{"id":"4208ce42-c1e1-4733-8bbb-9c6b68873410","arxiv_id":"2505.23051","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Resistive MHD simulations of accretion onto a spinning black hole find a magnetically arrested disk for all resistivities tested, with a proposed average plasma-beta below one as the MAD indicator.","lead":"Astrophysicists ran computer simulations of magnetized gas swirling into a spinning black hole, adding electrical resistance to the plasma. They found the disk becomes a magnetically arrested state in every case and proposed a simple average magnetic-pressure measure to identify that state.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The β_ave ≤ 1 MAD criterion is an undefined, floor-sensitive volume average; no sensitivity test shows it is independent of the atmosphere floors or domain size.","rationale":"The reader's weakest_assumption correctly identifies the load-bearing risk. The phrase 'spatial average plasma beta across the computational domain' is never defined precisely, and the domain is filled with floor-dominated atmosphere, so the volume average is sensitive to numerical floor choices. The central claim (all resistive models are MAD and β_ave ≲ 1 signals this) rests on a single correlation between two time series in one family of simulations; no convergence study in floors or domain size is provided. The proposed test would settle the floor/domain dependence. If the test shows the threshold is stable, the claim would be substantially strengthened; if not, the abstract's β_ave ≲ 1 statement should be qualified as simulation-dependent. The MAD classification itself is supported by φ_acc values well above the nominal threshold in all 12 models, and the 2D/3D comparison is internally consistent, so the concern is specifically the new β_ave diagnostic. Given the paper's otherwise standard numerical methodology and use of a publicly available code, the reader's CONDITIONAL verdict remains appropriate; no change is needed.","tokens_in":17450,"tokens_out":6923,"duration_ms":69172,"concrete_test":"Recompute β_ave from the saved snapshots using three definitions: (i) volume-weighted over the full domain as currently used; (ii) mass-weighted over the full domain; (iii) volume-weighted over cells with ρ > 10 ρ_floor and r < 50 r_g. For each, record the time when β_ave first drops below 1 and compare with the time φ_acc first exceeds 50 in Fig. 2b. Also repeat one high-η and one low-η run with ρ_floor and P_floor each lowered by an additional factor of 10. If the β_ave ≤ 1 crossing time shifts by more than ~300 t_g, or if the β_ave value at the φ_acc = 50 crossing changes by more than a factor of 2, the criterion is floor/domain-dependent rather than a robust physical MAD indicator.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline diagnostic, β_ave ≲ 1 for MAD, is introduced in Section 3 via Fig. 2c, but no equation defines how β_ave is averaged over the computational domain. The domain is mostly tenuous atmosphere with floor values ρ_floor = 10^-6 ρ0 and P_floor = 10^-8 P0 (Section 2.6), so a volume average over the entire domain is dominated by cells whose pressure is at or near the floor. In the MAD state the funnel/jet regions have very strong fields and floor-level gas pressure, making β very small there; the volume average can therefore cross 1 because of the floor pressure choice, independent of the condition of the disk body. The paper presents only a single time-series correlation (Fig. 2c vs 2b) from one family of simulations, with no sensitivity test to floor values, domain size, or weighting scheme (volume vs mass vs disk-restricted). The final discussion acknowledges resolution limits for reconnection/plasmoid studies but does not mention this limitation. Thus the central claim that β_ave ≤ 1 is a physically motivated MAD indicator is not established; the correlation could be a numerical artifact of the atmosphere floors.