{"id":"0db64147-b9bf-4688-a700-12d3f1ddf3c5","arxiv_id":"1908.11806","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Turbulence and ambipolar diffusion act as complementary drivers of disk formation in magnetized core collapse, producing persistent but strongly magnetized disks.","lead":"Simulations of collapsing magnetized cloud cores show that turbulence creates a large disk early, while ambipolar diffusion allows that disk to survive, and the two effects work better together than either alone. The result sharpens the long-running debate over how magnetic fields shape disk formation and how strongly magnetized young protoplanetary disks are.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The AD timestep floor in §2.2(ii) caps ambipolar diffusivity in low-density, moderately magnetized cells—exactly where AD-driven flux redistribution sets disk survival; the persistence threshold and plasma-β values are therefore not yet tested against unfloored AD.","rationale":"The reader's weakest-assumption analysis identifies the same load-bearing concern: the AD timestep floor. My independent read of §2.2(ii) and equation (7) agrees that this is the most insecure step in the central argument. The floor does not introduce an internal inconsistency, and the broad parameter trends (monotonic with QA, consistent with prior ideal-MHD turbulence work) give real but incomplete support. The concern is about the quantitative content of the main claims: the specific AD threshold for persistent disks and the plasma-β values that drive the paper's proposed tension with protoplanetary-disk simulations. Because the floor acts in low-density, moderately magnetized cells, it can suppress exactly the flux-redistribution mechanism that the paper argues enables disk survival and demagnetization. The paper reports only that a tiny mass fraction is affected, which is not the right diagnostic for magnetic flux transport; a small amount of mass can carry a dynamically important amount of flux. The concrete test I propose would either confirm the threshold and β values or reveal that they are floor artifacts. In the latter case, the qualitative complementary picture would survive but the quantitative claims would need revision. Therefore the appropriate verdict remains CONDITIONAL, matching the reader's assessment; no verdict change is needed.","tokens_in":32655,"tokens_out":7231,"duration_ms":70433,"concrete_test":"Re-run models M1.0AD0.3 and M1.0AD3.0 with the AD timestep floor reduced by a factor of 100 (or, if feasible, with an implicit or super-time-stepping AD scheme that removes the floor), keeping all other parameters and resolution identical. Compare disk cell counts, λ_d, and plasma-β at M* = 0.20 and 0.25 M⊙ against Fig. 16 panels (o)–(q) and (u)–(w). If M1.0AD0.3 develops a persistent disk, or if β in M1.0AD3.0 changes by more than a factor of 2, the persistence threshold and magnetization claims are not robust to the floor treatment.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's quantitative claims—that 'relatively strong' ambipolar diffusion is needed for persistent disks and that formed disks have plasma-β of order a few tens or smaller—depend on the AD timestep floor introduced in §2.2(ii). Equation (7) shows Δt_AD ∝ ρ^{3/2}/B^2, so the floor is triggered precisely in low-density, moderately magnetized envelope cells where AD acts fastest. In those cells the code reduces η_A from its physical value. The authors state only that a tiny amount of mass is affected, but the affected cells are the same region where AD is supposed to redistribute magnetic flux outward and weaken magnetic braking. If the floor suppresses that redistribution, then the simulated disks may retain more flux than the physical AD would allow, making disks appear harder to form and more strongly magnetized than they really are. The threshold QA ≈ 1 for disk persistence and the derived β values are thus not yet separated from this numerical approximation. The qualitative picture—turbulence helps early disk formation, AD helps late survival—would likely survive a floor-free test, but the quantitative threshold and the '2–3 orders of magnitude' magnetization tension could shift substantially.