{"id":"6c78c8fb-dc6d-4831-a5fd-fe67a50a31c1","arxiv_id":"2501.16169","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A global model that includes dust torque predicts that low-mass planets forming inside the water ice line can migrate outward, while planets forming beyond it are barely affected.","lead":"This paper adds the torque from dust and pebbles into a global model of planet formation, following small planetary embryos as they grow by gas and pebble accretion. It finds that for planets forming inside the water ice line, this dust torque can reverse migration and push them outward, changing their final masses and compositions.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Clamping inner-disk Stokes numbers to the BLP18 grid minimum (St=0.01) means the outward-migration result is obtained at the boundary of the torque map, not from resolved data; extending the map to St≲1e-3 is needed before the quantitative tracks can be trusted.","rationale":"The reader's weakest assumption is exactly the load-bearing point. Re-reading Sec. 2, Sec. 5, and Appendix A confirms that the inner-disk tracks are computed with Stokes numbers clamped to the minimum grid value, and that the multi-species check does not resolve the issue because all inner-disk species are below the grid minimum. The paper's central quantitative claim, outward migration for planets forming inside the water ice line, rests on this boundary value rather than on resolved torque data. The two torque prescriptions (BLP18 and Chrenko et al. 2024) are consistent with each other in this regime only because both are clamped, so that consistency does not provide independent support. I do not view this as a reason to reject the paper: there is external evidence that dust torque can be positive for low Stokes numbers (BLP18; Regaly 2020; Chrenko et al. 2024; Hou & Yu 2024, 2025), and the authors are transparent about the limitation. But the specific quantitative tracks, final masses, and composition changes should be treated as conditional until the torque map is extended to St<0.01 and the inner-disk runs are repeated without clamping. The reader already reached CONDITIONAL with moderate confidence, so my concern does not change the verdict; it sharpens the justification for that conditionality.","tokens_in":16023,"tokens_out":6755,"duration_ms":65464,"concrete_test":"Run local 2D hydrodynamical simulations with the BLP18 setup at St=1e-4, 3e-4, and 1e-3 for planet masses 0.01, 0.05, and 0.1 M⊕ at the reference radius (with and without dust back-reaction at ϵ=0.1), and build an extended torque map. Rerun the fiducial and metal-rich PLANETALP tracks of Secs. 3.2-3.4 using this map evaluated at the actual local St instead of clamping to 0.01. If the positive torque persists and is within a factor of two of the clamped value, the central claim survives; if the torque drops substantially or changes sign, the outward migration is an artifact of the boundary clamp. A cheaper check is to re-run with the dust torque set to zero for St<0.01; if outward migration disappears, the result is wholly dependent on the clamp.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The inner-disk outward-migration result is computed at the boundary of the BLP18 torque grid, not from resolved data. In the fiducial and metal-rich disks, the mass-weighted mean Stokes number inside the water ice line is 10^-4 to 10^-3 (Fig. 1, Sec. 3.1; Fig. 8), while the BLP18 grid starts at St=0.01. As stated in Sec. 2, the code clamps both planet mass and Stokes number to the minimum grid values rather than extrapolating, and Appendix A confirms that inside the ice line every dust species has Stmax<0.01, so the multi-species sum is exactly the clamped value. The same boundary value is used for the Chrenko et al. (2024) prescription in Sec. 4 because no extrapolation is allowed (Sec. 5). Consequently, the claimed outward migration of planets starting at 0.5-2 au, the mass reductions in the fiducial run, and the ice-line crossings in the metal-rich run all rest on assigning the torque computed at St=0.01 (and at the minimum planet mass) to a regime two orders of magnitude lower in Stokes number. Agreement between the BLP18-based and Chrenko-based runs is not independent confirmation of this point, since both are clamped. The authors acknowledge the limitation in Sec. 5, but it is not peripheral: it is the exact step that produces the paper's headline effect. The torque at St=10^-3 could be smaller, larger, or opposite in sign; the current model cannot distinguish.