{"id":"b9349866-4fd4-492f-aaa4-eb9fb45e4029","arxiv_id":"1908.01358","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A z~2.2 massive galaxy shows a rotating multi-phase disk, with ionized gas at ~200 km/s at 13 kpc and molecular gas at ~320 km/s at 1 kpc, plus an AGN-driven outflow that may help shut down star formation.","lead":"This paper measures how gas moves inside a massive, distant galaxy that is forming stars and feeding a black hole, combining data from three telescopes. It finds a rotating disk that stretches far out and a fast outflow from the center, then asks whether the black hole is emptying the galaxy's gas tank.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Wind-vs-rotation degeneracy is under-tested: a smoothed M82 outflow pv diagram matches the data, yet no quantitative outflow model is fitted, so V_ion and the Tully-Fisher placement are not uniquely determined.","rationale":"The reader's weakest_assumption is exactly the wind-vs-rotation degeneracy, and I agree this is the most load-bearing issue. The paper is transparent about the degeneracy and even performs the M82 smoothing experiment, but the dismissal of the wind scenario is qualitative and relies on orientation priors rather than a quantitative model comparison. The apparent similarity of the smoothed M82 wind pv diagram to the GMASS pv diagrams is a genuine red flag; the counter-argument that the wind would have to be aligned with the stellar major axis is not a kinematic exclusion because wide-angle or misaligned outflows can produce projected gradients along arbitrary axes. The central-peak claim is additionally fragile: Section 4.2 explicitly reports that the inner two ring velocities are unreliable, and V_CO at R ~ 1 kpc is only ~1.3-2 sigma above V_ion at R ~ 13 kpc, so the 'rotation curve peaks centrally' phrasing overstates the evidence. These issues reinforce the CONDITIONAL verdict: the multi-phase disc and AGN-outflow detections are plausible, but the rotation velocities and the Tully-Fisher placement should be presented as conditional on the rotation interpretation. A quantitative wind-model comparison, or a high-fidelity CO velocity-field fit, would provide the missing test and, if successful, would strengthen the paper toward full acceptance.","tokens_in":27146,"tokens_out":9538,"duration_ms":110259,"concrete_test":"Fit the SINFONI [O III] and KMOS Halpha data-cubes with a parametric biconical outflow model (same PSF and noise) and compare its evidence against the tilted-ring disc model from 3DBarolo, e.g., via a Bayesian information criterion or nested-sampling evidence ratio. If the outflow model fits the pv diagrams as well as or better than the disc model, the rotation velocities and derived masses/Tully-Fisher placement are not uniquely supported. As a complementary check, run 3DBarolo on the ALMA CO(6-5) cube with the same tilted-ring setup: if the CO velocity field is a clean rotating disc with PA matching 94 degrees and V ~ 320 km/s at R ~ 1 kpc, the molecular-phase rotation interpretation is secure, but the ionized gas still requires the outflow test.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that GMASS 0953 is a rotation-dominated multi-phase disc with a centrally peaked rotation curve depends on interpreting the [O III] and Halpha velocity gradients as rotation. Section 4.3 concedes that after smoothing to GMASS resolution, the M82 wind pv diagram 'appears roughly similar' to the observed pv diagrams; the rejection of the wind rests only on the assertion that a wind would be perpendicular to the disc (along the minor axis) whereas the gradient follows the major axis. That argument is not kinematic: wide-angle or misaligned AGN winds can project gradients along the observed axis, and with a PSF of ~0.6-0.7 arcsec (only ~2 resolution elements across the inner disc) the two interpretations are degenerate. If the gradient is an outflow cone, V_ion = 203 km/s is not a rotation velocity; then the dynamical mass lower limit, the 'centrally peaked rotation curve' combining V_CO and V_ion, and the z = 0 baryonic Tully-Fisher placement in Section 5 all lose their basis. The paper's own test in Section 4.2 also flags the inner two rings as unreliable, so the central peak rests mainly on V_CO at R ~ 1 kpc being only ~1.3-2 sigma above V_ion at R ~ 13 kpc.