{"id":"dd2b1c1c-6471-4b76-919d-26552b663dea","arxiv_id":"2505.02917","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"An orbit-superposition method recovers bar pattern speeds of edge-on barred galaxies from mock MUSE-like data, with true values inside 1-sigma intervals in 10 of 12 side-on cases and in 3 of 4 end-on cases under a strong-bar prior.","lead":"Astronomers have a new way to measure how fast a galaxy's central bar rotates even when the galaxy is seen edge-on, a situation where older methods fail. Tests on simulated galaxies recover the bar's pattern speed with about 10% uncertainty for side-on bars and 14% for end-on bars with an added assumption.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"1σ confidence intervals are computed with the potential fixed at the best fit (Sect. 4.1), so the quoted average 10% uncertainty likely understates degeneracies between Ωp and the mass–potential parameters, undermining the headline precision claim.","rationale":"The reader's verdict of CONDITIONAL is appropriate, and their rationale already notes the uncertainty-calculation issue in passing. However, the reader's designated 'weakest assumption' is the bar–disk projected alignment, whereas I find the fixed-potential confidence-interval construction to be the more load-bearing concern for the paper's central quantitative claim. The mock validation convincingly shows small bias in Ωp for side-on bars; the remaining question is whether the quoted 1σ uncertainties are meaningful. The paper's uncertainty method (Sect. 4.1) is a conditional Monte Carlo: it holds the potential fixed and only refits orbit weights. In the presence of strong degeneracies between the stellar mass-to-light ratio and the dark matter parameters (Fig. 5, Fig. C.1), the true profile likelihood in Ωp should be broader than the threshold derived from a fixed potential. The concrete test proposed would directly estimate the full-parameter scatter and settle whether the 10% precision claim is robust. If the test confirms wider scatter, the paper's precision claim would need to be revised, but the method's ability to recover Ωp without large bias (the qualitative claim) would likely remain. Thus the verdict stays CONDITIONAL: promising method, but the quoted confidence intervals require stronger validation before the method is used on real data. The reader's weakest assumption about bar–disk misalignment is a separate, valid concern about real-data applicability, but it does not directly undermine the mock-based central claim as much as the uncertainty calculation does.","tokens_in":35476,"tokens_out":21449,"duration_ms":225551,"concrete_test":"Recompute the 1σ confidence interval for one side-on case (e.g., Au-23-85-50) by generating 20 independent noise realisations of the mock kinematic maps and re-running the full seven-parameter search (including the potential parameters) for each realisation. Compare the standard deviation of the best-fit Ωp values across realisations with the quoted interval (21+3−3 km s−1 kpc−1). If the run-to-run scatter exceeds the quoted interval by more than ≈50%, the fixed-potential perturbation method understates the uncertainty and the 10% precision claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central quantitative claim is that side-on edge-on bars yield Ωp with an average 1σ uncertainty of 10% (Sect. 5.1). This precision estimate depends on how the model uncertainties are computed. In Sect. 4.1, the authors perturb the kinematic maps 1000 times, fix the gravitational potential at the best-fit parameters, and re-solve only the orbit weights. The width of the resulting χ2 distribution is then used as Δχ2_CL, the threshold for defining 1σ confidence intervals on all free parameters. This procedure ignores that Ωp is determined jointly with the potential parameters. The parameter exploration in Fig. 5 shows strong degeneracies among M⋆/L, dark matter c, M200, γ, qdisk, and φ; Fig. C.1 explicitly demonstrates a large degeneracy between stellar mass and dark matter mass. Since the rotating potential depends on the mass distribution, changes in these parameters can partially compensate for a different Ωp while yielding similar projected kinematics. The fixed-potential perturbation cannot capture these degenerate directions, so the resulting Δχ2_CL is not the correct threshold for the profile likelihood over the full seven-parameter space. Consequently, the quoted 1σ intervals (e.g., Ωp = 21+3−3 km s−1 kpc−1 for Au-23-85-50) and the headline average uncertainty of 10% are likely underestimates. The coverage statistics (10/12 cases within 1σ) are not sufficient to validate the precision claim, because the intervals themselves may be too narrow. This concern directly affects the central claim of the paper: the method appears to recover Ωp well, but the reported accuracy is not established by the current uncertainty analysis.