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper incorporates the dust torque from Benitez-Llambay & Pessah (2018, BLP18) into the PLANETALP global planet formation model, which includes axisymmetric disk evolution with viscous accretion and X-ray photoevaporation, dust growth and transport, and concurrent pebble and gas accretion. The authors compute planet formation tracks for three disk models (fiducial Mdisk=0.05 Msun, massive Mdisk=0.1 Msun, and metal-rich Z=0.02), comparing runs with and without the dust torque, and also apply the recent Chrenko et al. (2024) prescriptions that include the dynamical effect of pebble accretion. The main findings are that the dust torque drives substantial outward migration for planets initially inside the water ice line, with consequences for their final masses, formation timescales, and compositions, while planets starting beyond the ice line are only mildly affected. The paper concludes that the dust torque should not be ignored in models of planet formation.","tokens_in":16322,"tokens_out":3282,"duration_ms":31434,"significance":"If the qualitative result holds, this is a valuable step toward including dust-driven torques in global models of planet formation. The paper makes good use of two independent torque prescriptions and consistently applies a conservative no-extrapolation policy, and it is transparent about the limitations of the underlying hydrodynamical maps. The qualitative conclusion that the dust torque can reverse inward migration of low-mass planets inside the ice line is plausible and worth testing with more detailed simulations. However, the quantitative tracks in the inner disk depend critically on clamping the Stokes number to the minimum of the torque grid, which is exactly the regime where the model is least constrained. The significance of the paper will be much stronger if the torque maps are extended to lower Stokes numbers and lower planet masses, or if the clamping is validated by additional simulations.","major_comments":[{"comment":"","section":"Sec. 2, Dust-to-gas Mass Ratio"},{"comment":"","section":"Sec. 4 and Fig. 11"},{"comment":"","section":"Sec. 5, The Dynamical Impact of Pebble Accretion and Sec. 5, Limitations"}],"minor_comments":[{"comment":"","section":"Caption of Fig. 1"},{"comment":"","section":"Captions of Figs. 5 and 10"},{"comment":"","section":"Footnote 3"},{"comment":"","section":"Sec. 5, paragraph on embryo formation"},{"comment":"","section":"Title of Sec. 5"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of the journal and the qualitative claim is interesting. The main weakness is that the central inner-disk result is computed at the boundary of the torque grid, and the comparison with Chrenko et al. does not resolve this because both inputs are clamped. I would like the authors to address this head-on, ideally with a sensitivity test or by extending the torque maps to lower Stokes numbers. The manuscript is well written and the limitations are acknowledged, but the load-bearing nature of the clamping issue requires major revision before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is worth taking seriously, but read the fine print on the Stokes-number clamp. The paper does the first genuinely global implementation of dust torque in a pebble-accretion formation model, with self-consistent dust growth, disk evolution, and two independent torque prescriptions. That is a real step beyond Guilera et al. 2023, which used simpler alpha disks. The central physical mechanism — dust drifts inward, piles up inside the ice line, and produces a positive torque that can reverse inward migration of low-mass cores — is presented clearly and is consistent across three disk setups and both torque maps. Credit where due: the authors are unusually transparent about limitations; Section 5 and Appendix A spell out exactly where the model is being pushed past its input grid.\n\nThe soft spot is not peripheral. Inside the water ice line, the mass-weighted mean Stokes number is 10^-4 to 10^-3 (Fig. 1, Fig. 8), while the BLP18 torque map starts at St=0.01. The code clamps to the grid minimum rather than extrapolating, so the headline outward migration of planets starting at 0.5-2 au is computed at the boundary of the torque table, not from resolved data. The stress-test note is right: agreement between the BLP18-based and Chrenko-based runs is not independent confirmation, because both are clamped the same way. The torque at the true Stokes numbers could be smaller, larger, or opposite in sign. The authors acknowledge this (Sec. 5) and frame the result as expected significance rather than measured, which is honest, but it means the quantitative tracks — final masses, ice-line crossings, compositions — are conditional on an assumption that has not been tested.