{"id":"a9c522fb-d720-4a30-8122-6c845ec03bee","arxiv_id":"2502.00161","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A thin-disk protoplanetary evolution code is extended with a calibrated magnetic wind prescription, producing smaller, less massive disks whose synthetic ALMA sizes and masses move toward observed Class II disk values.","lead":"This paper adds magnetic disk winds to a computer model of planet-forming disks and compares the simulated disks with ALMA telescope surveys. It matters because winds may explain why observed disks are smaller and lose mass faster than the usual viscous evolution picture predicts.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed agreement with ALMA demographics hinges on the constant wind-strength calibrator Cβ=1000; a factor-of-few change moves the model from 'unreasonably small disks' to 'favorable comparison', so the comparison is not robust to the paper's own stated uncertainty.","rationale":"The paper is a model-development study with an unusually honest discussion of its own limitations; the qualitative result that winds produce smaller, less massive disks and the inferred lever arm λ≈1–2 are genuine and physically plausible. My concern is not that the wind model is wrong, but that the specific quantitative claim—'synthetic observations compare favorably with ALMA survey data'—is anchored to a constant Cβ=1000 that is unmeasured and admitted to be possibly non-constant. The internal comparison of WI-3 (Cβ=1000) and WI-3a (Cβ=5000) shows that the observable diagnostics swing by more than an order of magnitude in disk size and mass within the permitted uncertainty. Thus, without a sensitivity analysis or an independent determination of Cβ, the favorable comparison is an existence proof ('some Cβ works') rather than a validation of the model. This is exactly the condition the reader identified, and it justifies a CONDITIONAL verdict: the framework is acceptable, but the central claim should be presented with robustness bounds or an independent calibration. The age mismatch (models run to 1 Myr versus observed regions up to 3 Myr) is also a limitation, but the paper acknowledges its direction, and the Cβ sensitivity is the more decisive issue for the headline claim.","tokens_in":42385,"tokens_out":7420,"duration_ms":75818,"concrete_test":"Fix CH=1.0 and rerun WI-3a with Cβ=2000 and Cβ=8000, keeping all other inputs and the ProDiMo pipeline fixed. Compute the synthetic Band 6 R_dust, 12CO R_gas, and 13CO flux at 0.5 and 1 Myr, and compare against the 10th–90th percentile ranges of the Lupus/Chamaeleon samples used in Figs. 4–5. If either Cβ variant leaves these observed ranges or shifts R_dust by more than roughly a factor of 2, the central claim is not robust to the stated Cβ uncertainty. Alternatively, set Cβ from independent constraints—for example, non-ideal MHD vertical-field profiles or meteoritic paleomagnetic B_z limits—and repeat the comparison without refitting; if the resulting disks fall outside the observed distributions, the favorable comparison is calibration-dependent.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central validation is the claim that wind models 'compare favorably' with ALMA Class II surveys. This claim is carried by the two free parameters Cβ and CH in Eqs. (25)–(26). Because the mass-loss and torque scalings are steep (∝(Cβ β0)^-0.46 and ∝(Cβ β0)^-0.66), a constant Cβ=1000 is not a minor rescaling: it changes integrated wind loss and torque by roughly 25–100× relative to Cβ=1. The model's own parameter study demonstrates the consequence: with Cβ=1000, CH=0.5 and αMRI=10^-3, the Band 6 dust radius collapses to ~1.4 au (WI-3, Table 4), far outside observed ranges; with Cβ=5000, CH=1.0 (WI-3a), the same diagnostics enter the observed ranges (R_dust,1.3mm=51 au, R_gas,12CO=268 au). The difference between 'unreasonably small' and 'favorable' is therefore a factor of 5 in an unmeasured parameter. Section 4 explicitly concedes Cβ 'may not be constant in space and time' and that β0 estimates 'vary by orders of magnitude.' Nothing in the paper independently fixes Cβ: it is chosen so that Cβ β0 ~ 10^4, but the allowed range 10^2–10^8 spans the entire behavior of the model. The favorable comparison is thus a postdiction of the calibration, not an independent test. Independent support exists (the inferred lever arm λ≈1–2 in Sect. 3.4 is consistent with MHD simulations and observations), but that validates the torque-to-mass-loss ratio, not the absolute wind strength that sets disk sizes and masses.