it the basis of a conditional verdict, I see no reason to move the verdict; the condition should be stated explicitly in the abstract.","tokens_in":19056,"tokens_out":9273,"duration_ms":110381,"concrete_test":"Re-run the n=2 and n=3 models with a vertically varying cooling time, t_cool(R,Z) = β(Z) Ω_K^{-1}, with β ≲ 0.1 for |Z| ≳ 2H and β ≳ 10 near the midplane, preserving the same initial temperature profile; measure the saturated volume-averaged α_Rφ from Eq. (6) and recompute the synthetic 12CO residual maps. If α_Rφ drops below ~10^-4 while ring segments survive, the alpha claim in the abstract should be presented only as an upper limit, while the observability claim can stand; if the segments vanish, both quantitative and observability claims need to be downgraded.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim—saturated α_Rφ ≳ 10^-3 in thermally stratified disks, an order of magnitude above the isothermal case—depends on the locally isothermal equation of state, which is equivalent to an infinite cooling rate. In real protoplanetary disks the cooling time is finite and height-dependent: the optically thick midplane cools slowly, and previous radiation-hydrodynamic and two-temperature simulations (Fukuhara et al. 2023; Pfeil et al. 2023; Melon Fuksman et al. 2024; Zhang et al. 2024) find VSI damped where t_cool exceeds the critical value. Since α_Rφ in Eq. (6) is a volume average over the whole disk, a quiescent midplane would lower it substantially, potentially eroding the 'more than an order of magnitude stronger' comparison. The observability part is more robust because the synthetic residuals are dominated by the surface layers (Sections 3.2, 4.2), where cooling is fast and VSI can persist. Thus the weakest point is not internal inconsistency but the extrapolation from an idealized cooling-free model to real disks; the paper itself flags this in Section 4.4, but the abstract presents the α value without that qualification.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents three-dimensional hydrodynamic simulations of the vertical shear instability (VSI) in protoplanetary disks, extending Barraza-Alfaro et al. (2021) to vertically stratified temperature profiles and following the authors' linear analysis (Paper I). Three locally isothermal, inviscid models with thermal stratification ratio n = Tatm/Tmid = 1, 2, 3 (n = 1 being the isothermal baseline) are evolved with FARGO3D. The authors report that stratification enhances VSI growth and saturation: the volume-averaged Reynolds stress reaches alpha_Rphi ~ 1 x 10^-4 for n = 1 versus ~ 1 x 10^-3 and ~ 2 x 10^-3 for n = 2 and 3, with the saturated turbulence concentrated near the disk surfaces. Synthetic ALMA observations, produced with RADMC3D and the standard bettermoments/eddy/syndisk pipeline, show that after subtracting the fitted Keplerian model, VSI-induced velocity perturbations appear as quasi-axisymmetric rings in isothermal disks and as ring segments in stratified disks. Residual amplitudes are roughly 50-100 m/s at 20 degrees inclination, the ring segments remain visible up to 45 degrees in the n = 3 model, and the residual amplitude increases with tracer optical depth (12CO stronger than 13CO and C18O). The authors argue that these signatures may explain kinematic features such as the rings observed in MWC 480.","tokens_in":19303,"tokens_out":17405,"duration_ms":171155,"significance":"If the results hold, the paper makes two contributions. First, it establishes a quantitative, physically grounded link between vertical thermal stratification and VSI vigor in the idealized locally isothermal limit, with the n-dependence of the growth rates and wavelengths following the linear theory of Paper I. Second, and more robustly, it produces falsifiable kinematical predictions for ALMA: ring segments rather than full rings, visibility at inclinations up to 45 degrees, and a tracer-opacity dependence that a detection experiment can test. The computational work is careful and reproducible: public codes (FARGO3D, RADMC3D, syndisk, bettermoments, eddy) are used throughout; Appendix A presents resolution (15 vs 30 cells per scale height) and meridional boundary-condition convergence tests, showing the saturated alpha_Rphi to be insensitive to both; and the nonlinear state is emergent rather than fitted, so there is no circularity in the central results. The observability conclusions are the more durable part of the paper because the residuals are dominated by the surface layers, where finite cooling times are least damaging.","major_comments":[{"comment":"The headline quantitative claim, that \"the turbulence stress reaches alpha_Rphi >= 10^-3, more than an order of magnitude stronger than the isothermal case,\" appears in the Abstract without qualification, and Section 5 item 1 repeats the values alpha_Rphi = 1 x 10^-4, 1 x 10^-3, and 2 x 10^-3 as results. The simulations, however, use a locally isothermal (instantaneous-cooling), inviscid equation of state, and Section 4.4 itself states that the VSI amplitudes are \"likely to be upper limits compared to the ones in real disks.