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a series of 3D non-ideal MHD simulations of disk formation in rotating, magnetized molecular cloud cores, combining for the first time initial turbulence and ambipolar diffusion (AD) during the protostellar mass accretion phase, using the Athena code with sink particles and a zoom-in grid. The authors survey a parameter grid with turbulent Mach numbers M = 0, 0.5, 1 and AD coefficients QA from 0 to 10 times the standard value. They find that turbulence enables the early formation of a relatively large disk, that AD is required for such a disk to persist to later times, and that the two effects work constructively: turbulence helps at early times and AD promotes survival. They also report that AD-enabled disks in laminar runs tend to fragment, with fragmentation suppressed by turbulence. A second central claim is that the formed disks inherit strong magnetic fields, with plasma beta on the order of a few tens or smaller, 2-3 orders of magnitude below typical values adopted in protoplanetary disk simulations.","tokens_in":1641,"tokens_out":1688,"duration_ms":59570,"significance":"If the results hold, this is a useful step forward in the disk formation debate, being among the first studies to treat turbulence and ambipolar diffusion together in the protostellar accretion phase with a sink treatment. The systematic scan over QA and Mach number, the identification of the diffusion-DEMS in 3D, and the quantification of disk magnetization are valuable contributions. The paper also provides useful diagnostics linking the AD timestep floor to its potential impact, and it is careful to note where its conclusions are tentative. However, the main quantitative claims -- the AD strength threshold for persistent disks and the plasma-beta values -- are directly coupled to a numerical approximation whose effect has not been tested. This makes the significance conditional on a relatively inexpensive numerical check.","major_comments":[{"comment":"The AD timestep floor lowers the ambipolar diffusivity in low-density, moderately magnetized cells, which are exactly the cells where AD acts fastest and where AD-driven flux redistribution is most important. Equation (7) shows that Δt_AD is smallest precisely at low ρ and moderate B, and the floor caps η_A there. The authors state that only a tiny amount of mass is affected, but the affected cells are in the envelope where the flux must diffuse outward to reduce magnetic braking; the mass in those cells can be small while their dynamical effect is large. The paper does not test how the results change when the floor is removed or made less restrictive, nor does it report the actual reduction factor of η_A in the affected cells. The threshold QA ≈ 1 for persistent disks (Section 5) and the derived disk plasma-β values (Section 6.1) therefore remain entangled with this numerical approximation. I request a convergence test with the floor removed or substantially relaxed for at least the key models M0.0AD1.0 and M1.0AD1.0, together with a quantitative report of the mass fraction and the reduction factor of η_A in the affected cells.","section":"§2.2(ii), Eq. (7)"},{"comment":"The zoom-in resolution is about 10 au, and the 'well-developed disk' criterion requires more than 100 disk cells. With typical disk radii of a few tens to ~100 au, the disks are resolved by only a small number of cells across, and the disk-selection criteria (f = 2, density threshold 3.8e-15 g cm^-3) are applied at this coarse scale. The resolution study in Section 6.2 is only a lower-resolution uniform-grid run that reproduces the same qualitative trend; it does not demonstrate convergence of the disk magnetization diagnostics (λd and β) or of the disk size. Given that the paper's quantitative claim about plasma-β being 2-3 orders of magnitude below protoplanetary disk values is likely resolution-sensitive, I ask for at least one higher-resolution zoom-in run (e.g., of M1.0AD3.0 or M1.0AD10.0) to show that the reported β and λd values do not change substantially with resolution.","section":"§2.4, §6.1"},{"comment":"The classification of a disk as present or absent relies on specific thresholds (50 and 100 disk cells) that the authors themselves call 'somewhat arbitrary.' The quantitative conclusion that a persistent disk requires QA ≳ 1 is based on these thresholds, which are not justified by any sensitivity test. While the morphological maps support the qualitative trend, the specific threshold value in QA should be presented as a function of the disk definition, or a broader set of selection parameters should be tested to show that the qualitative conclusion is robust. This is not a request for an entirely new study, but a statement of the uncertainty associated with the stated threshold.","section":"§6.1, Figs. 15-16"}],"minor_comments":[{"comment":"Please specify the numerical value of the AD timestep floor (the minimum allowed Δt_AD or the equivalent maximum η_A) and the algorithm used to enforce it, so that readers can assess the strength of the