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a phenomenological model of magnetic disk winds for global protoplanetary disk evolution in the thin-disk limit, based on the shearing-box fitting formulae of Bai (2013). The wind prescription is implemented as mass and angular momentum sink terms in the FEOSAD code, with corrections for stellar mass, FUV luminosity, shearing-box height, and a constant factor C_beta intended to compensate for the overestimated vertical field in the ideal-MHD calculation. Six models are evolved from collapsing cores through Class II, and synthetic Band 6 continuum and CO line observations are generated with ProDiMo. The authors find that wind-driven models produce smaller and less massive disks than purely gravitoviscous models and argue that, after adjustment of C_beta and C_H, the synthetic sizes, masses, fluxes, and spectral indices compare favorably with ALMA survey demographics.","tokens_in":20,"tokens_out":6866,"duration_ms":131775,"significance":"If the calibration can be made robust, the paper would be a valuable step: it couples wind-driven angular momentum loss with self-gravity, MRI-dead-zone physics, and two-population dust evolution, and it provides a concrete pipeline from global hydrodynamic simulations to ALMA-observable quantities. The inferred steady lever arm lambda ~ 1-3 in Sect. 3.4 is a genuine independent consistency check, and the qualitative trends (smaller, lower-mass disks with winds; strong sensitivity of the outcome to the viscous alpha) are useful predictions. The main limitation is that the headline \"favorable comparison\" currently rests on two calibration parameters, C_beta and C_H, whose plausible ranges span the entire range of model behavior; the comparison is therefore a postdiction until robustness and independent constraints are demonstrated.","major_comments":[{"comment":"The central comparison with ALMA demographics is a postdiction rather than an independent test. C_beta and C_H are set in Section 2.2 by requiring the integrated wind mass-loss rate to be of the order of the accretion rate, a requirement motivated by observed outflow mass-flux estimates; the same broad set of observational properties is then used to claim that the synthetic disks 'compare favorably' with ALMA size and mass surveys. Because the wind mass loss and torque scale as power laws of C_beta*beta0 (with exponents -0.46 and -0.66 in Eqs. (25)-(26)), the normalization essentially determines the disk sizes and masses that are later compared with the data. The paper should either present the comparison explicitly as a calibration fit with the number of degrees of freedom stated, or validate the model against an observable not used in the calibration (for example, radial accretion-rate profiles, outflow kinematics, or a different star-forming region not included in Table B.1).","section":"Section 3.3 and Eqs. (25)-(26)"},{"comment":"The result is extremely sensitive to C_beta, and the manuscript does not quantify this sensitivity. With C_beta=1000 and C_H=0.5, model WI-3 gives R_dust,1.3mm = 1.4 au and R_gas,12CO = 114 au, which the authors themselves call unreasonably small; with C_beta=5000 and C_H=1.0, model WI-3a gives R_dust,1.3mm = 51 au and R_gas,12CO = 268 au and is claimed to be favorable. A factor of five in an unmeasured parameter therefore moves the model from outside all observed ranges to inside them. Section 4 concedes that C_beta 'may not be constant in space and time' and that beta0 estimates 'vary by orders of magnitude'; those statements, combined with the steep power-law exponents, mean the model cannot yet be considered validated. Please add a robustness study varying C_beta (and C_H) over the observationally motivated range, and provide an independent physical argument for why a single constant factor suffices to correct the ideal-MHD B_z.","section":"Table 4, Sect. 4, Eq. (17)"},{"comment":"The 'favorable comparison' is selective: model WI-3a, the only low-alpha wind model that reproduces the observed sizes, still does not fit all observables. The synthetic 13CO and C18O line fluxes in the last two panels of Fig. 5 remain systematically above the observed distributions, and in Fig. 6 the WI-3a track lies beyond the R_g = 4 R_d line, as the text acknowledges ('the ratio of gas to dust radius is somewhat larger than Rg = 4Rd line'). The conclusion should be reworded to state which diagnostics are reproduced and which are not, and a quantitative goodness-of-fit metric (e.g., fraction of survey objects within the model tracks, or chi-square over the diagnostics) should be provided instead of the qualitative phrase 'compare favorably'.","section":"Section 