\" Because alpha_Rphi in Eq. (6) is a volume average that includes the midplane, and the cited radiation-hydrodynamic and two-temperature simulations (Fukuhara et al. 2023; Pfeil et al. 2023; Melon Fuksman et al. 2024; Zhang et al. 2024) find the VSI damped wherever the cooling time exceeds the critical value, the abstract's number will reasonably be read as a prediction for real disks unless the upper-limit status is stated there. Please move the qualification into the Abstract and Section 5, and preferably add a short quantitative estimate, for instance the mass or volume fraction of the computed disk in which the cooling time exceeds the local critical value using a cooling-time prescription from the cited literature, to indicate the expected reduction in the volume-averaged alpha_Rphi. The relative comparison between n = 1, 2, and 3 within the model family is unaffected and needs no change.","section":"Abstract; Section 4.4; Section 5 item 1"},{"comment":"The observability conclusions, that VSI signatures \"can potentially be observable by ALMA\" and \"remain visible at disk inclinations as high as 45 degrees,\" rest on visual inspection of single-epoch synthetic maps with no quantitative detection criterion. The residual maps are produced by subtracting a 13-parameter fitted Keplerian model, and the turbulence itself is time-dependent: the stress profiles in Fig. 7 are averaged over t = 150-300 torb, while the images use only the epoch t = 300 torb. A matched-filter or SNR measurement of the ring-segment pattern in the noise-added cubes, ideally evaluated against a quiescent control disk and across several epochs in the saturated state, is needed to support the detectability claim. This is especially important at i = 45 degrees (Fig. 10) because the authors attribute the spurious radial features at i = 35 degrees (Fig. 9) to emission-surface fitting mismatches, an effect that should worsen with inclination.","section":"Section 3.2; Fig. 10; Section 5 item 2"}],"minor_comments":[{"comment":"The caption labels the models \"Tatm/Tmid = 0, 1, 2\"; this should read n = 1, 2, 3, since n = 0 is never defined in the text.","section":"Fig. 1 caption"},{"comment":"The phrase \"only vr, vphi, or vphi\" duplicates vphi; the middle panel shows vtheta, so this should read \"only vr, vtheta, or vphi.\"","section":"Fig. 11 caption and Section 4.1"},{"comment":"\"Synthethic Observations\" is a misspelling of \"Synthetic Observations.\"","section":"Section 3.2 heading"},{"comment":"The isotope ratio is written as \"[16C]/[18O] ~ 560\"; 16C is not the relevant species, and this should be [16O]/[18O] to match Wilson & Rood (1994).","section":"Section 2"},{"comment":"\"Kelvin-Helomholtz-like parasitic instability\" should be \"Kelvin-Helmholtz-like,\" and \"Rossyby instability\" should be \"Rossby instability.\"","section":"Section 3.2; Section 4.3"},{"comment":"The phrase \"the models with n = 1, 2m and 3\" contains a typo (\"2m\") and should read \"n = 1, 2, and 3.\"","section":"Appendix A"},{"comment":"The sentence \"we expect that the amplitudes ... is likely to be upper limits\" has a subject-verb agreement error and should read \"the amplitudes ... are likely to be upper limits.\"","section":"Section 4.4"},{"comment":"The claim that the ring segments \"are visible at disk inclinations as high as 45 degrees\" is demonstrated only for the n = 3 model; please qualify the statement as applying to n = 3 or show the i = 45 degrees case for n = 2 as well.","section":"Abstract; Fig. 10"},{"comment":"The resolution study spans only a factor of two (15 vs 30 cells per scale height), whereas Flores-Rivera et al. (2020) recommend at least 64 cells per scale height; the presented test does support convergence of the saturated alpha_Rphi, but the authors should state explicitly which diagnostics they regard as converged and why the factor-of-two test suffices for the observability claims, which depend on the resolved mode structure.","section":"Appendix A.1"},{"comment":"The Urpin & Brandenburg (1998) entry contains a duplicated DOI: \"10.1111/j.1365-8711.1998.01118.x10.1111/j.1365-8711.1998.01118.x.