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents an orbit-superposition method, built on the barred-galaxy implementation of Tahmasebzadeh et al. (2021, 2022), for estimating the bar pattern speed Ωp of edge-on barred galaxies. The method is tested on 16 MUSE-like mock data sets constructed from three Auriga simulated galaxies observed at inclinations θT ≥ 85° and bar azimuthal angles φT covering both side-on (φT ≥ 50°) and end-on (φT ≤ 30°) configurations. The central claims are: for side-on bars, the recovered Ωp encloses the true pattern speed ΩT within the 1σ (68%) confidence levels in 10 of 12 cases, with an average 1σ uncertainty of 10%; for end-on bars, uncertainties degrade, but imposing a strong-bar prior (pbar ≤ 0.50) improves the average uncertainty to 14% with the true value covered in three of four cases; and across all models the 2σ (95%) intervals always cover ΩT. The true pattern speeds are taken from independent measurements (the TW method and high-cadence snapshot tracking from Fragkoudi et al. 2020, 2021), so the recovery test is not circular. The paper also includes robustness checks on the bar/disk separation radius, on the treatment of BP/X-shaped structures, and on dark matter halo constraints.","tokens_in":35805,"tokens_out":4347,"duration_ms":52929,"significance":"If the claims hold, this is a valuable methodological advance: it would open edge-on barred galaxies, which are inaccessible to the Tremaine-Weinberg method, to dynamical pattern-speed measurement, with direct applications to surveys such as GECKOS. The paper’s strengths include the use of realistic Auriga simulations, independent validation of the recovered Ωp against published true values, a careful seven-parameter search illustrated by the full parameter-space figures, and explicit robustness tests. The reported 10% uncertainty for side-on cases is an attractive headline figure. However, the precision claim depends on how model uncertainties are computed, and that computation (Sect. 4.1) is the main point of concern. The end-on improvement also rests on a post hoc strong-bar prior. With the uncertainty-calibration issue addressed, the paper would be a solid contribution; in its current form the headline uncertainty is not fully demonstrated.","major_comments":[{"comment":"The 1σ confidence threshold Δχ2_CL is derived by perturbing the kinematic maps 1000 times and re-solving only the orbit weights, with the gravitational potential fixed at the best-fit parameters. This procedure does not sample the seven-parameter degeneracies that are visible in Fig. 5 and Fig. C.1, particularly among M⋆/L, c, M200 (Mdm,10kpc), γ, qdisk and φ. Since Ωp is fitted jointly with these parameters, changes in the potential can partially compensate for a different Ωp while preserving similar projected kinematics, so the fixed-potential bootstrap is likely to understate the true uncertainty. The quoted 1σ intervals (e.g. Ωp = 21+3−3 km s−1 kpc−1 for Au-23-85-50) and the headline average uncertainty of 10% therefore need support from a joint treatment. I recommend either a joint bootstrap over all seven parameters or, more directly, a coverage calibration in which many noise realisations of each mock are generated and the full parameter search is re-run for each realisation, with the fraction of true ΩT inside the resulting 68% intervals reported. This would test whether the intervals are correctly calibrated rather than relying on the current 10-of-12 count, which is weak evidence for calibration on only 12 cases.","section":"Sect. 4.1 and Figs. 5, 7, C.1"},{"comment":"The improved end-on results (average uncertainty 14%, true ΩT within 1σ in three of four cases) are obtained after imposing the strong-bar prior pbar ≤ 0.50, a constraint that was not part of the original model and was introduced after the unconstrained runs showed degraded recovery. Because the same four end-on mock data sets are used both to motivate the prior and to evaluate its success, the quoted success rate for the constrained models is optimistic; in a real application the analyst will not know in advance whether an edge-on galaxy hosts a strong bar. I recommend presenting the pbar ≤ 0.50 runs explicitly as an exploratory test, or validating the strong-bar prior on an independent set of end-on mocks (including galaxies with weaker or intermediate bars), before it is used as a recipe for real observations.","section":"Sect. 5.2 and Fig. 8"},{"comment":"The deprojection assumes that the bar major axis is aligned with the disk major axis in the observing plane (ψbar ≈ ψdisk = 90°) and that the disk major axis aligns with the x′-axis. The paper notes that the misalignment is usually small, but it does not test the sensitivity of the recovered Ωp to a small but non-negligible misalignment. Since the deprojected 3D stellar density and hence the rotating potential depend on this assumption, a misaligned bar in a real edge-on galaxy could bias the inferred pattern speed. Please add a test with mock data rotated by, for example, 5° and 10° relative to the assumed alignment, and quantify the resulting bias in Ωp. This is needed to support the stated applicability