{"id":"cbace833-404e-4ea1-809a-ee8aa0a9f5f8","arxiv_id":"2506.09150","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Stellar bars can convert up to 50% of a classical bulge's stars into bar-supporting disk-like orbits, potentially explaining the scarcity of classical bulges.","lead":"Using N-body simulations, this paper finds that stellar bars can trap up to half of a galaxy's classical bulge stars into disk-like resonant orbits, rendering the bulge nearly indistinguishable from the disk. Comparing with 210 MaNGA galaxies, the authors find that slower bars are associated with weaker bulges, suggesting bars erode classical bulges over cosmic time.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 50% attrition ceiling is unproven because the simulated 'classical bulges' are extremely diffuse (half-mass radius ~7.6 kpc vs disk scale length 2.7 kpc), so the h2:1 trapping may be an artifact of an unrealistically extended bulge.","rationale":"The reader's CONDITIONAL verdict remains appropriate: the simulation's core result is reproducible across four bulge fractions and survives a reinitialized-bulge control, and the MaNGA correlation points in the expected direction. However, I find a sharper, more internal weakness than the gas-free simplification emphasized by the reader: the initial conditions for the 'classical bulge' are unrealistically extended. With Hernquist scale length a=3.15 kpc, the half-mass radius is ~7.6 kpc, three times the disk scale length. Thus most of the 'bulge' mass lives at radii where the bar's ILR sweeps through the disk. The 50% h2:1 trapping rate and the low remaining P(Bulge) fraction could both be consequences of this radial overlap rather than of genuine secular attrition of a compact classical bulge. This is not a disagreement with consensus; it is a missing axis in the model parameter space that is directly load-bearing for the paper's headline quantitative conclusion. The proposed compact-bulge rerun is a clean, feasible check: if the trapping fraction drops substantially, the 'at most half' statement must be qualified or withdrawn. Since the reader already conditions acceptance on additional robustness tests, my recommendation does not change the verdict, but it specifies a concrete test that should be part of the condition.","tokens_in":22659,"tokens_out":10779,"duration_ms":121632,"concrete_test":"Rerun the 8% and 16% bulge models with identical total masses, disk, and halo parameters, but replace the Hernquist bulge (a=3.15 kpc) with a compact classical bulge, e.g. Hernquist a=0.5 kpc or a Sersic n=4 profile with effective radius ~1 kpc. Measure the h2:1 trapped fraction and the pseudo-observational P(Bulge)>0.5 fraction at t=10 Gyr. If the trapped fraction is substantially below the ~50% found with a=3.15 kpc (e.g. <30%), the claim that an observed classical bulge in a barred galaxy retains at most half its initial population is not supported. Also check whether the residual kinematically distinguishable bulge fraction rises above the ~12% reported; if so, the paper's 'nearly all erased' inference is weakened.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The quantitative central claim—that any observed classical bulge in a barred galaxy may retain at most half of its initial population—rests on four N-body models whose 'classical bulge' initial conditions are extremely diffuse. The bulge is a Hernquist sphere with scale length a=3.15 kpc, implying a half-mass radius of roughly 7.6 kpc, i.e. nearly three times the disk's exponential scale length of 2.7 kpc. A bulge with half its mass outside ~7.6 kpc is not a central classical bulge; it is a stellar-halo-like spheroid whose outer parts occupy exactly the disk radii through which the slowing bar's inner Lindblad resonance sweeps. The observed rise to ~50% h2:1-trapped stars may therefore be dominated by these disk-overlapping orbits rather than by a generic transformation of a compact, dispersion-supported classical bulge. The paper does not vary bulge concentration, so the 'at most half' conclusion is a one-point extrapolation. The pseudo-observational result that only ~11-14% of the initial bulge remains kinematically distinguishable is likewise coupled to this choice, because a diffuse initial bulge will naturally resemble the disk in the KDE-based classifier. A compact, realistic classical bulge could be far less affected, which would invalidate the stated upper bound. The reinitialized-bulge control is a good check, but it reinitializes the same diffuse bulge and therefore does not address this concentration assumption.