{"id":"60416cfe-9b27-4f79-9dfa-286e0cb3a4b7","arxiv_id":"2411.18182","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"A model-dependent orbit superposition reconstruction of APOGEE bulge stars finds the bulge is a 4:3 mix of thick and thin disc populations and is metal-rich relative to the surrounding disc when viewed in 3D.","lead":"This paper applies an orbit superposition method to tens of thousands of giant stars from the APOGEE survey to reconstruct a full three-dimensional, chemokinematic map of the Milky Way bulge, including regions heavily obscured by dust. It argues that the bulge is a mix of old thick-disc and younger thin-disc stars with a 4:3 mass ratio, and that it is actually metal-rich once stars far from the plane are included.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The X-shape and orbital decomposition are inherited from the adopted Sormani et al. (2022) potential, and the mass ratio and metallicity maps are conditional on that potential and on orbit-painted abundances; a potential-replacement stress test is needed.","rationale":"The reader correctly identifies the adopted potential as the weakest assumption. I agree because the density-fitting step makes spatial morphology an input, not an output; the only parts of the analysis that are genuinely data-driven are the abundance-to-orbit assignments, which additionally need phase mixing. The paper has real independent support: the raw and reconstructed MDFs resemble each other, the method was tested on mock data in Paper I, and the authors are explicit about the equilibrium caveat. Those supports do not, however, establish robustness to potential choice. This is a conditional-accept-level concern, not a rejection: it is addressable by systematic variation of the potential, which the paper does not provide. No internal inconsistency was found in the central argument.","tokens_in":34452,"tokens_out":5756,"duration_ms":59259,"concrete_test":"Recompute the full orbit-superposition reconstruction inside R<3.5 kpc using an alternative potential that still matches published constraints but differs in bar parameters, e.g., the Portail et al. (2017) model, and with the Sormani et al. (2022) potential at pattern speeds 30 and 44 km/s/kpc, keeping all APOGEE/Gaia inputs and the fitting procedure unchanged. Then compare the recovered high-/low-alpha mass ratio, the five GMM component centers and weights, and the face-on metallicity maps. If the 4:3 ratio or component centers move by more than the quoted statistical precision (e.g., ratio by >0.1 or [Fe/H] centers by >0.05 dex), the headline claims are materially potential-dependent and the paper should report systematic error bars rather than single numbers.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 2.2 states that orbit weights are fitted to the 3D stellar density of the analytic Sormani et al. (2022) potential, and that potential already contains the X-shaped/boxy bar. Consequently the recovered bulge density, X-shape, orbital-frequency families, and the claim that gradients 'trace the X-shaped component' (Section 5.2) are not independent measurements: they are a re-projection of the assumed mass model. The chemical results (4:3 high-/low-alpha mass ratio, five-component GMM decomposition, accreted-mass estimate) do add information beyond the density fit, but only under the additional assumptions of a constant 37 km/s/kpc pattern speed, dynamical equilibrium, and phase-mixed abundance painting, i.e. that APOGEE stars on an orbit are representative of every point of that orbit. With a different but plausible potential/pattern speed, the orbital weights and the mapping of metal-rich populations into the bar could change. The authors themselves write in Section 5.1 that the 3D density is defined by the adopted potential by construction, so this is a genuine limitation rather than an overlooked one. What is missing is an exploration of how much the headline numbers move when the central assumption is varied.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper applies the orbit superposition (Schwarzschild) method, previously developed in Papers I and II, to APOGEE DR17 giant stars with Gaia astrometry to reconstruct the 3D density, kinematics, and chemical abundance structure of the Milky Way bulge. The headline results are: (i) the bulge is composed of two main populations, a metal-poor high-alpha thick disc and a metal-rich low-alpha thin disc, with a mass ratio of 4:3; (ii) a five-component 2D GMM decomposition of the [Fe/H]-[Mg/Fe] plane, including a most metal-poor component of likely ex-situ origin with mass about 8e8 M_sun; (iii) the bulge metallicity gradients are spatially variable and trace the X-shaped/boxy density structure; and (iv) although the bulge has slightly subsolar mean metallicity, it is metal-rich relative to the surrounding disc once its full vertical extent is considered. The paper frames these results as evidence that the Milky Way is a typical secularly formed barred galaxy.","tokens_in":34599,"tokens_out":3462,"duration_ms":34340,"significance":"The orbit superposition approach is a promising technique for correcting survey footprint and selection-function biases, and the paper demonstrates its application to a large spectroscopic sample. If the central claims hold, the paper would provide a coherent chemo-kinematic picture of the bulge that supports the bar-buckling scenario and places the Milky Way among external barred galaxies with metal-rich boxy/peanut bulges. The reconstruction of footprint-independent maps, e.g., the face-on metallicity maps in Fig. 15, is a useful methodological contribution. However, the significance is substantially conditional on breaking the circularity with the adopted potential: the orbit weights are fitted to the Sormani et al. (2022) analytic density, which already contains the X-shaped bar, so the recovered morphology and orbital families are partly inherited from the input. The chemical results (mass ratio, GMM decomposition, accreted mass) are less circular, but they lack systematic uncertainty estimates and robustness tests against the adopted potential and pattern speed.","major_comments":[{"comment":"The orbit weights are adjusted to reproduce the analytic 3D stellar density of Sormani et al. (2022), which already contains the X-shaped/boxy bar. Consequently, the 'successful recovery' of the X-shape and boxy morphology in Figs. 1-3 and the orbital family decomposition in Figs. 4-5 are largely a re-projection of the input mass model. The manuscript itself acknowledges this in Section 5 (the sentence beginning 'By construction, the 3D density distribution of the MW bulge is defined by the adopted potential'), but the framing in Section 3.1 and in the Summary ('successfully reconstruct the 3D stellar density structure ... including capturing the distinct X-shaped/boxy structure') overstates the independence of the result. I request either a reframing as a consistency check or, preferably, a stress test with an alternative potential (e.g., one without a boxy/peanut bar, or with a different bar strength/pattern speed) to demonstrate that the orbit superposition plus APOGEE data alone would not artificially produce the X-shape.","section":"Section 2.2 and Section 3.1"},{"comment":"The headline numbers -- the 4:3 high-alpha/low-alpha mass ratio and the 8e8 M_sun accreted-mass estimate -- are quoted without error bars or sensitivity analysis. The mass ratio depends on the arbitrary choice of the high-/low-alpha boundary shown as the white line in Fig. 9, and the accreted mass depends on the GMM membership probabilities, which the authors admit are contaminated. The 50 resamplings only provide scatter in the GMM component centres, not in the model selection or in the derived masses. I ask the authors to provide uncertainty estimates and to test how the mass ratio and accreted mass change when the alpha boundary is varied within a reasonable range, and when the bar pattern speed is varied within the plausible 30-45 km/s/kpc range.","section":"Section 5.3"},{"comment":"The claim that 'no universal metallicity gradient value can characterise the MW bulge' and the conclusion that the bulge is metal-rich relative to the surrounding disc (Fig. 15) rely on density-weighted abundance maps that are constructed from the same orbit superposition. Since the density field is fitted to the adopted potential, the spatial variations of the gradients, including the claim that radial gradients 'closely trace the X-shaped bulge density structure' (Fig. 14), may be partly inherited from the input density rather than from the APOGEE chemistry. A concrete test would be to paint the same APOGEE abundances onto orbits integrated in a different (e.g., axisymmetric or non-X-shaped) potential and compare the resulting gradient maps. Without such a test, the reader cannot assess how much of the chemo-morphological correlation is driven by the assumed potential.","section":"Section 5.1 and Section 5.2"}],"minor_comments":[{"comment":"There is a typo in the final paragraph: 'the MW bugle story' should be 'the MW bulge story'.","section":"Section 6"},{"comment":"In the discussion of extremely metal-rich populations, 'do not suggest the precents of the classical bulge' should be 'do not suggest the presence of a classical bulge'.","section":"Section 4"},{"comment":"The word 'metallcity' in the paragraph beginning 'The fact that we observe a gap' should be 'metallicity'.","section":"Section 5.3"},{"comment":"The notation for the orbital frequency ratio is inconsistent: the text uses fr/fx and fz/fx, but the caption of Fig. 4 defines the top panel as 'in-plane orbital frequency ratio, fr/fx' and the bottom as 'fz/fx'. Please define all frequencies once in the text and use consistent notation throughout.