to real edge-on galaxies such as the GECKOS sample.","section":"Sect. 3.1.1"}],"minor_comments":[{"comment":"In the text after the left panel of Fig. 8, the sentence describing the revised models says the uncertainties are 'slightly reduced in the cases with (85°,30°) and (89°,10°), and significantly reduced in the case with (89°,10°)'; the repeated (89°,10°) is presumably a typo, and one of the two cases should be a different viewing angle.","section":"Sect. 5.2"},{"comment":"The symbol δ′j is used in Eqs. (5) and (6) but defined only afterwards; please define δ′j = 1 − q′j2 just before Eq. (5).","section":"Eqs. (5)–(7)"},{"comment":"The assumed stellar mass-to-light ratio (M⋆/L)T = 2 is stated without units; please specify that it is in solar units, as is implied by the rest of the paper.","section":"Sect. 2.2"},{"comment":"The first row of Fig. 5 appears to plot the bar azimuthal angle but the axis label is rendered as only 'p' in the PDF; please check the figure labels for garbled symbols.","section":"Fig. 5"},{"comment":"In the R0 sensitivity test, the case labels in Fig. E.1 and the text are consistent, but the caption of Fig. E.1 should state which coloured lines correspond to smaller and larger R0 before the text refers to them, to make the figure self-contained.","section":"Sect. 6.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of A&A and builds sensibly on the authors' prior work; I see no citation or novelty concern. The central accuracy result for side-on bars is credible because the true pattern speeds come from independent measurements. My main worry is the uncertainty calibration in Sect. 4.1: the fixed-potential bootstrap is unlikely to capture degeneracies between Ωp and the potential parameters, and the coverage statistics (10/12) are too sparse to validate the precision claim. I would not require new science, but the authors should either implement a joint bootstrap or provide a coverage-calibration experiment; without it the headline '10% uncertainty' is not supported. The post hoc pbar ≤ 0.50 prior in the end-on case should also be reframed or validated on independent mocks. I recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis paper is worth your time if you care about bar pattern speeds and the GECKOS survey. The genuinely new thing: it extends the authors' orbit-superposition framework to edge-on barred galaxies, with free deprojection and BP/X handling, and validates it on 16 mock data sets from three Auriga galaxies. That is a concrete capability the TW method cannot cover, so it fills a real observational gap.\n\nWhat the paper does well: the side-on bar recovery is convincing. Ten of twelve cases put the true Omega_p inside the 1-sigma interval, with an average uncertainty around 10%, and they compare against independent pattern-speed measurements from Fragkoudi et al. — so the result is not baked in by construction. The end-on cases are honestly presented: the original models struggle, and the stronger bar prior (pbar <= 0.50) improves things but still misses one of four at 1-sigma. They also probe robustness to the bar/disk separation radius, BP/X treatment, and dark matter halo assumptions. That is the kind of testing a methods paper should include.\n\nThe soft spots: the uncertainty calculation is the biggest one. In Sect. 4.1 they fix the potential at the best fit, perturb the kinematic maps, and re-solve only the orbit weights to set Delta chi^2_CL, then use that threshold on the full parameter-space chi^2 landscape. That ignores degeneracies between Omega_p and the mass–potential parameters (M*/L, dark matter c, M200, gamma, qdisk, phi), which the paper itself shows are substantial (Fig. 5 and C.1). So the quoted 1-sigma intervals are probably not rigorous profile-likelihood intervals. That said, the coverage statistics do not show undercoverage — 10/12 is actually slightly above 68% — so the intervals seem roughly calibrated in practice. I would call this a methodological concern rather than a fatal flaw, but it needs to be addressed or at least discussed more carefully.\n\nThe other limitation is the assumption that the bar major axis aligns with the disk major axis in the observing plane. The authors note the misalignment is usually small but never test how a few degrees of misalignment would bias Omega_p. For real edge-on galaxies, that is exactly the sort of systematic you would worry about.\n\nOverall: a solid, honest methods paper. The side-on case looks ready for real data; the end-on case is honestly flagged as harder. I would send this to review. A good referee will push on the uncertainty calculation and ask for a misalignment test, but the core validation is strong enough to merit that effort.