approximation.","section":"§2.2(ii)"},{"comment":"There is a typo: 'pniched' should be 'pinched'.","section":"Fig. 10 caption"},{"comment":"The text uses 'plamsa-β' and 'polaroidal' (e.g., in the caption of Fig. 15); these should be 'plasma-β' and 'poloidal'.","section":"Figs. 15-16 captions"},{"comment":"The text contains the typo 'AD coeffiecent' (should be 'AD coefficient').","section":"§5, near Fig. 13"},{"comment":"The comparison to Gray et al. (2018) refers to a 'large misalignment between the turbulence-induced angular momentum (which is set to zero for our simulation as a whole) and the magnetic field.' Please clarify how the net angular momentum was removed from the turbulent velocity field; a short description of the projection method would be useful.","section":"§6.2"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of MNRAS and the qualitative conclusions are supported by the systematic parameter scan. The main concern is the AD timestep floor, which has the potential to bias the quantitative thresholds and the plasma-β values in the direction that strengthens the paper's headline claims. The required tests (removing the floor for a couple of models, and a higher-resolution run) are feasible within the manuscript's scope and would make the conclusions substantially more convincing."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"I read this one with the stress-test note in hand, and the note is right: the AD timestep floor is the paper's real weak spot. Equation (7) shows the floor kicks in where eta_A is largest—low-density, moderately magnetized cells—which is precisely where AD should be redistributing flux and weakening magnetic braking. The authors say only a tiny mass is affected and move on. That is not enough, because the disk-persistence threshold (QA about 1) and the headline plasma-beta values (a few tens or smaller) both rest on how much flux AD actually removes. A floor-free test, or at least a sensitivity run with a less aggressive floor, would separate the physics from the approximation. If that test shifts the threshold by even a factor of a few in QA, the quantitative version of the story changes, even if the qualitative picture survives.\n\nWhat the paper does well is real. This is the first 3D sink-treatment study I know of that carries both core-scale turbulence and ambipolar diffusion through the protostellar accretion phase. The parameter scan over QA is systematic, the trends are monotonic, and the limiting cases line up with earlier work: ideal MHD turbulence makes transient disks, laminar AD alone makes persistent disks only when diffusion is strong. The complementary picture—turbulence builds the early disk, AD lets it persist, and turbulence suppresses the fragmentation that otherwise plagues AD disks—comes through clearly. The negative effect of turbulence (it makes AD disks more strongly magnetized and therefore harder to survive) is new and worth taking seriously. The paper also includes a genuine attempt at a quantitative disk definition, reports lambda_d and beta for each disk cell selection, and acknowledges its own simplifications: isothermal gas, one-power ionization equilibrium, coarse 10 au resolution, and only a crude convergence check. No circularity, no fitting to targets.\n\nSoft spots beyond the timestep floor are proportionate. Resolution is coarse enough that small, unresolved disks could hide in the no-disk runs; the authors say this themselves. The disk selection factor f=2 is arbitrary but standard. The plasma-beta tension with protoplanetary disk simulations is provocative but inherited from these approximations, so I would not yet use it to rewrite initial conditions in disk evolution models.\n\nBottom line: this deserves a serious referee. The central claim is likely robust, but the quantitative thresholds and magnetization numbers need the floor-free check and ideally higher resolution before being cited as constraints. I would bring it to reading group and would likely cite it for the complementary formation picture and the turbulence-induced magnetization effect.","headline":"A solid, systematic simulation study showing turbulence and ambipolar diffusion work together to build persistent but strongly magnetized disks; the main caveat is an AD timestep floor that may under-cook the diffusion in exactly the cells that matter.","tokens_in":33518,"tokens_out":1354,"would_cite":true,"duration_ms":15593,"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":"Turbulence and ambipolar diffusion work together to form large, persistent, strongly magnetized protostellar disks in 3D collapse simulations.","keywords":["protostellar disk