{"id":"3526e55c-1a91-46ab-b685-d7f78c98bfd6","arxiv_id":"2501.00788","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"SMC young stars trace a rotating, highly inclined east-west stretched disk while old stars trace a non-rotating ellipsoid, with a new infalling feature (SEA) appearing in the residuals.","lead":"The paper models the Small Magellanic Cloud's motion and shape using Gaia proper motions of nine star populations, finding that stars younger than 400 million years rotate as a stretched disk while older stars do not rotate. A generalist reader might care because one dwarf galaxy now shows a sharp, age-stratified picture of how structure forms during tidal interaction with a larger neighbor.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Old-population orientation angles (i, Theta) may be prior-driven rather than data-driven when v_f is fixed to zero; the age-gradient morphology claim needs a likelihood-flatness check.","rationale":"The strongest claim is two-part: young (<400 Myr) populations form a rotating disk, and old populations form a non-rotating flattened ellipsoid with different orientation. The first part has real support: coherent rotation profiles for YMS1/2 and CLSY, COM proper motions consistent with N21/D20, and the 3D red-giant fit recovers vsys near the assumed value. The second part is the load-bearing risk. Even if the V02 formalism includes the perspective terms that make i/Theta weakly identifiable at v_f=0, the paper does not demonstrate that this weak signal dominates the reported 1-2 degree constraints; the quoted residual RMS and known Gaia DR3 systematics are of the same order as the expected depth-induced PM modulation. Because the age-gradient morphology conclusion is explicitly driven by the old-population i/Theta values, this must be checked. I therefore keep the reader's CONDITIONAL verdict: the paper is valuable but the morphological-evolution headline needs an identifiability demonstration. My agreement with the reader is partial: the uniform-translation phrasing is stronger than what the V02 full projection would imply, but the core concern stands.","tokens_in":19703,"tokens_out":11623,"duration_ms":121554,"concrete_test":"Profile-likelihood flatness check: for the RGB (or RC) dataset, fix v_f=0 and compute the maximum log-likelihood over a dense grid in (i, Theta), profiling vt and theta_t at each grid point with the authors' V02 likelihood. If the range of ln L over i=40-90 deg and Theta=150-260 deg is less than about 2, the 1-2 deg uncertainties in Table 1 are not data-driven, so the old-population orientations and the age-gradient morphology claim should be re-framed as unconstrained; if it is much larger, the reader's concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline age-gradient morphology rests on the viewing angles reported for the non-rotating populations in Table 1 (YMS3, RGB, RC, CLSO). In the V02 model with v_f fixed to zero, the rotation term vanishes; the modeled PM is then dominated by a constant COM shift, and any remaining dependence on i/Theta enters only through perspective/depth modulation of the ~420 km/s COM motion, at the ~(z/D0) level (~0.1 mas/yr for a 5 kpc depth at 62 kpc). This is comparable to the quoted residual RMS (0.07-0.22 mas/yr) and to Gaia DR3 correlated systematics, so it is not evident that the 1-2 degree uncertainties in Table 1 are data-driven rather than prior-driven. The no-rotation variant is selected by 'convergence' (Sections 3.2, 4.2) rather than by a model comparison, and no posterior/profile diagnostics over i/Theta for the non-rotating fits are shown. The Theta-age trend is additionally fragile: CLSO (Theta=240 deg) is the main driver of the increase and is admitted to be sparse, while the Red Giants give i=66 deg, breaking the monotonic i decrease. A profile-likelihood check will settle whether the old-population orientations are actually identified.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper models Gaia DR3 proper motions of nine SMC stellar populations using the van der Marel et al. (2002) kinematic framework, with MCMC fits that either include or exclude a rotation component. The central claim is an age-dependent structural transition: young populations younger than ~400 Myr (YMS1, YMS2, CLSY, and CLSI) are consistent with a rotating, highly inclined disk with asymptotic rotation velocity vf ~ 49-89 km/s and scale radius Rf ~ 6-9 kpc, while older populations (YMS3, RGB, RC, CLSO, Red Giants) show negligible rotation and a flattened ellipsoidal morphology. The paper also reports that the inclination decreases and the position angle of the line of nodes increases with age, estimates a line-of-sight extension of ~30 kpc, and identifies four residual proper-motion anomalies (EA, SEA, SA, WA), including a newly claimed infalling feature (SEA).","tokens_in":19858,"tokens_out":6390,"duration_ms":65385,"significance":"If the age-dependent transition from a rotating young disk to a non-rotating old ellipsoid is real, it would be an important constraint on the SMC's tidal and dynamical evolution and on its interaction history with the LMC. The COM proper-motion estimates agree broadly with recent literature (Table 2), and the residual-anomaly maps, especially the newly identified SEA, provide a useful observational baseline for simulations. The paper's main limitation is that the headline structural gradient rests on viewing angles for non-rotating populations