\"","section":"References"},{"comment":"The \"time-averaged velocity perturbations\" quoted for the midplane are not tied to a stated averaging window; please specify the interval (presumably t = 150-300 torb, as in Fig. 7).","section":"Section 3.1"}],"recommendation":"major_revision","confidential_remarks":"To the editor: this is a careful, internally consistent numerical study, and I consider the requested changes achievable without new production runs; the one optional item (a quantitative cooling-time estimate) may require a modest literature-based calculation. I could not verify Paper I (Yun et al. 2024, \"submitted\"), which supplies the disk models, the shear parameter q, and the mode terminology; the authors should be encouraged to cite the arXiv version of Paper I if one is available. The use of public codes and the convergence tests are strengths, and the comparison to Barraza-Alfaro et al. (2021) is fair. The main issue is that the abstract's headline number is presented as a result rather than as an upper limit in an idealized model; this is a framing problem, not a defect in the simulations themselves."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a careful piece of work. The new result is that vertical thermal stratification strengthens saturated VSI turbulence by an order of magnitude or more (alpha_Rphi ~1e-3 to 2e-3 vs ~1e-4 isothermal) and changes the expected ALMA morphology from axisymmetric rings to ring segments, with residuals of 50-100 m/s visible up to i=45 deg. That is a genuinely new extension of Barraza-Alfaro et al. (2021) and of their own Paper I linear analysis. The comparison to MWC 480 is suggestive rather than proof, but it is a reasonable hook.\n\nWhat the paper does well: the numerics are carefully specified and the appendix tests resolution (15 vs 30 cells/H) and meridional boundary conditions, showing the saturated stress is insensitive. The synthetic observations are thorough: three CO isotopologues, three inclinations, a proper Keplerian fit with MCMC over emission surface parameters. And they are transparent about the main caveat: Section 4.4 states clearly that viscosity, finite cooling time, buoyancy, and magnetic fields can weaken the VSI, and that the quoted amplitudes are likely upper limits. That honesty is real.\n\nNow the soft spots. The biggest one is exactly what the authors admit: the locally isothermal equation of state is equivalent to infinite cooling rate. Recent radiation-hydrodynamic and two-temperature simulations (Fukuhara et al. 2023, Pfeil et al. 2023, Melon Fuksman et al. 2024, Zhang et al. 2024) find VSI damped where t_cool exceeds the critical value, i.e. in the optically thick midplane. Since alpha_Rphi is a volume average, the real number could be substantially lower than 1e-3. The stress-test is right that the observability part is more robust, because the synthetic residuals are dominated by surface layers where cooling is short. Still, I would like the abstract not to present alpha_Rphi >= 1e-3 without the 'upper limit' qualifier that appears only in Section 4.4.\n\nSecond, the resolution test goes to 30 cells/H, but Flores-Rivera et al. (2020) argued 64 cells/H is needed to capture the VSI properly. The saturated alpha looks converged, but the transient evolution is resolution-dependent in the isothermal case. It would be stronger if they pushed to higher resolution in at least the n=3 model.\n\nThird, minor: no code or data release, which makes reproduction harder, and the synthetic pipeline assumes the inclination and distance are known exactly. Neither is a fatal issue.\n\nWho should read this: anyone working on VSI, disk turbulence, dust evolution, or ALMA disk kinematics. It deserves a serious referee; the central qualitative conclusion is solid and the quantitative limitation is at least flagged. My recommendation: send it to review, and ask the authors to modulate the abstract and to add a sentence acknowledging the 64 cells/H concern. That would make the paper fully honest.","headline":"Solid, well-executed simulation study: thermal stratification boosts VSI turbulence an order of magnitude and changes its ALMA morphology, though the absolute alpha values are upper limits under instantaneous cooling.","tokens_in":19844,"tokens_out":2639,"would_cite":true,"duration_ms":27151,"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":"Thermally stratified protoplanetary disks drive vertical shear instability turbulence more than an