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper uses a suite of four isolated, gas-free N-body disk galaxy simulations with live dark matter halos and bulge-to-disk mass fractions of 0%, 4%, 8%, and 16% to quantify the secular interaction between a growing stellar bar and a classical bulge. Using orbital frequency analysis, the authors find that up to 50% of the stars initially in the classical bulge become trapped in the h2:1 (inner Lindblad resonance) family over 7-9 Gyr, and that this fraction is reproduced when the bulge is reinitialized to isotropic initial conditions at the time the bar becomes strong. A Bayesian pseudo-observational classifier applied in (r, z, v_z, v_phi) phase space implies that only 11-14% of the initial bulge remains kinematically distinguishable from the disk at the end of the runs. In a sample of 210 MaNGA barred galaxies, slow bars show lower bulge prominence than fast bars (Anderson-Darling >3.2 sigma for the pattern-speed split, 2.9 sigma for the R split). The authors conclude that bars erode classical bulges and that any observed classical bulge in a barred galaxy may contain at most half of its primordial population fraction.","tokens_in":22939,"tokens_out":17545,"duration_ms":165027,"significance":"If the 50% trapping fraction is robust, the paper identifies a genuinely important secular channel: classical bulges are not passively long-lived tracers, and the observed scarcity of classical bulges (including in the Milky Way) could be at least partly a consequence of bar-driven kinematic transformation rather than of formation history alone. The main strengths are that the headline fraction emerges from forward N-body evolution rather than from a fitted model, that the controlled reinitialization experiment addresses the dependence on bar-formation history, and that the authors confront their simulations with an independent MaNGA sample. However, the headline 'at most half' inference rests on a single, unusually diffuse bulge profile and on an observational correlation that has a plausible alternative causal reading; both limit the strength of the claim as currently stated. The paper is therefore a solid, well-posed dynamical study whose main generalization needs additional robustness work.","major_comments":[{"comment":"Section 2.1 specifies the classical bulge as a Hernquist sphere with scale length a = 3.15 kpc, which implies a three-dimensional half-mass radius of (1+sqrt(2))a approximately 7.6 kpc, about 2.8 times the disk scale length of 2.7 kpc. Most of the 'bulge' mass therefore sits at radii that overlap the disk, precisely the region through which the inner Lindblad resonance sweeps as the bar slows. The headline claim of Section 5, that any observed classical bulge in a barred galaxy may contain at most half of its initial population, rests on this single, unusually diffuse bulge realization; the paper does not vary bulge concentration. A compact classical bulge typical of observations (effective radius near 1 kpc) could trap a very different fraction, and the pseudo-observational confusion of Section 3.3 is also amplified by this choice. Please add at least one compact-bulge run (for example a = 0.5-1 kpc) or explicitly restrict the claim to diffuse bulges.","section":"2.1, 3.1, 5"},{"comment":"The observational comparison in Section 3.2 does not uniquely support the attrition scenario, because the paper's own models contain the reverse causal chain: the 16% Bulge model reaches the slow-bar regime later than models with smaller bulges (Figure 2, Section 3.1). If massive bulges delay bar slowing, then the MaNGA trend of lower bulge prominence among slow bars (Anderson-Darling >3.2 sigma for the Omega_bar split, 2.9 sigma for the R split) is expected even if bars never erode bulges. The interpretation in Section 4 that slow bars are older and therefore have had more time to erode their bulges requires controlling for this selection effect; as written, the statement in Section 5 that the observations support the key findings overstates the constraining power of the comparison. A concrete test would be to check whether the trend survives a match on stellar mass and bulge-to-total ratio, or to predict a bar-aligned rotating component in classical-bulge kinematics that the size correlation alone would not produce.","section":"3.2, Fig. 2"},{"comment":"Section 3.3 reports that at the end of the runs only 11%, 12%, and 14% of the initial bulge is kinematically distinguishable from the disk for the 4%, 8%, and 16% Bulge models, and the reader is not told what fraction would already be distinguishable at t = 0 under the same classifier. The nearly flat initial P(Bulge) distribution shown in the left panel of Figure 7 suggests a non-trivial baseline confusion that is built into the diffuse bulge and the 4D KDE rather than caused by the bar; the bar-driven confusion should be quoted as the change relative to that baseline. In addition, the conclusion in Section 5 that approximately 50% of each initial