whose identifiability is not demonstrated, and on a model-selection criterion based on convergence rather than statistical comparison. These issues are local and testable, so the central claim is defensible if the requested diagnostics support it.","major_comments":[{"comment":"The choice between the rotating and non-rotating model variants is made by \"convergence in the posterior distribution\" (Sections 3.2, 4.2), not by a model-comparison statistic. For YMS3, RGB, RC, and CLSO, a non-converging rotating fit does not by itself demonstrate that the rotation amplitude is zero; it may indicate that the data are unable to constrain the rotation parameters. I request a quantitative comparison between the two variants for each population, for example via BIC/AIC or cross-validated log likelihood, together with posterior distributions of vf for the old populations with the rotation term left free. This is load-bearing for the conclusion that the old populations have negligible rotation.","section":"Sections 3.2 and 4.2"},{"comment":"For populations in which vf is effectively zero, i and Theta enter the V02 proper-motion model only through the perspective/depth modulation of the COM motion. For the old populations this modulation has an amplitude of roughly (z/D0) times the ~420 km/s COM motion, i.e., ~0.1 mas/yr for a few-kpc depth, comparable to the quoted residual RMS (0.07-0.22 mas/yr) and to Gaia DR3 systematics. The 1-2 degree uncertainties quoted for i and Theta of YMS3, RGB, RC, and CLSO in Table 1 are therefore not evidently data-driven; a likelihood-flatness check over i and Theta for the non-rotating fits (e.g., with vf fixed to zero) should be shown. Because the claimed line-of-sight extension of ~30 kpc in Section 5.3 is a geometric projection of exactly these fitted angles, this issue propagates to the LOS-depth claim as well.","section":"Section 3.1, Table 1"},{"comment":"The claimed monotonic age trend in the viewing angles is not robust. In Table 1, i decreases from ~82 deg for YMS1 to ~58 deg for RGB and RC, but the Red Giants (D14) have i = 66.01 deg, breaking the monotonic decrease, and the increase in Theta from ~190 deg to 240 deg is largely driven by CLSO, which the text in Section 5.1 itself describes as sparse and centrally concentrated. The paper should present a robustness test omitting CLSO and the Red Giants, or explicitly demote the age-gradient morphology claim from a principal result to a tentative trend.","section":"Section 5.1, Table 1"},{"comment":"The inference that the young SMC is \"rotation-supported\" compares the fitted asymptotic velocity vf (49-89 km/s) with the residual rotational dispersion sigma_rot (9-11 km/s) computed about the same fitted rotation curve. Because the rotation curve is fit to the same data, the residual dispersion is minimized by construction, making this comparison circular as a disk-support diagnostic. Please compare the rotation amplitude with the total velocity dispersion of the young populations (including correlated errors), and report the model-comparison statistic between rotating and non-rotating fits for YMS1, YMS2, CLSY, and CLSI.","section":"Section 5.2"}],"minor_comments":[{"comment":"The rotation angles for the 3D views are inconsistent: the text gives R1 = 90 deg followed by R2 = 20 deg for panel (b), while the caption lists (R1,R2) = (90,20) for panel (b) and (R1,R2) = (90,120) for panel (c); the sentence \"R2 = 90\" in the text appears to be a typo and should be corrected.","section":"Section 5.3, Figure 10 caption"},{"comment":"The four residual anomalies (EA, SEA, SA, WA) are identified visually, without a quantitative significance threshold or a signal-to-noise map of the residuals. Since the SEA is highlighted as a new infalling feature, please add a significance map so the anomaly is not just an eyeball detection.","section":"Section 4.1, Figure 4"},{"comment":"The text states that Theta ranges from ~185 deg to ~202 deg for most populations, but Table 1 lists RGB with Theta = 207.66 deg, RC with 202.00 deg, and CLSO with 240.44 deg; these numbers should be reconciled in the text.","section":"Section 5.1"},{"comment":"The uniform priors for i, theta, vt, and vf are described but their numerical ranges are not given; since the reported errors are posterior credible intervals, the prior ranges should be explicitly listed.","section":"Section 3.2"}],"recommendation":"major_revision","confidential_remarks":"This is a competent analysis with a plausible and interesting central claim. The requested diagnostics—profile likelihoods for the non-rotating fits, formal model comparison for the rotation component, and a robustness test for the CLSO/Red Giant contribution to the age gradient—are standard and should be feasible with the existing MCMC machinery. I do not see grounds for rejection, but the paper should not be accepted until the identifiability issue for the old-population viewing angles is addressed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nIf the central claim holds, this paper settles the old spheroid-vs-disk debate by saying the SMC has been both: a rotating disk for populations younger than ~400 Myr, and a flattened, non-rotating ellipsoid for older stars. That is genuinely new, as are the nine-population unified V02 fit, the reported age trends in i and Theta, and the identification of the SEA residual feature. Credit is due: the COM proper motions agree with N21, D20, and Z18; the 3D red-giant fit recovers a systemic velocity very close to the fixed value; the young-population rotation signal is coherent; and the derived LOS extension is broadly consistent with previous distance work.