\n\nBest,\n[Your name]","headline":"Extends orbit-superposition to edge-on barred galaxies with a genuinely convincing side-on validation; end-on is honestly weaker and the uncertainty calculation has a methodological gap that deserves referee attention.","tokens_in":36459,"tokens_out":3576,"would_cite":true,"duration_ms":40465,"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":"An orbit-superposition model recovers the bar pattern speed $\\Omega_{\\rm p}$ of edge-on galaxies to about 10 percent (side-on bars) and 14 percent (end-on bars with a strong-bar prior), with 2$\\sigma$ intervals always covering the truth.","keywords":["pattern speed","barred galaxies","edge-on galaxies","orbit-superposition modelling","galactic dynamics","integral field spectroscopy","BP/X-shaped bulge"],"falsifier":"Take one of the mock galaxies, re-project it with the bar deliberately misaligned from the disk line of nodes by 5$^\\circ$–10$^\\circ$ while keeping the true $\\Omega_{\\rm T}$ fixed, run the same pipeline, and check whether the recovered $\\Omega_{\\rm p}$ still falls within the claimed 1$\\sigma$ interval; if it does not, the method's accuracy is limited to aligned bars. A complementary check is to apply the method to an edge-on galaxy whose pattern speed is already known from an independent, low-inclination measurement and compare the two values.","tokens_in":35233,"feed_emoji":"🌀","tokens_out":9384,"duration_ms":97887,"temperature":0.7,"pith_summary":"This paper extends orbit-superposition dynamical modelling to edge-on barred galaxies and tests whether the bar pattern speed $\\Omega_{\\rm p}$ can be recovered from integral-field kinematics. On 16 mock data sets built from three simulated galaxies with known pattern speeds, viewed at inclinations of 85$^\\circ$ or more, side-on bars recover $\\Omega_{\\rm p}$ with an average 1$\\sigma$ uncertainty of about 10 percent, with the true value inside the 1$\\sigma$ interval in 10 of 12 cases. End-on bars are harder: unconstrained models reach about 30 percent uncertainty when the bar is nearly perpendicular to the line of sight, but a prior that the bar is strong (axis ratio $p_{\\rm bar}\\le 0.50$) cuts the average 1$\\sigma$ uncertainty to 14 percent and keeps the true value inside 1$\\sigma$ in three of four cases. The significance is that nearly edge-on galaxies—roughly half of all barred disks—have so far been inaccessible to pattern-speed measurement, and this method is designed for exactly that geometry.","feed_headline":"Edge-on galaxy bar speeds recovered to ~10 percent","feed_subtitle":"Orbit-superposition models pull bar rotation from edge-on kinematics, where standard methods fail.","key_machinery":"The load-bearing machinery is a rotating triaxial barred potential reconstructed from the projected light: the surface brightness is fit by multi-Gaussian expansion (MGE), Gaussians are split into bar and disk at a separating radius $R_0$, and each component is deprojected under the assumption that the bar major axis coincides with the disk line of nodes, with the bar position angle fixed by physical priors (non-negative luminosity, bar axis ratio $0.25\\le p_{\\rm bar}\\le 0.7$, and boxy/peanut/X-shaped prominence). Orbits are sampled in the x-z plane of the rotating frame, including retrograde orbits, and integrated about 200 periods; non-negative least squares assigns orbit weights that reproduce the surface brightness and the Gauss–Hermite kinematic maps. The pattern speed $\\Omega_{\\rm p}$ is just the figure-rotation rate of the bar component in a seven-parameter space (disk axis ratio, bar azimuth, stellar mass-to-light ratio, dark-matter concentration, dark-matter mass, inner density slope, and $\\Omega_{\\rm p}$), and the confidence intervals come from 1000 perturbed realisations of the kinematic data.","core_discovery":"The central claim is that an edge-on view does not prevent orbit-superposition modelling from measuring bar pattern speed, as long as the bar is not seen almost end-on. In the side-on geometry (bar azimuthal angle $\\varphi_{\\rm T}\\ge 50^\\circ$), the model-recovered $\\Omega_{\\rm p}$ brackets the true value at 1$\\sigma$ in 10 of 12 mock cases, with an average 1$\\sigma$ uncertainty of 10 percent; all 12 cases are covered at 2$\\sigma$. In the end-on geometry ($\\varphi_{\\rm T}\\le 30^\\circ$), precision degrades substantially—up to about 30 percent at $\\varphi_{\\rm T}=10^\\circ$—and unconstrained models tend to reinterpret a strong end-on bar as a weak side-on bar; adding the prior $p_{\\rm bar}\\le 0.50$ restores average 1$\\sigma$ uncertainty to 14 percent and recovers the truth in three of four cases. Across every model in the paper, the 2$\\sigma$ confidence interval contains the true $\\Omega_{\\rm T}$.","pith_inferences":["Because $\\Omega_{\\rm p}$ enters the fit roughly as projected bar-end velocity divided by projected bar length, the recovered $\\Omega_{\\rm p}$ should be nearly independent of the recovered bar azimuth; a dedicated test varying $\\varphi_{\\rm T}$ while holding $\\Omega_{\\rm T}$ fixed would show how far this cancellation holds, an experiment the paper does not run.","The same degeneracy that makes end-on bars hard—strong end-on bars masquerading as weak side-on bars—could be broken by adding kinematic bar diagnostics such as the sign of the $h_3$–$V$ correlation, avoiding