bulge is classified through its kinematics as consistent with the stellar disk populations is not what Section 3.3 reports for the joint criterion of h2:1 and P(Disk) > 0.5: the end-of-run fractions are 22%, 41%, and 31% for the 4%, 8%, and 16% models, respectively. The conclusion should report the joint numbers or clarify that the 50% refers to h2:1 trapping alone.","section":"2.4, 3.3"},{"comment":"The limitation statement in Section 5, that gas, star formation, and external interactions are not included and may alter or suppress this transformation process, is an honest acknowledgment, but it directly bounds the central inference as it applies to real galaxies. The 'at most half' claim should be framed as conditional on isolated, gas-free secular evolution, and the MaNGA correlation should be described as indirect support rather than as verification of the mechanism.","section":"5"}],"minor_comments":[{"comment":"In Section 2.3, Omega_x is described as the azimuthal frequency measured with respect to the bar frame, but the resonance condition is written as 2 Omega_r = Omega_x; please define Omega_x explicitly as Omega_phi - Omega_bar so that the ILR condition is unambiguous.","section":"2.3"},{"comment":"The bulge-prominence weights in Equation (3) are adopted without a sensitivity test; please show that the fast/slow comparisons in Section 3.2 are insensitive to the exact values or cite a validation of the scheme.","section":"Eq. (3)"},{"comment":"The binned gradient in Figure 6 is comparable in size to the quoted per-bin uncertainties; please add a rank-correlation statistic between bulge prominence and each bar-speed metric, with its significance, rather than relying only on the two-way Anderson-Darling splits.","section":"3.2, Fig. 6"},{"comment":"Please report the initial-condition fraction of bulge particles already classified as P(Disk) > 0.5 by the pseudo-observational classifier, so that the bar-driven confusion in Section 3.3 can be separated from the baseline confusion apparent in the nearly flat initial P(Bulge) histogram of Figure 7.","section":"3.3, Fig. 7"},{"comment":"Please correct the typographical issues: 'the the circular velocity' in Section 2.2, 'is varies the least' in the Figure 3 caption, 'furest extent' in the Figure 7 caption, and the axis label 'nbar rotation' in Figure 3.","section":"2.2, Fig. 3, Fig. 7"},{"comment":"The bulge-prominence definition relies on 'Garland et al. (in prep)'; please provide a stable reference or reproduce the vote-fraction weights and thresholds so that the analysis is fully checkable.","section":"2.5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well within the scope of ApJ and the simulation analysis is careful. My main reservation is the gap between the simulated bulge model (a very diffuse Hernquist spheroid with a = 3.15 kpc) and the classical bulge of the headline claim; a compact-bulge companion run would make the 50% result much more convincing. I also recommend that the authors soften the causal interpretation of the MaNGA correlation, since their own models predict that massive bulges delay bar slowing. The relationship to the companion paper (McClure et al. 2025) is acknowledged, but the editor may wish to confirm that the novel analysis here is sufficiently distinct in scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [colleague],\n\nThe headline result here—up to 50% of a classical bulge's stars getting trapped in the bar's h2:1 resonance, leaving only ~11–14% kinematically distinguishable—is genuinely new and the reinitialized-bulge control is a good check. I think the paper is worth taking seriously. But there is a soft spot that matters: the simulated classical bulges are extremely diffuse. A Hernquist bulge with scale length 3.15 kpc has a half-mass radius of about 7.6 kpc, nearly three times the disk's 2.7 kpc scale length. That is not a central classical bulge; it is a halo-like spheroid whose outer parts sit exactly where the slowing bar's ILR sweeps. So the 50% trapping number may be inflated by disk-overlapping orbits, and the paper does not vary bulge concentration. The reinitialized control reuses the same diffuse bulge, so it doesn't answer this. The 'at most half' conclusion is a one-point extrapolation.\n\nWhat the paper does well: the 50% fraction is reproducible across four bulge masses; the frequency analysis is checked against bar amplitude growth; and the MaNGA comparison (210 galaxies) shows the same direction in both Omega_bar and R, even if the significance is only 2.9–3.2 sigma. The authors also explicitly acknowledge the gas-free, isolated-model limitation in Section 5. Self-citation to their prior work is appropriate here since this builds directly on it.