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"GMASS 0953 is a massive (M* ~ 1.15e11 Msun), z = 2.226 star-forming galaxy with an obscured AGN. The paper combines GNIRS slit spectroscopy, SINFONI and KMOS integral-field data, and published ALMA CO(6-5) measurements to study the kinematics of the ionized and molecular gas. The [O III] and Halpha velocity gradients are modelled as a rotating disc with 3DBarolo, yielding a consistent outer rotation velocity V_ion = 203+17/-20 km/s at R ~ 13 kpc. Combined with V_CO = 320+92/-53 km/s at R ~ 1 kpc from Talia et al. (2018), the authors infer a multi-phase disc with a rotation curve that peaks in the centre, a dynamical mass lower limit of 1.3e11 Msun, and consistency with the z = 0 baryonic Tully-Fisher relation. They also identify an AGN-driven ionized outflow from a broad blueshifted [O III] component and evaluate outflow rates and depletion timescales, concluding that total gas depletion occurs on ~1e8 yr timescales while emphasizing the uncertainties in the AGN's role.","tokens_in":27411,"tokens_out":10328,"duration_ms":98555,"significance":"If the rotation interpretation is correct, this is one of the few multi-phase kinematic studies of a massive z ~ 2 galaxy, and it supports the existence of kinematically settled systems at early epochs. Strengths include independent 3DBarolo fits to two emission lines that give consistent outer rotation curves, a robust outflow detection in two independent datasets, and a generally transparent discussion of limitations, including the wind degeneracy and the unreliable inner rings. The principal weakness is that the strongest conclusions - the central rotation-curve peak, the dynamical mass, and the Tully-Fisher placement - depend on the rotation assumption, which is not uniquely constrained by the data, and on a V_CO-to-V_ion comparison whose significance is not quantified against the authors' own caveats. The paper is therefore a solid measurement and case study, but its headline claims are provisional.","major_comments":[{"comment":"The wind-versus-rotation degeneracy is acknowledged but never quantitatively tested. The only test offered is the qualitative M82 comparison in Fig. 8, which the authors state 'appears roughly similar' after smoothing, and the wind interpretation is then rejected because the wind would move perpendicular to the disc. With a PSF of 0.59-0.70 arcsec and a ring width of 0.35 arcsec, the inner few kpc are resolved with less than one PSF element, and wide-angle or inclined biconical outflows can project velocity gradients along the major axis. Because V_ion, the dynamical mass lower limit, and the Tully-Fisher placement in §5 all assume rotation, the authors should add a quantitative test (e.g., mock biconical-wind cubes observed with the same PSF, inclination, and velocity binning) or present the rotation-derived quantities explicitly as conditional on the assumed geometry.","section":"§4.3, §5"},{"comment":"The claim that the rotation curve peaks in the central regions rests on comparing V_CO = 320+92/-53 km/s (Talia et al. 2018) with V_ion = 203+17/-20 km/s at R ~ 13 kpc, a difference of roughly 1.2-2.1 sigma depending on how the asymmetric error bars are propagated. The authors themselves state in §4.2 that the first two rings of the ionized rotation curve are not constrained by the pv diagrams and that their errors are underestimated. The paper should provide an explicit significance estimate that includes the systematic uncertainties (fixed inclination, unresolved CO core, and ring correlation) and, if the significance remains low, present the centrally peaked rotation curve as tentative rather than a firm result.","section":"§4.2, §5"},{"comment":"The baryonic Tully-Fisher placement uses V_ion as the circular velocity without an asymmetric-drift correction, even though the velocity dispersion upper limits from the 3DBarolo fits are large (sigma < 160 km/s for [O III] and sigma < 90 km/s for Halpha; §4.2). If the true circular velocity is higher than V_ion, the galaxy's position on the z = 0 relation would shift substantially. The authors should show how the BTFR conclusion depends on V_c by considering the range of asymmetric-drift corrections allowed by the data, or restrict the claim to the measured rotation velocity.","section":"§5"}],"minor_comments":[{"comment":"The manuscript contains several typos that should be corrected: 'hypotesis' (§4), 'phoenomena' (§4.3), 'changelling' and 'blushifted' (Conclusions), and 'espress' and 'abudance' (Appendix D).","section":"Throughout"},{"comment":"Please specify how the M82 data were smoothed to 4.65 kpc (kernel shape and width) and whether the velocity binning was matched to the [O III]/Halpha data, so that the comparison is reproducible.","section":"Fig. 8"},{"comment":"The abstract uses 'may host an ionized disc' while the conclusions state 'we found evidence of a multi-phase disc'; the confidence level should be harmonized, especially if the wind degeneracy remains unresolved.","section":"Abstract/Conclusions"},{"comment":"In the sentence comparing V_CO with the Halpha inner value, give the radius and uncertainty of the Halpha inner value explicitly, since the reader would otherwise need to reconstruct it from Fig. 7.","section":"§5"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is careful and honest about many of its limitations, and the data are well suited to a case study; however, the headline kinematic conclusions are stated more strongly than the current tests support. I would welcome a revised version with a quantitative wind-model comparison and a significance estimate for the central peak. No concerns about overlap or citation practice; the use of Talia et al. (2018) CO data is transparent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this is a solid, honest single-object paper worth reading, but the boldest conclusion—a centrally peaked, multi-phase rotation curve—is not as secure as the abstract makes it sound.