3.3, Figs. 5 and 6"}],"minor_comments":[{"comment":"The sentence introducing the parameters says 'the only free parameters are C_H and C_beta,' but several additional choices are unconstrained and not listed as parameters: the FUV exponents in Eqs. (19)-(20), the 0.1 g cm^-2 surface density threshold for wind activation, the 1.2 factor in the centrifugal criterion, and the 5% per-cell mass-loss cap. Please list these in a parameter table and comment briefly on their influence.","section":"Section 2.2"},{"comment":"The equation for the lever arm uses Sigma_w,tr before the variable is defined; define Sigma_w,tr as the wind-driven radial mass transport rate in the text preceding Eq. (31).","section":"Section 3.4, Eq. (31)"},{"comment":"The phrase 'the last two panels of Fig. 5 show the 13CO and C18O line fluxes at 150 au' is ambiguous: the fluxes are computed for a disk at a distance of 150 pc, not at a radius of 150 au. Please clarify the wording.","section":"Section 3.3"},{"comment":"In the Fig. 4 caption, the half-violin plots are described as 'gas (red) and dust (red) radii'; one of the two should presumably be a different color (e.g., blue) to match the panels, and the same inconsistency appears in the Fig. 5 caption.","section":"Figure captions"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a model-development paper with a companion paper announced; the present version would be strengthened by an explicit statement of what is newly validated relative to the companion, and by a robustness/calibration section. I would not require the authors to scan C_beta exhaustively, but the current 'favorable comparison' language should be matched to the demonstrated diagnostics. The absence of a data/code availability statement for the FEOSAD-ProDiMo interface may also be worth raising with the authors, as the interface is central to the paper's reproducibility."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a useful model-building paper, and the authors are honest about its limits, but the headline claim of favorable agreement with ALMA is a postdiction of two free parameters, not an independent test.\n\nWhat is genuinely new: the assembled wind prescription in Eqs. (25)–(26), with stellar-mass, FUV, and plasma-beta corrections to the Bai (2013) fitting formulas, implemented self-consistently in FEOSAD with dust co-evolution and ProDiMo synthetic observations. That is real infrastructure. The inferred lever arm λ ≈ 1–2 is a legitimate independent check, because it follows from the torque-to-mass-loss ratio rather than from the absolute wind strength, and it matches MHD simulations and outflow observations.\n\nThe soft spot is where the reader puts it. C_beta and C_H are constrained so that the integrated wind mass loss is of order the accretion rate, and the same observational datasets (ALMA sizes and masses) are then used as the target. The paper even says C_beta may not be constant in space or time and that beta estimates vary by orders of magnitude. The parameter sensitivity is steep: changing C_beta from 1000 to 5000 and C_H from 0.5 to 1.0 takes the WI-3 dust radius from 1.4 au to 51 au. A factor five in an unmeasured parameter separates 'unreasonably small' from 'favorable.' That is not a robust validation. Also, the adjusted model WI-3a still does not fit the CO fluxes; the paper acknowledges this.\n\nCredit where due: the presentation is explicit, the equations are complete, the parameter study is small but sufficient to expose the issue, and the lever-arm check is a step beyond pure curve-fitting. The qualitative shrinking of disks under winds is plausible and consistent with prior advective disk models.\n\nThis paper deserves serious refereeing. I would send it to review with a request to reframe the comparison as a calibration exercise, not a validation, and to push for a testable prediction — for example, from non-ideal MHD or by withholding part of the sample during calibration.