\n\nMinor secondary issues: no code or data are released, so the tracks cannot be reproduced easily; there is no convergence or uncertainty analysis. The 2D torque maps also neglect migration and dust back-reaction, which the authors note. For the outer-disk planets the conclusions are robust because the dust torque is weak there regardless. The inner-disk result is the one that needs independent torque maps extended to low St before it can be taken as quantitative.\n\nWho is this for: pebble-accretion modelers and anyone interpreting super-Earth populations. It deserves a serious referee; the implementation effort and transparency are real contributions even if the inner-disk numbers are provisional.","headline":"First genuine global implementation of dust torque in a pebble-accretion formation model, with a real caveat: the headline inner-disk outward migration is computed at the boundary of the torque grid, so the direction is plausible but the quantitative tracks are conditional.","tokens_in":16894,"tokens_out":1536,"would_cite":true,"duration_ms":14181,"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":"The paper argues that the dust torque from drifting pebbles can dominate and reverse inward migration for low-mass planets forming inside the water ice-line, so planet formation models should include it.","keywords":["protoplanetary disks","planetary migration","dust torque","pebble accretion","planet formation","water ice line","planet-disk interactions"],"falsifier":"Run a 3D hydrodynamical simulation of a roughly 0.1–0.4 M⊕ planet at 0.5–1 au with dust Stokes numbers of $10^{-4}$ to $10^{-3}$, a dust-to-gas ratio around 0.1, and free planetary migration; if the net torque is negative or the planet does not migrate outward despite the enhanced dust, then the clamping of the torque map to St = 0.01 is the cause and the central result fails.","tokens_in":15802,"feed_emoji":"🪐","tokens_out":5871,"duration_ms":51296,"temperature":0.7,"pith_summary":"The paper tries to show that the torque exerted by dust and pebbles on a growing planet is not a negligible correction: inside the water ice-line it can dominate the total torque early in a planet's life and push the planet outward, reversing the usual inward migration. Using a global disk model that evolves gas, dust, and pebbles and lets planets grow by pebble and gas accretion, the authors find that this outward push changes final masses, formation timescales, and sometimes composition. For planets born beyond the ice line, the dust torque merely slows inward migration because the dust-to-gas ratio drops quickly there. If true, models of terrestrial and low-mass planet formation that ignore the dust torque will misplace planets and misestimate their growth histories.","feed_headline":"Pebble dust can reverse low-mass planet migration early","feed_subtitle":"In a global formation model, the dust torque dominates inside the water ice-line, changing final masses.","key_machinery":"The load-bearing object is the dust torque map of BLP18, a two-dimensional grid of the torque as a function of planet mass and particle Stokes number (the dimensionless measure of how strongly a particle is coupled to the gas), computed for a reference disk and scaled by the local dust-to-gas ratio relative to 0.01. In the global model the total torque is Γtot = Γg + Γd, with Γg from Jiménez & Masset (2017) and Γd obtained by bilinear interpolation in planet mass and mass-weighted mean Stokes number, rescaled by ϵP/ϵ0. The mechanism that carries the argument is the inward drift of ice-rich pebbles: beyond the ice line particles grow to larger Stokes numbers and drift rapidly, enriching the inner disk in dust, which raises the positive dust torque there. A conservative clamp at the minimum grid values (St = 0.01 and the lowest tabulated planet mass) avoids extrapolation, so the inner-disk results rest on the torque at the grid boundary.","core_discovery":"The central claim is that the dust torque, computed from the asymmetric distribution of pebbles around an embedded planet, can be a dominant component of the total torque for low- and intermediate-mass planets forming inside the water ice-line. As pebbles drift inward from the outer disk, they raise the dust-to-gas mass ratio in the inner region by more than an order of magnitude within the first million years; the positive dust torque then exceeds the negative gas Lindblad torque up to planet masses of roughly 0.4 M⊕ in the fiducial case, producing outward migration. The outcome depends on the destination: in the fiducial and massive disks, outward-moving planets reach regions of lower dust-to-gas ratio, grow less efficiently, and end up less massive; in a metal-rich disk, they stay in high-ratio regions, reach larger masses, and can acquire envelopes. When the torque prescriptions include pebble accretion, the effect strengthens for low-mass, low-Stokes-number planets, allowing them to cross the ice line and accrete icy pebbles. The paper concludes that dust torque should