order of magnitude stronger than isothermal disks, and the resulting ring-segment velocity residuals should be detectable at high…","keywords":["protoplanetary disks","vertical shear instability","hydrodynamic simulations","thermal stratification","turbulence","velocity residuals","synthetic observations","angular momentum transport"],"falsifier":"A radiation-hydrodynamic rerun of the $n=2$ and $n=3$ models with finite cooling times that saturates $\\alpha_{R\\phi}$ below about $10^{-4}$, or an interferometric observation of a strongly stratified disk at $i=20°$ in 12CO that shows no 50–100 m/s ring segments, would falsify the central claim.","tokens_in":18865,"feed_emoji":"🌀","tokens_out":13911,"duration_ms":125704,"temperature":0.7,"pith_summary":"Thermally stratified protoplanetary disks — with a hotter atmosphere above a cooler midplane — drive the vertical shear instability (VSI), a purely hydrodynamic source of turbulence, much harder than vertically isothermal disks do. In three-dimensional simulations, the VSI begins growing within a few orbits and saturates with a Reynolds stress-to-pressure ratio $\\alpha_{R\\phi} \\approx 1\\times10^{-3}$ (for an atmosphere twice as hot as the midplane) and $\\approx 2\\times10^{-3}$ (three times as hot), compared with $\\approx 1\\times10^{-4}$ in the isothermal case. The saturated turbulence is strongest near the disk surfaces, and after subtracting Keplerian rotation it appears in synthetic millimeter images as ring segments rather than full rings, with velocity residuals of roughly 50–100 m/s at 20° inclination and visibility up to 45°. The authors argue that this makes VSI turbulence detectable in real disks and that the height-dependent, longer-wavelength patterns observed in some disks are naturally explained by thermal stratification. They caution that the models assume instantaneous cooling, so these amplitudes are likely upper limits.","feed_headline":"Stratified disks boost VSI turbulence tenfold","feed_subtitle":"Hot-atmosphere disks should reveal 50–100 m/s ring segments in gas motion, even at 45° inclination.","key_machinery":"The load-bearing object is the vertical shear parameter $q = -R \\, \\partial\\ln\\Omega/\\partial Z$, the fractional change of orbital angular velocity with height. Realistic disks with a hotter atmosphere have a non-monotonic $q$ profile that peaks near $|Z| \\approx Z_q/2$ (about 1.5 scale heights), stronger than the linearly growing $q$ of an isothermal disk; the simulations compare $n=1,2,3$, where $n = T_{\\rm atm}/T_{\\rm mid}$. This enhanced shear shortens the growth time, lengthens the preferred radial wavelength (the measured $\\lambda_R/H$ grows from about 3.3 to 10.7 with $n$), and drives the two-stage growth of surface modes followed by body modes. Saturation is measured by $\\alpha_{R\\phi}$, the volume-averaged Reynolds stress to thermal pressure, and the observability claim rests on synthetic velocity residual maps $v_0 - v_{\\rm mod}$ obtained by fitting and subtracting a Keplerian disk model from the line-of-sight velocity centroids.","core_discovery":"The central claim is that vertical thermal stratification does not merely modify the vertical shear instability — it strengthens it. In the stratified models ($n \\equiv T_{\\rm atm}/T_{\\rm mid} = 2$ and $3$), the instability grows earlier and faster than in the isothermal $n=1$ model, reaching a saturated turbulence level $\\alpha_{R\\phi}$ of about $1\\times10^{-3}$ and $2\\times10^{-3}$, respectively, more than an order of magnitude above the isothermal value of $1\\times10^{-4}$. The perturbation velocities in the radial and azimuthal directions grow relative to the vertical ones as stratification increases, so the turbulence is less purely meridional. When the saturated velocity fields are put through synthetic line radiative transfer and a Keplerian model is subtracted, the residual maps show quasi-axisymmetric rings in the isothermal disk but fragmented ring segments in stratified disks; these segments survive up to an inclination of 45° and grow stronger when the tracer line is optically thick (12CO) because it probes the surface layers where the VSI is most active.","pith_inferences":["If surface layers of real disks are this turbulent, dust settling and grain-growth timescales in the upper layers could differ from midplane values, changing predictions for where planetesimals form.","The fitted wavelength scaling ($\\lambda_R/H$ increasing with $T_{\\rm atm}/T_{\\rm mid}$) could be turned around observationally: measuring the radial spacing of ring segments