\n\nThe other soft spots are minor in comparison: the observational correlation is supporting, not a direct test of the mechanism, and the data/code are only available on request.\n\nMy bottom line: this deserves a serious referee, but the referee should ask for at least one compact-bulge model to see whether the 50% ceiling holds when the bulge is concentrated. For readers working on bar-bulge coevolution, it's useful even as an upper bound with caveats. I'd bring it up in reading group, and I'd cite it, but with the concentration caveat attached.","headline":"The 50% trapping number is new and the control is good, but the simulated bulges are too diffuse to support the 'at most half' conclusion yet.","tokens_in":23513,"tokens_out":3207,"would_cite":true,"duration_ms":29152,"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":"Stellar bars trap up to half of a classical bulge into disk-like orbits.","keywords":["galaxy bars","classical bulges","secular evolution","resonant trapping","N-body simulations","orbital dynamics","MaNGA galaxies","bulge attrition"],"falsifier":"Find a sample of old, slow-bar galaxies in the same mass range as the models whose bulges are still large and dispersion-dominated; if a substantial population of such galaxies exists, the claim that slow bars erode classical bulges to at most half their initial mass would fail. Alternatively, run one of the models with a live gas component and star formation: if the trapped h2:1 fraction drops well below 50%, the secular-attrition mechanism is not robust under realistic conditions.","tokens_in":1635,"feed_emoji":"🌌","tokens_out":1677,"duration_ms":66398,"temperature":0.7,"pith_summary":"The paper argues that a long-lived stellar bar is not a passive companion to a classical bulge: it actively erodes the bulge by trapping its stars into bar-supporting resonances, so that an observed bulge in an evolved barred galaxy may contain at most half of the stars the galaxy originally built. In isolated N-body simulations with live dark halos, about 50% of the initial bulge population ends up on 2:1 inner Lindblad resonance orbits with disk-like kinematics, making much of the classical bulge observationally indistinguishable from the disk. The authors test this against 210 MaNGA barred galaxies and find that slower bars, which are older, are preferentially paired with weaker bulges. This offers a secular explanation for the long-noted scarcity of classical bulges in local disk galaxies and for the Milky Way's small classical bulge fraction, without requiring that those bulges never formed.","feed_headline":"Stellar bars trap half of classical bulges into disk orbits","feed_subtitle":"Simulations show old slow bars leave only a remnant bulge; MaNGA galaxies confirm the trend.","key_machinery":"The load-bearing object is the h2:1 inner Lindblad resonance of the stellar bar, identified by orbital-frequency analysis with the naif code: stars satisfying $2\\Omega_r = \\Omega_\\phi$ in the bar frame are the bar-supporting population. The paper uses this frequency selection to track what fraction of the initial classical bulge becomes trapped, and a Bayesian kinematic classifier on ($r$, $z$, $v_z$, $v_\\phi$) to ask how often those trapped bulge stars would be mistaken for disk stars. The reinitialization experiment, where bulge particles are replaced by their original isotropic counterparts at the strong-bar epoch, shows that trapping proceeds independently of the initial bar-formation episode.","core_discovery":"The central claim is that classical bulges are not long-lived relics of hierarchical assembly once a stellar bar forms: the evolving bar transfers angular momentum to the bulge and resonantly traps up to half of its stars on h2:1 orbits (two radial oscillations per azimuthal period), converting an initially isotropic, dispersion-supported population into a rotating, disk-like one. By the time a bar becomes slow, only about 12-14% of the initial bulge remains kinematically classifiable as bulge in the simulations, and the bar-trapped bulge stars are concentrated in the central bar and pseudobulge region. The paper further claims that in 210 MaNGA galaxies, slow bars are associated with smaller bulges, consistent with bulge attrition over long secular timescales.","pith_inferences":["Extension: if gas is included, the trapped fraction may differ; gas can both weaken the bar and suppress trapping while also funneling material to the center to mimic bulge growth, so the MaNGA trend could be partly environmental rather than purely secular.","Extension: chemical tagging in the Milky Way's bar region should reveal a population of old bulge-origin stars on h2:1 orbits that kinematics alone would call disk; measuring that fraction would test the 50% number directly in our own Galaxy.","Extension: bars forming at high redshift should already have