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents global resistive MHD simulations of magnetized accretion flows around a spinning black hole, using the PLUTO code with an effective Kerr potential. The authors run two-dimensional axisymmetric and three-dimensional models with uniform resistivities from η=0.1 down to the ideal-MHD limit, and compare mass accretion rates, normalized horizon magnetic flux, MRI quality factors, stresses, variability, and jet power. They report that all models reach the magnetically arrested disk (MAD) state according to the standard horizon-flux criterion φ_acc ≳ 50, and they propose that a spatial average plasma beta β_ave ≲ 1 over the computational domain is an alternative indicator of the MAD state. Additional results include reduced MRI turbulence and jet power at high resistivity, plasmoid formation at low resistivity, and no clear resistivity-variability correlation.","tokens_in":17675,"tokens_out":10556,"duration_ms":107481,"significance":"If substantiated, the β_ave ≲ 1 criterion would be an inexpensive diagnostic for classifying MAD versus non-MAD states in global simulations, and the resistivity-dependent trends (turbulence suppression, jet-power reduction, plasmoid formation) are timely for interpreting EHT and GRAVITY observations. The paper also provides a useful 2D versus 3D comparison at higher effective resolution than typical GRMHD runs. The main weakness is that the proposed β_ave diagnostic is not defined operationally and lacks sensitivity tests; as presented, the central claim is not sufficiently supported.","major_comments":[{"comment":"The manuscript never defines how β_ave is computed. The text says \"spatial average plasma-beta parameter across the entire computational domain,\" but no equation or weighting is given (e.g., volume-weighted arithmetic mean of β, ratio of volume-averaged pressures, or mass-weighted average). Because the domain is largely filled by the low-density atmosphere at the numerical floors ρ_floor=10^-6 ρ0 and P_floor=10^-8 P0 (Section 2.6), a whole-domain average can be sensitive to the floor values; the paper presents no sensitivity test to floors, domain size, or weighting scheme. The conclusion that β_ave ≲ 1 signals the MAD state is therefore not established and could be an artifact of the atmosphere treatment.","section":"Section 3, Fig. 2c"},{"comment":"The threshold β_ave=1 is read off from the same set of simulations used to classify the MAD state via φ_acc≥50. This is a post-hoc correlation rather than an independent test. To support the claim, the authors should validate the threshold on a held-out set of simulations or on subregions of the same runs (e.g., excluding the atmosphere), and show that the crossing time of β_ave=1 tracks φ_acc=50 under different floor settings.","section":"Section 3, Fig. 2c vs 2b"},{"comment":"The MRI quality factors and the Maxwell/Reynolds stress profiles are computed at a single time t=8500 tg and, for the 3D models, at a single azimuthal slice φ=0. In a turbulent flow these diagnostics fluctuate strongly in time and azimuth; the reported factors-of-1000 differences between 2D and 3D models near the black hole are not robust without time/azimuth averaging. This does not affect the time-series conclusions from Fig. 2, but it weakens the quantitative turbulence comparison.","section":"Section 3, Figs. 3 and 4"},{"comment":"The normalized flux φ_acc is defined in Gaussian units but the y-axis of Fig. 2b is labeled \"code unit.\" The conversion between the code's magnetic field units (where P_mag=B²/2) and the Gaussian-unit threshold φ_acc=50 is not stated. Without this conversion the reader cannot verify whether the plotted curves actually cross the MAD threshold; please clarify the units and conversion factors.","section":"Section 2.7 and Fig. 2b"}],"minor_comments":[{"comment":"The atmosphere profile ρ_atm = ρ_floor r^{-3/2}, P_atm = P_floor r^{-5/2} appears to use the same floor constants as the numerical floors in Section 2.6; for r>1 this puts the initial atmosphere below the floor, so the floor will override it. Please clarify how the atmosphere and the floor are meant to interact.","section":"Section 2.3"},{"comment":"The computational domain extent (r_min, r_max, z_max) is not stated; only the inner boundary and resolution are given. The domain size is needed to interpret the phrase \"entire computational domain\" and the volume averages.","section":"Section 2.5"},{"comment":"The discussion of limitations focuses on resolution for reconnection and plasmoids, but does not mention the sensitivity of β_ave to numerical floors; this omission should be addressed in the revision.","section":"Section 4"},{"comment":"The reference \"SÄ dowski\" should be \"Sądowski\".","section":"References"},{"comment":"The