\n\nWhat is new: the independent 3DBarolo fits to [O III] and Halpha give consistent rotation curves, the explicit ionized/molecular comparison is useful, and the AGN-driven outflow is detected robustly in two independent datasets. The authors credit Wisnioski et al. 2018 and Talia et al. 2018 properly, so the citation pattern is not a problem. The data handling is transparent: they show the fits, admit the inner rings are unreliable, give upper limits on velocity dispersion, and spell out the factor-of-20 uncertainty in the outflow rate.\n\nThe main soft spot is the wind-vs-rotation degeneracy, which is handled qualitatively. Section 4.3 shows that the M82 H I wind pv diagram, smoothed to GMASS resolution, looks roughly similar to the observed gradients, then rejects the wind because it would move along the minor axis rather than the major axis. That is not a kinematic test: wide-angle or inclined winds can produce major-axis gradients, and with only about two resolution elements across the inner disc the two interpretations are genuinely degenerate. A simple outflow-cone model fit to the cube would have settled the matter or at least quantified the ambiguity. If the gradient is not rotation, then V_ion is not a rotation velocity, and the dynamical mass, the Tully-Fisher placement, and the centrally peaked narrative all lose their footing.\n\nThe central-peak claim is also weaker than the abstract suggests. The inner rings are explicitly distrusted, and V_CO at ~1 kpc is only about 2 sigma above the Halpha inner point, with large errors. I would frame that as consistent with a central peak, not a demonstration of one. A third, minor issue: several input galaxy properties come from unpublished work (Dalla Mura et al., in prep.), which should be replaced before publication.\n\nWho this is for: observers working on z~2 kinematics, multi-phase gas, and AGN feedback. It is also a good teaching case for 3DBarolo and for how not to over-interpret low-resolution velocity gradients.\n\nRecommendation: this deserves a serious referee. I would send it out. The authors need to soften the central-peak claim, add a quantitative wind test or clearly label the result as model-dependent, and replace the in-prep references. That is a revision, not a rejection.","headline":"A careful, transparent single-object study; the two-phase comparison is real, but the rotation-curve-peak and Tully-Fisher claims rest on a wind-vs-rotation degeneracy the authors test only qualitatively.","tokens_in":28018,"tokens_out":2511,"would_cite":true,"duration_ms":29127,"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":"GMASS 0953, a massive galaxy at $z\\simeq2.2$, is a multi-phase rotating disc: ionized gas rotates at $203$ km/s at 13 kpc, molecular gas rotates faster in the centre, and the baryonic mass sits within the local Tully-Fisher scatter.","keywords":["galaxies: active","galaxies: evolution","galaxies: kinematics and dynamics","galaxies: high-redshift","AGN-driven outflows","multi-phase galaxy discs","baryonic Tully-Fisher relation","integral field spectroscopy"],"falsifier":"A concrete test would be sub-kpc-resolution integral-field or ALMA observations of the inner few kiloparsecs: a rotating disc should show the classic spider-diagram velocity field, with line-of-sight velocities following $V_{\\rm rot}\\,\\sin i\\,\\cos\\theta$ around the fixed major axis, whereas an outflow cone should show a biconical pattern with the highest velocities along the minor axis and no matching high-velocity signature in the CO disc.","tokens_in":26940,"feed_emoji":"🌌","tokens_out":15138,"duration_ms":132884,"temperature":0.7,"pith_summary":"The paper sets out to show that GMASS 0953, a massive star-forming galaxy at $z\\simeq2.226$ that harbours an obscured active galactic nucleus, is a rotation-dominated system rather than a chaotic merger or a wind-dominated outflow. Combining near-infrared measurements of [O III] and H$\\alpha$ with existing CO(6-5) data, it builds a kinematic picture in which the ionized gas rotates at $V_{\\rm ion}=203^{+17}_{-20}\\,{\\rm km\\,s}^{-1}$ at $R\\simeq13$ kpc while the molecular gas rotates at $V_{\\rm CO}=320^{+92}_{-53}\\,{\\rm km\\,s}^{-1}$ on the $\\sim1$ kpc scale, so the rotation curve peaks in the central region. If true, this matters because it would show that at least some massive galaxies were already dynamically settled at the peak epoch of galaxy growth, and that their baryonic content already matches the local Tully-Fisher relation. The paper also reports a kpc-scale AGN-driven ionized outflow, seen as a broad blueshifted [O III] wing at $\\Delta v=-535\\pm152$ km/s, and argues that the combined gas depletion timescale is about $10^8$ yr, so the galaxy could soon become passive.","feed_headline":"One massive galaxy at z≈2 rotates as a single multi-phase disc","feed_subtitle":"Ionized and