\n\nThe soft spot is not minor, though. For the non-rotating populations (YMS3, RGB, RC, CLSO, and the red giants in the no-rotation variant), i and Theta are nearly unidentifiable from PM alone: with vf fixed to zero, the modeled PM is close to a uniform translation, with only a small perspective modulation that is of the same order as the residual RMS and Gaia DR3 systematics. Yet Table 1 quotes 1–2 degree uncertainties on those angles. The no-rotation model is chosen by “convergence,” not by model comparison, and no posterior or profile-likelihood diagnostics over i/Theta are shown for the non-rotating fits. So the age gradient in viewing angles, which carries the morphological-evolution claim, might be prior-driven. The trend also has internal cracks: CLSO’s Theta ~ 240 is admitted to be a sparseness artifact, and the red giants break the monotonic i decrease. The four anomalies are picked by eye without significance thresholds, the Rf discrepancy with D18 is acknowledged but unresolved, and correlated Gaia PM systematics are not modeled.\n\nI think the reader’s conditional verdict is right. A profile-likelihood or prior-sensitivity check over i/Theta for the non-rotating populations would decide whether the headline is real. If those angles are identifiable, this is a strong contribution; if not, the morphological-evolution part needs substantial revision. Either way, the paper contains useful data and worthwhile analysis, and it deserves a serious referee — I would send it out with a request for that identifiability check and proper significance estimates for the anomalies.","headline":"Worth careful refereeing — new age-stratified kinematic map of the SMC, but the headline morphological-age trend rests on viewing angles that may not be identified from proper motions alone.","tokens_in":20640,"tokens_out":3424,"would_cite":true,"duration_ms":38422,"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":"Modeling nine stellar populations shows the Small Magellanic Cloud evolved from a non-rotating ellipsoid to a rotating disk as stars aged, with inclination decreasing and position angle increasing with age.","keywords":["Small Magellanic Cloud","Gaia DR3 proper motions","galaxy kinematics","stellar populations","disk morphology","tidal interaction","Magellanic Clouds","Markov Chain Monte Carlo"],"falsifier":"Refit the old, non-rotating populations (YMS3, RGB, RC, CLSO, and red giants) with a model that sets rotation to zero and lets i and Theta float freely; if their posterior distributions are flat or the extra parameters do not improve the fit, then the reported age-gradient in i and Theta, and the line-of-sight depths built on them, are not supported by the data.","tokens_in":19278,"feed_emoji":"🌌","tokens_out":6870,"duration_ms":60547,"temperature":0.7,"pith_summary":"The paper claims that the Small Magellanic Cloud has undergone an age-dependent structural transition: the oldest stellar populations form a slowly rotating, flattened ellipsoid, while populations younger than about 400 million years form a rotation-supported, highly inclined disk. The evidence comes from fitting a parametric kinematic model to Gaia DR3 proper motions of nine populations, including main-sequence stars, red giants, red clump stars, and three age groups of star clusters. The fit yields an asymptotic rotation velocity of roughly 49-89 km/s with a scale radius of 6-9 kpc and a velocity dispersion of about 9-11 km/s for the young disk, and it reproduces a line-of-sight extension of about 30 kpc. If correct, this would mean the SMC is not simply one shape but has been reshaped over time, likely by the interaction with the Large Magellanic Cloud, and several residual proper-motion anomalies, including a newly identified infalling region, trace that tidal history.","feed_headline":"SMC aged from a shapeless ellipsoid into a spinning disk","feed_subtitle":"Age-graded Gaia kinematics show young SMC stars rotating in a thin disk while old stars form a flattened, non-rotating ellipsoid.","key_machinery":"The load-bearing object is a parametric disk-kinematic model, applied to the proper-motion field of each stellar population. The model assumes the observed proper motion is the sum of a common center-of-mass translation and an internal rotation component, with the rotation following an arctangent-like profile governed by an asymptotic velocity vf and a scale radius Rf; the model also treats the disk inclination i and position angle of the line of nodes as free parameters. Parameters are estimated with an MCMC sampler, and for each population the paper compares a rotating and a non-rotating model variant, choosing the one whose posterior converges. The same framework is extended to a 3D version by adding line-of-sight velocities of red giants. Residual maps, computed as observed minus modeled proper motion, are then used to identify spatially coherent kinematic anomalies.","core_discovery":"The central discovery is that the SMC's kinematics and morphology depend on stellar age in a coordinated way. For the young main-sequence populations (YMS1 and YMS2) and the young and intermediate-age clusters (CLSY and CLSI), the proper-motion field is well described by a rotating disk: the model converges on a