color maps and line styles for resistivity values η=10^-4 and η=10^-5 are hard to distinguish; consider labeling each panel with the η value more prominently and using a perceptually uniform colormap.","section":"Figure 5 and Figure 8"}],"recommendation":"major_revision","confidential_remarks":"The paper's main contribution as presented is the β_ave criterion, but the current manuscript does not support it with an operational definition or sensitivity analysis. The resistive 2D/3D comparison is a useful contribution and the MAD classification via φ_acc is plausible. The requested revisions (define the average, add sensitivity tests, time/azimuth-average the turbulence diagnostics) are feasible within the scope of the manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick overview: this is a reasonable simulation parameter study, and the central result that all highly magnetized models reach the MAD state across resistivity eta = 0.1 to ~0 is well supported by the normalized horizon flux crossing the standard threshold (Fig. 2b). That part holds up. The genuinely new element is the beta_ave <= 1 MAD indicator, and this is the weak spot. The text in Section 3 says 'spatial average plasma-beta across the entire computational domain' but never defines the average. Given that most of the domain is a tenuous atmosphere sitting at the floors (rho_floor = 1e-6, P_floor = 1e-8, Section 2.6), a volume-weighted average can easily be controlled by cells with floor pressure, especially in the funnel where beta is tiny. The paper shows only a time-series correlation between beta_ave dropping below 1 and the horizon flux crossing 50, with no sensitivity test to floors, domain size, or weighting. The stress-test concern lands; I went back to the text and the definition is simply absent. This can be fixed, but as written the headline diagnostic is not established.\n\nWhat the paper does well: the 2D/3D comparison at matched resolution is useful and the resistive sweep is clean. The plasmoid identification in the low-resistivity 2D runs is plausible and the authors are honest that the 3D resolution is insufficient to resolve plasmoids. The null result on variability-resistivity aligns with Nathanail et al. 2024. The structure and citation pattern look fine; the prior Ripperda and Vourellis work is credited.\n\nSoft spots, in proportion: the jet power claim in Section 3.2 is not a direct measurement; it is the BZ formula with kappa = 0.1 applied to horizon flux, so saying 'low-resistivity models produce higher power jets' is essentially restating that they have more horizon flux. Also, the effective Kerr potential is pseudo-Newtonian, so the quantitative jet and MAD statements should be read as semi-relativistic. There is no convergence study, which is normal for global runs at this scale but matters more here because the beta_ave claim depends on the atmosphere.\n\nWho it is for: people running global resistive MHD simulations, and anyone wanting a cheap MAD diagnostic. With a proper definition and sensitivity runs, the beta_ave criterion could be useful. Right now it is a post-hoc threshold on an undefined volume average.\n\nRecommendation: send to peer review. A good referee can request the definition, a floor-sensitivity test, and a domain-restricted average, and the authors' core MAD result will likely survive the revision. It should not be desk-rejected.","headline":"A competent resistive-MHD parameter study whose robust MAD core is undercut by an undefined, floor-sensitive beta_ave diagnostic presented as the main new result.","tokens_in":18247,"tokens_out":2700,"would_cite":false,"duration_ms":28570,"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":"Resistive accretion flows around spinning black holes all settle into the magnetically arrested state, and a domain-averaged plasma beta at or below one marks the transition.","keywords":["resistive magnetohydrodynamics","magnetically arrested disk","plasma beta","black hole accretion","MRI turbulence","plasmoid formation","jet power","active galactic nuclei"],"falsifier":"Rerun one low- and one high-resistivity model with the density and pressure floors lowered by at least two orders of magnitude and with $\\beta_{\\rm ave}$ recomputed both over the full domain and over the disk body only; if the horizon flux still exceeds the MAD threshold while the