\n\nWho is this for: people building long-term disk evolution models with winds, and observers interpreting disk size demographics. It is a solid framework paper, but treat the ALMA agreement as a demonstration of tunability, not evidence.","headline":"Useful model-building paper with honest caveats, but the ALMA agreement is a postdiction of two free parameters, not an independent test.","tokens_in":43341,"tokens_out":2566,"would_cite":true,"duration_ms":26336,"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":"Magnetic disk winds, added to a gravitoviscous thin-disk model, make synthetic Class II protoplanetary disks smaller, less massive, and broadly consistent with ALMA survey demographics.","keywords":["protoplanetary disks","magnetocentrifugal disk winds","thin-disk MHD simulations","gravitoviscous disk evolution","synthetic ALMA observations","dust evolution and drift","magnetic lever arm"],"falsifier":"Measure the vertical magnetic field strength in Class II disks at radii from about 1 to 100 au using Zeeman or Faraday-rotation techniques; if the ratio of the simulation's ideal-MHD field to the measured field is not consistently near 1000, or if the required ratio varies with radius or time, the $C_\\beta$ scaling and the wind torques derived from it fail.","tokens_in":42168,"feed_emoji":"🌬️","tokens_out":8894,"duration_ms":86615,"temperature":0.7,"pith_summary":"The paper proposes that protoplanetary disk evolution in the Class II stage is not driven by turbulent viscosity alone, but by a tug-of-war between gravitoviscous spreading and magnetic disk winds that remove mass and angular momentum vertically. It constructs a global wind prescription from local shearing-box MHD results, embeds it in a thin-disk formation-and-evolution code that already includes self-gravity, adaptive viscosity, and two-component dust, and then post-processes the output with a radiation thermo-chemical code to produce ALMA-like continuum and CO maps. The wind-inclusive models produce Class II disks that are smaller and less massive than their gravitoviscous counterparts, and after adjusting two wind calibration parameters the synthetic sizes, dust masses, and line fluxes fall largely within the ranges of ALMA surveys of nearby star-forming regions. The authors argue this makes magnetic winds a necessary ingredient for explaining observed disk demographics, not merely a refinement of viscous theory.","feed_headline":"Magnetic winds shrink protoplanetary disks toward ALMA sizes","feed_subtitle":"Adding vertical winds to a gravitoviscous disk model makes synthetic Class II disks smaller, lighter, and broadly consistent with surveys.","key_machinery":"The load-bearing object is a pair of power-law fitting formulae (Eqs. 25 and 26) that give the wind mass-loss rate and Maxwell stress as functions of local surface density, sound speed, plasma $\\beta$, stellar mass, radius, and total luminosity. The Maxwell stress term enters the azimuthal momentum equation as a sink and vertically removes angular momentum, while the mass-loss term removes gas and a proportional amount of small dust. Their radial and luminosity dependencies come from shearing-box fits, with three modifications: a rescaling of radius to arbitrary stellar mass, a box-height factor $C_H H_g/r$, and FUV-luminosity factors that boost winds during accretion outbursts. The critical calibration is $C_\\beta = 1000$, applied because the ideal-MHD vertical field in the base code is over-strong; this is the assumption that sets the overall wind strength. Winds are active only inside the centrifugal radius and above a gas surface density threshold of $\\Sigma_g = 0.1$ g cm$^{-2}$.","core_discovery":"The central claim, stated on the paper's own terms, is that a semi-analytic magnetic disk wind model, built from the fitting formulae of local shearing-box simulations, generalized to arbitrary stellar mass and FUV luminosity, and calibrated by a single factor $C_\\beta = 1000$ multiplying the midplane plasma $\\beta$, can be embedded self-consistently in the FEOSAD thin-disk MHD code and yields disks that compare favorably with ALMA Class II surveys. In these models the winds act as sinks of gas, small dust, and angular momentum inside the centrifugal disk, producing advective evolution in which disks tend to contract rather than spread. The result is that wind disks are smaller and less massive than gravitoviscous-only disks; at low turbulent viscosity the default winds over-shrink disks to a few au, while adjusting $C_\\beta$ and $C_H$ recovers observed sizes. The paper claims that the synthetic observations from the adjusted model fall within the observed dust radius, gas radius, and dust mass distributions, and that the inferred global magnetic lever arm settles to $\\lambda \\approx 1$--$3$ during the Class II stage, in agreement with MHD simulations.","pith_inferences":["If $C_\\beta$ is actually not constant in radius, the dust-trapping rings and the shrinking-continuum-radius trend predicted here would shift; mapping the wind stress profile observationally through disk size versus age could constrain that radial variation.","A testable signature separating wind from viscous evolution is that millimeter-dust radii shrink with age while CO gas radii