not be ignored in planet formation models.","pith_inferences":["If the dust torque acts as modeled, planet formation models that omit it should predict systematically different final positions and masses for close-in low-mass planets; comparing the model's mass–period distribution against exoplanet demographics would be a direct test.","Because the model clamps the Stokes number to 0.01 inside the ice line, where the true mass-weighted mean is 10^-4 to 10^-3, the reported dust torque is a conservative estimate; extending the maps to lower Stokes numbers could strengthen, or if the torque turns over, weaken the outward migration result.","The paper's discussion of dust back-reaction suggests the torque may be even larger for Mars-mass planets, so the outward-migration window might extend to smaller embryos than the roughly 0.4 M⊕ threshold found in the fiducial run."],"forward_implications":["Planets that start inside the water ice-line can migrate outward early in their evolution, with final masses either lower or higher than in dust-free models depending on the dust-to-gas ratio of the region they enter.","In metal-rich disks, outward migration can keep inner planets in high dust-to-gas regions, allowing them to grow larger cores and acquire gas envelopes, and in some cases to cross the ice line and incorporate icy pebbles.","For planets initially beyond the ice line, the dust torque only delays inward migration and does not substantially alter their formation tracks, because the dust-to-gas ratio falls quickly and formation timescales are short.","Including pebble accretion in the torque prescription strengthens the outward migration of low-mass, low-Stokes-number planets inside the ice line, reinforcing the conclusion that dust torque belongs in global planet formation models."],"supporting_citations":[{"why":"Supplies the two-dimensional dust torque maps in planet mass and Stokes number that enter the model as Γd.","marker":"BLP18"},{"why":"Builds the smooth interpolated torque maps used here and the scaling with dust-to-gas ratio.","marker":"Guilera et al. 2023"},{"why":"Provides the analytical prescriptions for the dust torque including pebble accretion, used to test the sensitivity of the tracks.","marker":"Chrenko et al. 2024"},{"why":"Supplies the gas Lindblad and corotation torque prescriptions that combine with the dust torque in Γtot.","marker":"Jiménez & Masset 2017"},{"why":"Defines the pebble accretion and gas envelope growth model within the global formation framework.","marker":"Venturini et al. 2020b"},{"why":"Sets the fiducial low-viscosity disk model and the planet formation setup used for the main runs.","marker":"Guilera et al. 2021"}],"fun_headline_variants":["Dust torque flips low-mass planet migration outward","Pebble drift boosts dust torque, pushing planets outward","Inside ice line, dust torque dominates planet migration","Dusty inner disk can reverse planet migration","Pebble accretion amplifies dust torque for low-mass planets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The outward migration result assumes that dust torque maps computed in 2D, without planetary migration and dust back-reaction, and tabulated only down to a Stokes number of 0.01, remain valid when interpolated and linearly scaled to the low Stokes numbers and enhanced dust-to-gas ratios that actually occur inside the water ice-line in the model.","fun_headline_variants_meta":{"raw":{"variants":["Dust torque flips low-mass planet migration outward","Pebble drift boosts dust torque, pushing planets outward","Inside ice line, dust torque dominates planet migration","Dusty inner disk can reverse planet migration","Pebble accretion amplifies dust torque for low-mass planets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000551,"raw_usage":{"total_tokens":2663,"prompt_tokens":1012,"completion_tokens":1651,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":628,"completion_tokens_details":{"reasoning_tokens":1574}},"tokens_in":628,"tokens_out":1651,"duration_ms":12368,"temperature":1.0,"reasoning_tokens":1574,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T13:39:20.807732+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a 3D hydrodynamical simulation of a roughly 0.1–0.4 M⊕ planet at 0.5–1 au with dust Stokes numbers of $10^{-4}$ to $10^{-3}$, a dust-to-gas ratio around 0.1, and free planetary migration; if the net torque is negative or the planet does not migrate outward despite the enhanced dust, then the clamping of the torque map to St = 0.01 is the cause and the central result fails.","supporting_citations":[{"cited_title":"M., Miller Bertolami , M","cited_arxiv_id":null,"evidence_quote":"Sets the fiducial low-viscosity disk model and the planet formation setup used for the main runs."}],"review_version":1}