in residual maps would constrain the disk's vertical temperature stratification.","Comparing 12CO and C18O residual maps of the same disk could serve as a direct probe of the vertical shear profile, since the ratio of the two signals encodes where the instability operates.","A radiation-hydrodynamic version of these runs would show whether surface-dominated turbulence survives finite cooling; if it does, the observable ring segments might persist even when midplane VSI is suppressed."],"forward_implications":["In a thermally stratified disk, VSI-driven angular momentum transport reaches $\\alpha_{R\\phi} \\gtrsim 10^{-3}$, enough to make the instability a significant driver of radial gas accretion and mixing, not just a source of small velocity noise.","VSI kinematics are height-dependent: optically thick tracers like 12CO show stronger, more readily detected residuals than optically thin tracers like C18O that probe the midplane.","The ring-segment morphology of stratified VSI turbulence remains visible at inclinations up to 45°, whereas isothermal rings become hard to detect at high inclination because they are dominated by vertical motion.","The longer radial wavelength and height dependence of stratified VSI can reconcile model kinematics with ring-like velocity perturbations observed in real disks that isothermal models could not explain.","Because the simulations use instantaneous cooling, the reported $\\alpha$ and velocity amplitudes should be read as upper limits; finite cooling in optically thick regions is expected to weaken them."],"supporting_citations":[{"why":"Supplies the linear-theory predictions of VSI in stratified disks — surface versus body modes, enhanced growth rates, and wavelength scaling — that the simulations confirm and extend.","marker":"Paper I"},{"why":"Provides the vertically isothermal VSI simulations and the velocity-residual-map methodology that this paper extends to stratified disks.","marker":"Barraza-Alfaro et al. (2021)"},{"why":"Identifies the surface and body VSI modes and establishes the basic disk model for the instability.","marker":"Nelson et al. (2013)"},{"why":"Sets the finite-cooling-time criterion used to bound when VSI can grow, backing the paper's upper-limit caveat.","marker":"Lin & Youdin (2015)"},{"why":"Shows vertically varying cooling suppresses VSI in optically thick regions, supporting the expectation that real amplitudes are lower.","marker":"Fukuhara et al. (2023)"},{"why":"Reports observed ring-like velocity perturbations whose height dependence and length scale motivate the stratified VSI interpretation.","marker":"Teague et al. (2021)"},{"why":"Provides the emission-surface parameterisation used to fit the Keplerian model and isolate the velocity residuals.","marker":"Law et al. (2021)"}],"fun_headline_variants":["Stratified disks amplify VSI turbulence tenfold","Hot atmospheres strengthen disk turbulence and rings","Thermal Stratification Boosts VSI Turbulence and Visibility","VSI turbulence up 10x in stratified protoplanetary disks","Stratification enhances VSI, revealing ring segments at 45°"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The models treat the gas as locally isothermal with instantaneous cooling, the ideal limit for VSI growth; in real disks, finite cooling in optically thick regions damps the instability, so the simulated turbulence amplitudes are probably upper limits, as the authors state.","fun_headline_variants_meta":{"raw":{"variants":["Stratified disks amplify VSI turbulence tenfold","Hot atmospheres strengthen disk turbulence and rings","Thermal Stratification Boosts VSI Turbulence and Visibility","VSI turbulence up 10x in stratified protoplanetary disks","Stratification enhances VSI, revealing ring segments at 45°"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000648,"raw_usage":{"total_tokens":3008,"prompt_tokens":1013,"completion_tokens":1995,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":629,"completion_tokens_details":{"reasoning_tokens":1912}},"tokens_in":629,"tokens_out":1995,"duration_ms":14706,"temperature":1.0,"reasoning_tokens":1912,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T16:33:07.967547+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A radiation-hydrodynamic rerun of the $n=2$ and $n=3$ models with finite cooling times that saturates $\\alpha_{R\\phi}$ below about $10^{-4}$, or an interferometric observation of a strongly stratified disk at $i=20°$ in 12CO that shows no 50–100 m/s ring segments, would falsify the central claim.","supporting_citations":[],"review_version":1}