disk-restricted $\\beta_{\\rm ave}$ stays above one, the proposed criterion is an artifact of the floor-dominated averaging volume.","tokens_in":17228,"feed_emoji":"🕳️","tokens_out":6189,"duration_ms":58685,"temperature":0.7,"pith_summary":"This paper asks whether adding explicit, uniform resistivity to a highly magnetized accretion flow around a spinning black hole changes the flow's magnetic state. Using two- and three-dimensional resistive magnetohydrodynamic simulations, it finds that every model, from nearly ideal to \\eta = 0.1, ends up in the magnetically arrested disk (MAD) state, in which magnetic pressure near the horizon holds back the accreting gas. It then proposes a cheaper diagnostic than horizon flux for identifying that state: the volume-averaged plasma $\\beta$ over the whole computational domain, with MAD onset at $\\beta_{\\rm ave} \\lesssim 1$. The authors also report that higher resistivity suppresses magnetorotational-instability turbulence, that plasmoids appear only at low resistivity, that variability is not tied to resistivity, and that low-resistivity models produce the strongest jets. If the $\\beta$-averaging criterion holds, any simulation with access to a domain-wide pressure ratio could identify MAD accretion without resolving the horizon flux.","feed_headline":"Beta below one flags the magnetically arrested disk state","feed_subtitle":"Volume-averaged plasma beta at or below one marks MAD in 2D and 3D resistive black hole accretion runs.","key_machinery":"The load-bearing object is the pair of MAD indicators: the normalized magnetic flux threading the inner boundary, $\\dot{\\phi}_{\\rm acc}$, previously used to define the MAD threshold, and the newly proposed volume-averaged plasma $\\beta$ $\\beta_{\\rm ave}$ computed over the whole computational domain. The paper's argument runs on the correspondence between these two: whenever $\\dot{\\phi}_{\\rm acc}$ rises above the canonical value of about 50, $\\beta_{\\rm ave}$ falls to or below unity, so the averaged pressure ratio serves as a proxy for the horizon-flux criterion. The simulations are driven by a resistive MHD code with an effective Kerr potential, uniform resistivity, and a fixed torus set to MAD-like dimensions; the MRI quality factor, Maxwell and Reynolds stresses, and current density maps carry the secondary claims about turbulence and reconnection.","core_discovery":"The central claim is that resistivity does not destroy the magnetically arrested state: all resistive models considered, with uniform resistivity from about 0 to 0.1, reach the MAD state as measured by the normalized horizon magnetic flux $\\dot{\\phi}_{\\rm acc}$ crossing the canonical threshold of about 50 in Gaussian units. The paper's new proposal is that this state can be identified from the spatial average of the plasma $\\beta$, $\\beta_{\\rm ave}$, computed over the entire computational domain; the flow enters MAD when $\\beta_{\\rm ave} \\lesssim 1$, meaning magnetic pressure is comparable to or larger than gas pressure on average. The same simulations show that mass accretion rates are nearly equal in 2D and 3D until $t \\approx 1000\\,t_g$ and then diverge as non-axisymmetric MRI turbulence dominates in 3D, that high resistivity ($\\eta = 0.1, 0.01$) damps MRI turbulence, that low-resistivity runs show plasmoids and current sheets in the jet region, that variability of accretion rate and magnetic flux shows no clear trend with resistivity, and that jet power is roughly two orders of magnitude lower at $\\eta = 0.1$ than at $\\eta \\leq 10^{-3}$.","pith_inferences":["The $\\beta_{\\rm ave} \\lesssim 1$ threshold would be on firmer ground if recomputed as a mass-weighted or disk-restricted average; the current volume average includes floor-dominated atmosphere, and the authors do not report such a test.","If the threshold survives, it suggests a purely local pressure-balance criterion for MAD that could be applied to observed accretion flows via inferred magnetic and gas pressures, not just to simulations.","The absence of a variability-resistivity trend hints that observed timing variability in sources like Sgr A* is set by something other than the effective magnetic diffusivity, for instance the intermittent flux-eruption cycle common to MAD states.","The plasmoid formation at low resistivity in jets suggests that reconnection-powered