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The stress-test concern raised in the reader's report is legitimate and, on reading the manuscript, is not resolved: the authors themselves acknowledge the density is defined by the adopted potential by construction, but they do not provide the potential-replacement test that would establish how much of the chemo-kinematic results are independent of that assumption. The paper is otherwise carefully written and the method is internally consistent, so I do not see a fatal flaw; a major revision with the requested sensitivity analysis would make the headline claims credible."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this paper gives you the first mass-weighted, footprint-corrected chemo-kinematic view of the MW bulge, and the 2D GMM decomposition of the [Fe/H]-[Mg/Fe] plane is a new product. The 4:3 high-alpha/low-alpha mass ratio and the \"metal-rich bulge in 3D\" claim are the headline results, and they are worth engaging with.\n\nWhat is actually new: the orbit superposition method itself came from Papers I and II; the bulge application is the extension. The selection-function-free maps of chemistry, kinematics, and gradients are new, and the honest comparison with raw APOGEE MDFs (Fig. 10) shows the correction is modest — good practice. The GMM analysis is standard but carefully done, and the authors are explicit that the two intermediate components are not distinct populations.\n\nThe soft spots are real but not fatal. The main one is the circularity: the orbit weights are fitted to the Sormani et al. 2022 analytic density, which already has the X-shape, so the recovered X-shape, orbital families, and density maps are inherited rather than measured. The paper openly says this in Section 5.1, which I respect, but it means the orbital decomposition in Sections 3 and 4 is not an independent result. The chemical maps are conditional on that potential, the constant 37 km/s/kpc pattern speed, and orbit-painted abundances. The headline numbers — 4:3, accreted mass, gradient amplitudes — have no error bars and no systematic exploration of potential choice. That is the missing piece: how much do these numbers move with a different, still plausible potential or pattern speed? Also, no code or data is shipped, so full verification is not possible.\n\nI disagree slightly with the reader's framing on one point: the chemical claims are not circular in the same way as the density claims. The GMM decomposition and the mass ratio are conditional on the potential, but they are not re-productions of its input; they do add information. They just lack quantified systematics.\n\nWho gets value: bulge/bar researchers and chemo-dynamical modellers will cite the mass-weighted maps and the GMM decomposition. The paper deserves a serious referee. It should not be desk-rejected; it should go to review with the expectation that the authors quantify the potential dependence, add error bars, and ideally release the orbit catalogue or code. I would send it back for major revision on those grounds, but the core product is worth having in the literature.\n\nI'd bring this to reading group and cite it myself.","headline":"A genuinely useful, selection-function-free chemo-kinematic map of the bulge, but the headline numbers carry no error bars and the density/orbital results inherit the assumed Sormani potential.","tokens_in":35315,"tokens_out":1616,"would_cite":true,"duration_ms":18481,"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":"This paper argues that the Milky Way's bulge is built from two disc populations in a 4:3 mass ratio and is metal-rich relative to the surrounding disc once the full vertical extent is counted.","keywords":["Milky Way bulge","orbit superposition","Schwarzschild method","APOGEE","Gaia astrometry","thick disc / thin disc","metallicity distribution function","boxy/peanut bulge"],"falsifier":"Re-run the orbit-superposition reconstruction with an independently constructed potential that has no X-shaped bulge while matching all other observables; if the X-shape, the 4:3 thin-to-thick disc mass ratio, and the five-component chemical decomposition disappear or change drastically, the central results are artifacts of the chosen potential.","tokens_in":34158,"feed_emoji":"🌌","tokens_out":8983,"duration_ms":72638,"temperature":0.7,"pith_summary":"This paper tries to settle a long dispute about the Milky Way's bulge