formation","ambipolar diffusion","turbulence","magnetic braking","magnetohydrodynamics","plasma beta","star formation","pseudodisk"],"falsifier":"Re-run the same core-collapse simulations with the ambipolar-diffusion timestep floor removed or weakened and check whether the persistent disks in the standard-diffusion models (M0.0AD1.0 and M1.0AD1.0) still form and survive; if they disappear, the disk-persistence claim rests on the numerical floor rather than on ambipolar diffusion itself.","tokens_in":32452,"feed_emoji":"🌀","tokens_out":7213,"duration_ms":58274,"temperature":0.7,"pith_summary":"This paper uses three-dimensional magnetohydrodynamic simulations of collapsing rotating cores to argue that turbulence and ambipolar diffusion each help solve the magnetic braking catastrophe, but on different timescales. Turbulence alone forms a large, near-Keplerian disk early in the protostellar accretion phase, yet that disk is so strongly magnetized that it later gets destroyed by a magnetically dominated structure. Ambipolar diffusion alone can enable a persistent disk, but only if it is relatively strong, and such disks tend to fragment gravitationally. Together, turbulence and strong ambipolar diffusion produce large, persistent, and stable disks throughout the accretion phase: turbulence seeds the disk early, and ambipolar diffusion weakens and unpinches the field so magnetic braking does not kill it later. The disks that survive remain strongly magnetized, with plasma $\\beta$ of order a few tens or less, two to three orders of magnitude lower than values commonly adopted in protoplanetary disk simulations.","feed_headline":"Turbulence starts disks; ambipolar diffusion keeps them alive","feed_subtitle":"3D collapse simulations show the pair forms large, stable, strongly magnetized protostellar disks.","key_machinery":"The load-bearing machinery is ambipolar diffusion acting as a magnetic-flux redistribution agent, quantified by the diffusion coefficient $\\eta_A = Q_A B^2/(4\\pi\\rho^{3/2})$ with ion density $\\rho_i = C\\rho^{1/2}$, together with the turbulence-induced warping of the pseudodisk. Ambipolar diffusion lowers the vertical field strength $B_z$ near the protostar, creates a strong-field plateau called the 'diffusion-DEMS', and reduces azimuthal pinching as measured by the field curvature $\\kappa_\\phi$; both effects reduce the magnetic braking torque $-\\Gamma_z = (1/c)[(J\\times B)\\times r]_z$. The sink-particle treatment, which removes mass and momentum while preserving the magnetic flux, lets the simulations follow the accretion phase, while the warped pseudodisk provides the early disk by letting high-angular-momentum material retain its angular momentum as it falls toward the center.","core_discovery":"The central claim is that turbulence and ambipolar diffusion are complementary disk-formation agents in the main protostellar accretion phase. Sonic turbulence warps, but does not destroy, the magnetically flattened pseudodisk that feeds the protostar, and it enables a relatively large disk to form early whether or not ambipolar diffusion is present. That early disk does not survive unless ambipolar diffusion is strong enough: ambipolar diffusion redistributes magnetic flux outward and reduces azimuthal field-line pinching in the circumstellar region, lowering the magnetic braking torque that would otherwise strip the disk's angular momentum. In laminar runs, strong ambipolar diffusion produces disks that tend to fragment; initial turbulence suppresses that fragmentation. The paper therefore concludes that turbulence and ambipolar diffusion work together constructively, and that the disks formed this way inherit magnetic fields with plasma beta of order a few tens or smaller, substantially more magnetized than the initial conditions usually assumed in protoplanetary disk modeling.","pith_inferences":["If the strong magnetization persists through the protostellar phase, late-stage protoplanetary disk models may need initial poloidal fields with $\\beta$ near unity rather than $10^4$-$10^5$; this would change the balance between MRI and wind-driven accretion as well as planet migration rates.","A direct observational test is possible: ALMA Zeeman or dust-polarization measurements of young, embedded disks should find ordered poloidal fields with $\\beta$ in the range of a few to a few tens if this picture is right, whereas older T Tauri disks should be more demagnetized.","The AD timestep floor, which lowers $\\eta_A$ in low-density moderately magnetized cells, is the main numerical caveat; removing it or varying its threshold would show whether the persistence threshold in $Q_A$ is physical or an artifact."],"forward_implications":["A rotationally supported disk formed early by turbulence is transient in ideal MHD; its survival requires an ambipolar diffusivity at least of order the standard value $Q_A$.","Strong ambipolar diffusion alone yields persistent disks, but these disks are gravitationally unstable and fragment in the laminar runs; adding sonic turbulence removes the fragments.","The surviving disks have plasma $\\beta$ of order 1 to a few tens, with the poloidal field often dominating the toroidal field, so they are much more magnetized than typical protoplanetary-disk simulations assume.","Resolving the mismatch between these strongly magnetized young disks and the weaker fields used in later disk evolution models requires longer-term, higher-resolution simulations with more realistic thermodynamics and additional non-ideal MHD effects.","The magnetically induced pseudodisk remains the coherent backbone of the inner accretion flow even under sonic turbulence, so the disk-feeding pattern is warped but not destroyed.","The early disks induced by turbulence are initially very strongly magnetized, with plasma $\\beta$ below unity, which is why they cannot persist without ambipolar diffusion to weaken the field."],"supporting_citations":[{"why":"Supplies the sink-particle treatment that the paper modifies to follow the protostellar accretion phase.","marker":"Gong & Ostriker 2013"},{"why":"Provides the ambipolar-diffusion implementation in the MHD code used for the collapse simulations.","marker":"Chen & Ostriker 2014"},{"why":"Gives the analytic prediction of the AD-shock diffusion-DEMS that the laminar AD runs confirm in 3D.","marker":"Li & McKee 1996"},{"why":"Provides 2D axisymmetric AD simulations whose diffusion-DEMS is extended to and tested in 3D here.","marker":"Li et al. 2011"},{"why":"Showed the AD-induced structure becomes unstable in 3D, the comparison point for the interchange instability in these runs.","marker":"Krasnopolsky et al. 2012"},{"why":"Introduced the turbulence-warped pseudodisk picture and turbulent disk promotion that this paper extends to self-gravitating, sink-accreting cores.","marker":"Li et al. 2014b"},{"why":"Supplies the quantitative disk-cell selection criteria used to identify disks and measure their plasma beta.","marker":"Masson et al. 2016"},{"why":"Motivates the enhanced ambipolar-diffusivity models by showing small-grain depletion can raise the AD coefficient by one to two orders of magnitude.","marker":"Zhao et al. 2016"},{"why":"Identifies the ambipolar-diffusion timestep bottleneck in low-density, moderately magnetized cells that the paper's timestep floor addresses.","marker":"Mac Low et al. 1995"}],"fun_headline_variants":["Turbulence births disks, ambipolar diffusion keeps them","Sonic turbulence starts disks; diffusion ensures survival","Magnetic disks survive when turbulence and diffusion combine","Disk formation needs both warping turbulence and flux diffusion","From turbulent start to diffusive stability: disk recipe"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the numerical floor on the ambipolar-diffusion timestep, which artificially weakens diffusion in low-density, moderately magnetized cells where diffusion is fastest, does not alter the magnetic-flux redistribution that the paper credits with letting disks survive; if that floor is doing the work, the disk-persistence result is a numerical artifact.","fun_headline_variants_meta":{"raw":{"variants":["Turbulence births disks, ambipolar diffusion keeps them","Sonic turbulence starts disks; diffusion ensures survival","Magnetic disks survive when turbulence and diffusion combine","Disk formation needs both warping turbulence and flux diffusion","From turbulent start to diffusive stability: disk recipe"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000243,"raw_usage":{"total_tokens":1573,"prompt_tokens":1035,"completion_tokens":538,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":651,"completion_tokens_details":{"reasoning_tokens":463}},"tokens_in":651,"tokens_out":538,"duration_ms":5824,"temperature":1.0,"reasoning_tokens":463,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T10:06:48.712695+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the same core-collapse simulations with the ambipolar-diffusion timestep floor removed or weakened and check whether the persistent disks in the standard-diffusion models (M0.0AD1.0 and M1.0AD1.0) still form and survive; if they disappear, the disk-persistence claim rests on the numerical floor rather than on ambipolar diffusion itself.","supporting_citations":[],"review_version":1}