the need for an ad hoc $p_{\\rm bar}$ prior.","If real bars are systematically misaligned with the disk line of nodes by more than a few degrees, the fixed-alignment deprojection will bias the potential and hence $\\Omega_{\\rm p}$; constructing mock data with controlled misalignment angles would quantify the tolerance.","The reported confidence intervals are computed from orbit-weight re-solution under data perturbations, so they may understate systematics from potential-shape misspecification; cross-checks against pattern speeds from corotation features on real galaxies would test this."],"forward_implications":["Side-on edge-on bars have their pattern speeds recoverable to about 10 percent (1$\\sigma$), comparable to what existing orbit-superposition and line-of-sight methods achieve for non-edge-on galaxies.","End-on edge-on bars remain recoverable to about 14 percent average 1$\\sigma$ uncertainty if the bar is known or assumed to be strong, with $p_{\\rm bar}\\le 0.50$.","Boxy/peanut/X-shaped morphology is not required: the simulated galaxy without such a feature is recovered just as precisely as the two that have it.","The recovered $\\Omega_{\\rm p}$ stays within 2$\\sigma$ of the truth when the bar/disk separating radius $R_0$ is varied by factors 0.6–1.3, and when the dark-matter halo is allowed to shift within the stellar/dark degeneracy.","The method is positioned for direct application to real integral-field observations of edge-on galaxies, where the pattern speed has not previously been measurable."],"supporting_citations":[{"why":"Supplies the orbit-superposition machinery—potential construction, orbital sampling in the x-z plane, and non-negative least-squares weighting—that the present method adapts.","marker":"van den Bosch et al. (2008)"},{"why":"Develops the barred variant with a rotating bar potential and demonstrates about 10 percent $\\Omega_{\\rm p}$ recovery for non-edge-on galaxies, the starting point extended here to edge-on geometry.","marker":"Tahmasebzadeh et al. (2021, 2022)"},{"why":"Provides the three simulated galaxies whose known internal structure and pattern speeds serve as the truth for the mock data sets.","marker":"Grand et al. (2017)"},{"why":"Supplies the true pattern speeds for Au-18 and Au-23 from the TW method at 5 percent accuracy, used as the recovery target.","marker":"Fragkoudi et al. (2020)"},{"why":"Supplies the true pattern speed of Au-28 from high-cadence snapshots and confirms Au-18's value, fixing the comparison values.","marker":"Fragkoudi et al. (2021)"},{"why":"Shows that the standard TW method delivers accurate $\\Omega_{\\rm p}$ only for non-edge-on views, establishing the gap this paper fills.","marker":"Guo et al. (2019)"},{"why":"The multi-Gaussian expansion formalism used to fit and deproject the surface brightness into the triaxial bar and disk components.","marker":"Cappellari (2002)"}],"fun_headline_variants":["Edge-on bar speeds measured to 10% via orbit-superposition","Orbit-superposition recovers bar pattern speeds in edge-on galaxies","Bar speed from edge-on view: orbit-superposition gets 10% precision","Edge-on barred galaxies: pattern speed recovery improved to 10%","New orbit-superposition method recovers edge-on bar rotation speeds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The deprojection assumes the bar major axis lies exactly along the disk's line of nodes in the observing plane; if a real bar is misaligned by more than a few degrees, the 3D stellar density, the rotating potential, and the recovered $\\Omega_{\\rm p}$ could all be biased.","fun_headline_variants_meta":{"raw":{"variants":["Edge-on bar speeds measured to 10% via orbit-superposition","Orbit-superposition recovers bar pattern speeds in edge-on galaxies","Bar speed from edge-on view: orbit-superposition gets 10% precision","Edge-on barred galaxies: pattern speed recovery improved to 10%","New orbit-superposition method recovers edge-on bar rotation speeds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000923,"raw_usage":{"total_tokens":4033,"prompt_tokens":1095,"completion_tokens":2938,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":711,"completion_tokens_details":{"reasoning_tokens":2842}},"tokens_in":711,"tokens_out":2938,"duration_ms":21288,"temperature":1.0,"reasoning_tokens":2842,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:40:31.289198+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take one of the mock galaxies, re-project it with the bar deliberately misaligned from the disk line of nodes by 5$^\\circ$–10$^\\circ$ while keeping the true $\\Omega_{\\rm T}$ fixed, run the same pipeline, and check whether the recovered $\\Omega_{\\rm p}$ still falls within the claimed 1$\\sigma$ interval; if it does not, the method's accuracy is limited to aligned bars. A complementary check is to apply the method to an edge-on galaxy whose pattern speed is already known from an independent, low-inclination measurement and compare the two values.","supporting_citations":[],"review_version":1}