non-zero asymptotic velocity vf in the range ~49-89 km/s, scale radius Rf ~6-9 kpc, and small rotational dispersion sigma_rot ~9-11 km/s, indicating a rotation-supported, thin, highly stretched disk. For the older populations (YMS3, RGB, RC, CLSO, and red giants with line-of-sight velocities), the rotation signal is negligible and the proper-motion field is consistent with a non-rotating, flattened ellipsoidal distribution. The viewing angles from these fits show a monotonic trend: inclination decreases from about 82 to 58 degrees and position angle increases from about 180 to 240 degrees with increasing age. The paper also finds four residual proper-motion anomalies—East, South East, South, and West—which it interprets as tidal and infall signatures of the recent LMC interaction.","pith_inferences":["If the non-rotating old populations cannot actually constrain i and Theta from proper motions alone, the reported age gradient in viewing angles may reflect model priors or the adopted disk geometry rather than a real geometric sequence; a 3D analysis with full radial velocities for all populations would settle this.","The newly identified South East Anomaly, if truly infalling, predicts that the southeastern stellar population should show a distinct line-of-sight velocity offset and possibly a different age/metallicity distribution compared with the disk model's expectation.","The modeled ~30 kpc line-of-sight extension is a directly testable prediction: comparing distance moduli of red clump stars or Cepheids across the body of the SMC against the model's LOS distance map would confirm or refute the inferred geometry without relying on proper-motion modeling."],"forward_implications":["The SMC's morphology is not fixed: it can be simultaneously an ellipsoid (in its old stars) and a rotating disk (in its young stars), and the monotonic age trends in i and Theta imply a structural transition over the past ~1-2 Gyr.","The young disk is rotation-supported and thin, with sigma_rot of only ~10 km/s, extending more than 20 kpc in the disk plane and producing a line-of-sight depth of up to ~30 kpc.","The old populations are pressure-supported, with negligible rotation, and their modeled line-of-sight extension of ~11 kpc each side confirms the large depth previously inferred from distance indicators.","The East, South East, South, and West proper-motion anomalies are coherent kinematic signatures of the LMC interaction; the newly identified South East Anomaly, interpreted as infalling gas and stars, implies recent accretion onto the SMC.","The base kinematic models provide a reference frame for identifying minority outlier populations and for calibrating numerical simulations of the Magellanic system."],"supporting_citations":[{"why":"Supplies the parametric proper-motion model that decomposes the observed PM into center-of-mass motion and internal rotation.","marker":"V02"},{"why":"Provides the rotation-velocity parametrization and the prior for the scale radius Rf used in the fits.","marker":"D18"},{"why":"Supplies the cleaned Gaia DR3 SMC catalog from which the stellar populations are selected.","marker":"J23"},{"why":"Provides the red-giant sample with line-of-sight velocities used for the 3D model and the comparison of systemic velocity.","marker":"D14"},{"why":"Supplies the cluster catalog with ages, metallicities, and extinctions used to split clusters into three age groups.","marker":"D24"},{"why":"Defines the color-magnitude polygon selections used to separate YMS and RGB populations.","marker":"G21"},{"why":"Provides previous estimates of inclination and position angle and the slow-rotation result for the old population that the paper compares against.","marker":"Z21"}],"fun_headline_variants":["SMC's spiral disk is written in young stars only","Age tells the SMC's tale: from blob to spinning disk","SMC's young stars spin, old stars stand still","A young SMC spins, an old SMC doesn't","SMC's disk only appears in its youngest stars"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the viewing angles (inclination and position angle) measured for the non-rotating old populations are actually determined by the data, even though a non-rotating proper-motion field depends only on the center-of-mass translation and not on those angles.","fun_headline_variants_meta":{"raw":{"variants":["SMC's spiral disk is written in young stars only","Age tells the SMC's tale: from blob to spinning disk","SMC's young stars spin, old stars stand still","A young SMC spins, an old SMC doesn't","SMC's disk only appears in its youngest stars"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001024,"raw_usage":{"total_tokens":4407,"prompt_tokens":1126,"completion_tokens":3281,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":742,"completion_tokens_details":{"reasoning_tokens":3198}},"tokens_in":742,"tokens_out":3281,"duration_ms":21980,"temperature":1.0,"reasoning_tokens":3198,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T22:44:58.033118+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Refit the old, non-rotating populations (YMS3, RGB, RC, CLSO, and red giants) with a model that sets rotation to zero and lets i and Theta float freely; if their posterior distributions are flat or the extra parameters do not improve the fit, then the reported age-gradient in i and Theta, and the line-of-sight depths built on them, are not supported by the data.","supporting_citations":[],"review_version":1}