flares should be more prominent in low-diffusivity sources; correlating flaring activity with jet power in AGN samples would test this."],"forward_implications":["A volume-averaged plasma beta at or below unity can be used as a practical MAD indicator in global simulations, without computing horizon-threading flux.","Resistivity by itself does not select the magnetic state; even strongly diffusive flows accumulate enough flux to become MAD, so comparisons of MAD versus non-MAD flows should not be attributed to resistivity.","In any model with high resistivity, jet power is expected to be about two orders of magnitude weaker, because less magnetic flux accumulates at the horizon.","Simulations aiming at late-time accretion behavior must be three-dimensional after $t \\approx 1000\\,t_g$, when non-axisymmetric effects change the accretion rate and flux.","Low-resistivity flows are the places to look for reconnection-driven plasmoid formation in jets, though the resolution here is insufficient to confirm the plasmoid dynamics."],"supporting_citations":[{"why":"Defines the magnetically arrested disk state as the balance of accumulated magnetic pressure against the accretion ram pressure.","marker":"R. Narayan et al. 2003"},{"why":"Supplies the canonical MAD threshold $\\dot{\\phi}_{\\rm acc} \\gtrsim 50$ and the scaling of jet power with horizon flux.","marker":"A. Tchekhovskoy et al. 2011"},{"why":"Provides the PLUTO code with its resistive MHD module and numerical schemes used for all simulations.","marker":"A. Mignone et al. 2007"},{"why":"Provides the effective Kerr potential that replaces full general relativity in the simulations.","marker":"I. K. Dihingia et al. 2018"},{"why":"Establishes the resistive GRMHD framework for reconnection and plasmoid formation that this paper builds on.","marker":"B. Ripperda et al. 2019"},{"why":"Shows how plasmoids form and are advected in resistive current sheets, the pattern the paper sees at low resistivity.","marker":"B. Ripperda et al. 2020"},{"why":"Reports that resistivity has minimal effect on accretion-flow variability, the comparison claimed here.","marker":"A. Nathanail et al. 2024"},{"why":"Fixes the MAD torus dimensions, inner edge and pressure maximum, used for the initial equilibrium torus.","marker":"G. N. Wong et al. 2021"},{"why":"Supports the choice of torus size consistent with the MAD state in GRMHD simulations.","marker":"C. M. Fromm et al. 2022"}],"fun_headline_variants":["Resistivity doesn't prevent MAD state in black hole flows","Plasma beta below one marks MAD in resistive accretion","High resistivity damps MRI turbulence but not MAD state","Jet power drops sharply with higher resistivity in AGN flows","MAD state identified by average plasma beta in simulations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The criterion rests on the assumption that averaging the plasma beta over the whole simulation box, including the very low-density, low-pressure atmosphere used to fill empty space, measures the disk's true magnetization rather than being dominated by the artificial floor values.","fun_headline_variants_meta":{"raw":{"variants":["Resistivity doesn't prevent MAD state in black hole flows","Plasma beta below one marks MAD in resistive accretion","High resistivity damps MRI turbulence but not MAD state","Jet power drops sharply with higher resistivity in AGN flows","MAD state identified by average plasma beta in simulations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000273,"raw_usage":{"total_tokens":1726,"prompt_tokens":1125,"completion_tokens":601,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":741,"completion_tokens_details":{"reasoning_tokens":522}},"tokens_in":741,"tokens_out":601,"duration_ms":6570,"temperature":1.0,"reasoning_tokens":522,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:55:24.711647+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Rerun one low- and one high-resistivity model with the density and pressure floors lowered by at least two orders of magnitude and with $\\beta_{\\rm ave}$ recomputed both over the full domain and over the disk body only; if the horizon flux still exceeds the MAD threshold while the disk-restricted $\\beta_{\\rm ave}$ stays above one, the proposed criterion is an artifact of the floor-dominated averaging volume.","supporting_citations":[],"review_version":1}