by reconstructing its three-dimensional structure from the orbits of APOGEE giant stars. It claims the bulge is not a classical spheroid but a bar-built structure made mostly of two disc populations: a metal-poor, high-$\\alpha$ thick disc and a metal-rich, low-$\\alpha$ thin disc in a 4:3 mass ratio. It further claims that although the bulge's mean metallicity is slightly below solar, the bulge is metal-rich compared with the surrounding disc once the full vertical extent is included, in line with external barred galaxies. A five-component decomposition of the [Fe/H]-[Mg/Fe] plane identifies an accreted, ex-situ population of about $8\\times10^8\\,M_\\odot$ and two intermediate components that merely trace the transition between the two main discs. If these claims hold, the Milky Way is a typical secularly formed barred galaxy.","feed_headline":"4:3 mix of two discs shapes the Milky Way bulge","feed_subtitle":"Orbit-superposition maps from APOGEE and Gaia show the bulge is metal-rich next to its surrounding disc.","key_machinery":"The central mechanism is orbit superposition: each APOGEE giant star's orbit is integrated in a fixed analytic gravitational potential of the inner Milky Way (taken from Sormani et al. 2022, which already includes the X-shaped bar), with a constant bar pattern speed of 37 km/s/kpc. The orbits are assigned non-negative weights by fitting their combined 3D density to the stellar component of that same analytic potential, then each orbit is sampled at 500 phase-space points and the star's measured chemistry is painted along the orbit, spreading abundance information into regions the APOGEE footprint never observed. This converts the survey's patchy, midplane-censoring footprint into a complete, mass-weighted 3D chemo-kinematic model of the bulge. The weight-fitting step is what makes the reconstruction of the X-shape and orbital families, such as banana, pretzel, and longer bar orbits, possible, but it also means those structures are inherited from the adopted potential.","core_discovery":"On the paper's own terms, the discovery is that the present-day Milky Way bulge is built from the same two chemical discs that extend to the solar neighbourhood: an inner thick disc that is metal-poor with subsolar metallicity and high [Mg/Fe] (the high-$\\alpha$ population) and a thin disc that is metal-rich with supersolar metallicity and low [Mg/Fe], contributing in a 4:3 mass ratio inside 3.5 kpc. The paper argues that the apparent metal-poorness of the bulge in (l,b) projections is a projection effect: near the midplane the metal-rich, kinematically cold population has been pushed above the plane by the vertical bar instability, so once the full vertical extent is counted the bulge and the bar major axis are more metal-rich than the surrounding disc. It also finds that no single metallicity gradient describes the bulge: radial gradients trace the X-shaped density structure and vertical gradients trace the boxy component, which explains why surveys covering different fields have reported different gradient values. The most metal-poor of five 2D Gaussian components in the [Fe/H]-[Mg/Fe] plane carries the chemical signature of accreted stars and a mass near $8\\times10^8\\,M_\\odot$.","pith_inferences":["Editorial inference: a direct test the authors do not perform is to leave the bar pattern speed free rather than fixed at 37 km/s/kpc; the inferred resonance families, bar mass, and the 4:3 thin-to-thick disc ratio could shift if the adopted speed is wrong.","Editorial inference: because the X-shape is present in the input potential, the orbital-family decomposition is best read as a property of the assumed potential combined with the data, not as an independent measurement of the Milky Way's orbital structure.","Editorial inference: the same orbit-superposition method could be applied to external barred galaxies with integral-field spectroscopy, replacing the chemical painting step with stellar-population gradients from IFU data to test whether the metal-rich-bar pattern is universal.","Editorial inference: the metal-rich bar along the major axis implies that kinematic fractionation plus suppressed star formation, rather than in-situ enrichment, sets the abundance pattern; age-dating the metal-rich bulge stars, which the paper omits because of age-catalogue quality issues, could separate these explanations."],"forward_implications":["The bulge's two main populations are the inner extensions of the thick and thin discs, so the bulge's chemical history is largely the disc's chemical history, not a separate spheroidal formation event.","The 4:3 mass ratio between the metal-poor high-alpha and metal-rich low-alpha components in the bulge, if correct, is a constraint on