continue to grow; targeted ALMA observations of clusters with known ages could look for this divergence.","The calibration implies an unstated self-regulation conjecture: real Class II disks must cluster near $\\beta_0 \\sim 10^4$; Zeeman or Faraday-rotation measurements across a disk sample would test it directly."],"forward_implications":["Gravitoviscous-only evolution yields Class II disks that are typically too large and too massive; matching survey demographics appears to require wind-driven angular momentum removal.","With winds, disks evolve advectively: their sizes tend to shrink over time, and the final radius is set by the balance between gravitoviscous spreading and wind contraction.","For low turbulent $\\alpha$ ($10^{-4}$--$10^{-3}$), the default winds are so strong that disks collapse to a few au, so observational sizes constrain the wind calibration parameters $C_\\beta$ and $C_H$.","The inferred global magnetic lever arm stays near 1--3 throughout the Class II stage, supporting simpler long-term wind models that assume a constant lever arm.","Wind models lose more mass and evolve on shorter timescales than gravitoviscous models, leaving less gas for giant planet formation by the Class II stage."],"supporting_citations":[{"why":"Supplies the shearing-box fitting formulae for wind mass-loss rate and wind stress that the paper modifies.","marker":"Bai 2013"},{"why":"Documents how wind mass loss and stress scale with FUV penetration depth, motivating the luminosity corrections.","marker":"Bai & Stone 2013"},{"why":"Provides the FEOSAD thin-disk MHD code, its equations, and the gravitoviscous baseline into which winds are inserted.","marker":"Vorobyov et al. 2020b"},{"why":"Provides the ProDiMo radiation thermo-chemical code used to produce synthetic continuum and CO observations.","marker":"Woitke et al. 2016"},{"why":"Supplies the ALMA Lupus gas-radius and dust-radius distributions used for direct comparison.","marker":"Ansdell et al. 2018"},{"why":"Supplies the ALMA survey dust-mass distributions for Class II disks.","marker":"Manara et al. 2023a"},{"why":"Provides the observational constraint that total wind mass loss should be of the order of the accretion rate.","marker":"Pascucci et al. 2023"},{"why":"Supplies the advective wind-driven disk model predicting disk contraction, which the results confirm.","marker":"Tabone et al. 2022"},{"why":"Compiles the survey data and flux-to-mass conversion assumptions used in the synthetic mass estimates.","marker":"Miotello et al. 2023"}],"fun_headline_variants":["Magnetic winds shrink disks to ALMA-observed sizes","Wind-driven accretion contracts protoplanetary disks","Disk wind model reproduces ALMA Class II disk sizes","Magnetic disk winds reduce disk size and mass"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole wind torque and mass-loss calibration rests on one number: the factor $C_\\beta = 1000$ that converts the simulated ideal-MHD vertical field into the true midplane plasma $\\beta$, assumed constant in space and time.","fun_headline_variants_meta":{"raw":{"variants":["Magnetic winds shrink disks to ALMA-observed sizes","Wind-driven accretion contracts protoplanetary disks","Disk wind model reproduces ALMA Class II disk sizes","Magnetic disk winds reduce disk size and mass"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000429,"raw_usage":{"total_tokens":2263,"prompt_tokens":1087,"completion_tokens":1176,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":703,"completion_tokens_details":{"reasoning_tokens":1114}},"tokens_in":703,"tokens_out":1176,"duration_ms":10689,"temperature":1.0,"reasoning_tokens":1114,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T19:58:07.864558+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the vertical magnetic field strength in Class II disks at radii from about 1 to 100 au using Zeeman or Faraday-rotation techniques; if the ratio of the simulation's ideal-MHD field to the measured field is not consistently near 1000, or if the required ratio varies with radius or time, the $C_\\beta$ scaling and the wind torques derived from it fail.","supporting_citations":[{"cited_title":"2023, in Astronomical Society of the PacificConferenceSeries,Vol.534,ProtostarsandPlanetsVII,ed.S.Inutsuka, Y","cited_arxiv_id":null,"evidence_quote":"Provides the observational constraint that total wind mass loss should be of the order of the accretion rate."},{"cited_title":"C., & Kataoka, A","cited_arxiv_id":null,"evidence_quote":"Compiles the survey data and flux-to-mass conversion assumptions used in the synthetic mass estimates."}],"review_version":1}