eroded bulges by intermediate redshift, so a deficit of dispersion-dominated bulges in high-redshift barred galaxies may be visible in deep kinematic surveys.","Extension: the same resonant mechanism should also spin up part of the stellar halo, producing a flattened, slowly rotating halo component that could be searched for in existing halo surveys."],"forward_implications":["Any kinematically measured classical bulge in a barred galaxy with an old, slow bar should be read as a lower limit: the original bulge could be up to twice as massive as the observed component.","Kinematic-only decompositions of barred galaxies will systematically undercount classical bulges, so bulge-to-total ratios from surveys need chemical or photometric information to recover the true bulge mass.","Bulge attrition acts on short timescales: within about 3 Gyr of strong-bar formation, a third of the bulge stars are already trapped, so even moderately old bars can noticeably erode bulges.","The Milky Way's small classical bulge fraction is consistent with its long, slow bar, independent of assumptions about its merger history.","Observational samples should show a monotonic trend in which the slower, older the bar, the weaker the classical bulge; the MaNGA sample shows this at more than 3 sigma when bars are divided by pattern speed."],"supporting_citations":[{"why":"Provides the same simulation suite and the orbital-family analysis on which this paper builds.","marker":"McClure et al. (2025)"},{"why":"Establishes that angular momentum exchange at the -1:1 Lagrange and 2:1 inner Lindblad resonances drives the bulge spin-up.","marker":"Saha et al. (2012)"},{"why":"Defines fast and slow bar regimes through the corotation-radius ratio and the bar braking process used to classify bar ages.","marker":"Debattista & Sellwood (2000)"},{"why":"Supplies the 210 MaNGA barred galaxies with measured pattern speeds, corotation radii, and R values used for the observational comparison.","marker":"Géron et al. (2023)"},{"why":"Provides the galaxy initial-condition generator used to build the disk, halo, and bulge models.","marker":"Springel et al. (2005)"},{"why":"Supplies the density profile used for both the dark matter halo and the classical bulge.","marker":"Hernquist (1990)"},{"why":"Provides the naif frequency-analysis code used to identify the resonant orbits.","marker":"Beraldo e Silva et al. (2023)"},{"why":"Gives the integral-field pattern-speed method applied to the MaNGA observations.","marker":"Tremaine & Weinberg (1984)"},{"why":"Establishes the angular momentum exchange between the bar and surrounding stellar and dark components that underlies bar slowing.","marker":"Athanassoula (2003)"}],"fun_headline_variants":["Bars trap half of classical bulge stars onto disk orbits","Slow bars leave only remnant bulges in disk galaxies","Stellar bars convert up to 50% of bulge stars into disk","Bar resonances reshape classical bulges into disk stars","Old slow bars correlate with weaker bulges in MaNGA sample"],"cache_read_input_tokens":25600,"weakest_assumption_plain":"The load-bearing premise is that isolated, gas-free N-body models capture the dominant secular mechanism acting on bulges; if gas inflows, ongoing star formation, or mergers suppress or reverse the resonant trapping, the inference that observed classical-bulge scarcity reflects bar-driven attrition weakens, and the MaNGA correlation is only supporting evidence, not a direct test of the mechanism.","fun_headline_variants_meta":{"raw":{"variants":["Bars trap half of classical bulge stars onto disk orbits","Slow bars leave only remnant bulges in disk galaxies","Stellar bars convert up to 50% of bulge stars into disk","Bar resonances reshape classical bulges into disk stars","Old slow bars correlate with weaker bulges in MaNGA sample"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000214,"raw_usage":{"total_tokens":1390,"prompt_tokens":872,"completion_tokens":518,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":488,"completion_tokens_details":{"reasoning_tokens":434}},"tokens_in":488,"tokens_out":518,"duration_ms":6521,"temperature":1.0,"reasoning_tokens":434,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T04:55:54.420180+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a sample of old, slow-bar galaxies in the same mass range as the models whose bulges are still large and dispersion-dominated; if a substantial population of such galaxies exists, the claim that slow bars erode classical bulges to at most half their initial mass would fail. Alternatively, run one of the models with a live gas component and star formation: if the trapped h2:1 fraction drops well below 50%, the secular-attrition mechanism is not robust under realistic conditions.","supporting_citations":[],"review_version":1}