the relative masses of the thick and thin disc populations before the bar buckled.","Because the recovered metallicity gradients vary across the bulge and trace the X-shaped and boxy structure, different spectroscopic surveys of different fields can legitimately report different gradients, and there is no single bulge gradient to compare.","The small accreted component of about $8\\times10^8\\,M_\\odot$ for stars with [Fe/H] above about -1.2, together with the absence of a significant spheroid, reinforces the picture of the Milky Way as a secularly formed barred galaxy.","Future bulge surveys that cover the midplane regions APOGEE misses should see a metal-rich bar along the major axis, matching the paper's prediction."],"supporting_citations":[{"why":"Supplies the analytic 3D mass distribution and potential of the inner Milky Way that the orbit weights are fitted to reproduce.","marker":"Sormani et al. 2022"},{"why":"Provides the bar and bulge mass model and one of the key measurements of the 37 km/s/kpc bar pattern speed adopted in the orbit integration.","marker":"Portail et al. 2017"},{"why":"Gives the 3D X-shaped bulge density reconstruction that the adopted analytic potential is designed to match and against which the recovered structure is compared.","marker":"Wegg & Gerhard 2013"},{"why":"Defines the five-component ARGOS bulge metallicity distribution used as the reference for the number and positions of MDF peaks.","marker":"Ness et al. 2013a"},{"why":"Provides five MDF components from microlensed dwarf and subgiant stars; the paper compares its orbital-family MDFs to those peak locations.","marker":"Bensby et al. 2017"},{"why":"Earlier three-component GMM decomposition of APOGEE DR16 bulge data used as a comparison point for the new 2D GMM analysis.","marker":"Rojas-Arriagada et al. 2020"},{"why":"Four-Gaussian bulge decomposition used to frame the number of MDF components and the significance of the most metal-poor one.","marker":"Wylie et al. 2021"},{"why":"Establishes kinematic fractionation, the mechanism invoked to explain why colder, metal-rich populations show a sharper X-shape and are displaced vertically.","marker":"Debattista et al. 2017"},{"why":"Simulations predicting that boxy and peanut bulges are metal-rich; used as evidence that a metal-rich Milky Way bulge makes it a typical barred galaxy.","marker":"Fragkoudi et al. 2020"},{"why":"The contrary metal-poor-bulge result that the paper argues is an artifact of APOGEE midplane bias and incomplete selection-function correction.","marker":"Queiroz et al. 2021"}],"fun_headline_variants":["Bulge's two discs: thin and thick in 4:3 mix","No single metallicity gradient for the Milky Way bulge","Bulge is metal-richer than disc, two discs in 4:3","APOGEE+Gaia map bulge's two-disc recipe","X-shaped bulge: no universal gradient, says orbit-superposition"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The entire reconstruction assumes the adopted analytic model of the inner Milky Way's mass distribution, which already contains the X-shaped bar, is the true potential, and that the Galaxy is in dynamical equilibrium with a bar rotating at a fixed 37 km/s/kpc; because the orbit weights are fitted to reproduce that same model's density, the recovered X-shape, orbital families, and mass ratios are partly built into the input rather than independently derived from the data.","fun_headline_variants_meta":{"raw":{"variants":["Bulge's two discs: thin and thick in 4:3 mix","No single metallicity gradient for the Milky Way bulge","Bulge is metal-richer than disc, two discs in 4:3","APOGEE+Gaia map bulge's two-disc recipe","X-shaped bulge: no universal gradient, says orbit-superposition"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000272,"raw_usage":{"total_tokens":1738,"prompt_tokens":1154,"completion_tokens":584,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":770,"completion_tokens_details":{"reasoning_tokens":491}},"tokens_in":770,"tokens_out":584,"duration_ms":5750,"temperature":1.0,"reasoning_tokens":491,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T11:25:38.700959+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-run the orbit-superposition reconstruction with an independently constructed potential that has no X-shaped bulge while matching all other observables; if the X-shape, the 4:3 thin-to-thick disc mass ratio, and the five-component chemical decomposition disappear or change drastically, the central results are artifacts of the chosen potential.","supporting_citations":[{"cited_title":"M., Gerhard, O","cited_arxiv_id":null,"evidence_quote":"Four-Gaussian bulge decomposition used to frame the number of MDF components and the significance of the most metal-poor one."}],"review_version":1}