molecular gas spin together, placing the galaxy on the local Tully-Fisher relation.","key_machinery":"The central machinery is the tilted-ring disc model applied to integral-field data cubes through three-dimensional fitting, which compares an artificial rotating-disc cube with the observed cube and thereby corrects for beam smearing. Its kinematic backbone is the comparison of three emission-line tracers: [O III] $\\lambda5007$ and H$\\alpha$ trace the outer ionized disc out to 13 kpc, while CO(J=6-5) traces the unresolved molecular disc near 1 kpc, with all three gradients sharing a position angle aligned with the stellar major axis. The outflow-rate estimate rests on a simple conical-wind model that converts the luminosity of the blueshifted [O III] component into an expelled ionized-gas mass and divides it by a flow timescale set by the outflow radius and velocity.","core_discovery":"The central claim is that GMASS 0953 hosts a multi-phase rotating disc whose rotation curve rises toward the centre. The ionized gas, traced independently by [O III] $\\lambda5007$ and H$\\alpha$, shows a large-scale velocity gradient aligned with the stellar major axis; tilted-ring modelling gives a representative rotation velocity of $V_{\\rm ion}=203^{+17}_{-20}\\,{\\rm km\\,s}^{-1}$ at $R\\simeq13$ kpc. The molecular gas, traced by CO(J=6-5), shows the same spatial orientation but on a scale of $R\\simeq1$ kpc, with a higher central velocity $V_{\\rm CO}=320^{+92}_{-53}\\,{\\rm km\\,s}^{-1}$. The paper interprets this as evidence that the ionized and molecular phases belong to one disc, that the rotation curve peaks in the very central region, and that the dynamical mass within 13 kpc exceeds $1.3\\times10^{11}\\,M_{\\odot}$, consistent with the stellar mass. With baryonic mass $M_{\\rm b}\\simeq(1.47\\pm0.12)\\times10^{11}\\,M_{\\odot}$, the galaxy falls within the 0.11 dex scatter of the local baryonic Tully-Fisher relation, and a broad blueshifted [O III] wing at $\\Delta v=-535\\pm152$ km/s is presented as evidence of an AGN-driven outflow. The paper explicitly cautions that a galactic-scale wind could mimic the ionized velocity gradients and is not fully separable at the current spatial resolution, but argues that rotation is the more likely reading because the gradient follows the stellar major axis.","pith_inferences":["If the multi-phase disc is real, the combination of a high inner CO velocity and a lower outer ionized velocity implies a steeply falling rotation curve; a higher-resolution CO rotation curve would directly probe the bulge-to-disc transition and test the inferred central mass concentration.","The M82 comparison suggests that low-resolution, single-tracer kinematics of high-redshift galaxies may systematically misclassify outflow cones as discs, making multi-phase position-angle agreement a cheap discriminator for future surveys.","Because the outflow rate swings by about an order of magnitude depending on the dust-extinction correction, the prediction that GMASS 0953 quenches within about $10^8$ yr is not strongly tied to AGN feedback; if the unextincted estimate is correct, star formation alone would dominate the gas depletion.","If confirmed, the position of GMASS 0953 on the local baryonic Tully-Fisher relation would make it a useful benchmark for testing whether high-redshift kinematic samples suffer from selection biases toward the most settled galaxies."],"forward_implications":["Massive galaxies at $z\\simeq2$ can already be dynamically relaxed rotating discs, so disc settling does not have to wait until low redshift.","A rotation curve that peaks near 1 kpc and then declines implies a centrally concentrated mass distribution, consistent with the compact stellar structure of GMASS 0953.","Molecular and ionized gas trace the same disc, so high-resolution CO observations can recover central rotation velocities where seeing-limited ionized-gas measurements are unreliable.","With a total depletion timescale near $10^8$ yr, the galaxy would rapidly become passive if fresh gas accretion is shut off.","The kpc-scale blueshifted [O III] feature shows that AGN-driven gas ejection is operating in this source, even though its efficiency relative to star formation remains uncertain."],"supporting_citations":[{"why":"Supplies the CO(J=6-5) detection of the compact molecular disc and its rotation velocity $V_{\\rm CO}=320^{+92}_{-53}$ km/s, the inner anchor of the multi-phase rotation curve.","marker":"Talia et al. 2018"},{"why":"Provides the KMOS H$\\alpha$ data cube and an earlier kinematic model whose rotation curve the paper compares with its own.","marker":"Wisnioski et al. 2018"},{"why":"Supplies the SINFONI [O III] observations used to map the outer ionized disc.","marker":"Förster Schreiber et al. 2009"},{"why":"Provides the GNIRS slit spectrum in which the narrow and broad [O III] and H$\\alpha$ components are measured.","marker":"Kriek et al. 2008"},{"why":"Provides the three-dimensional fitting approach used to model the data cubes and correct for beam smearing.","marker":"Di Teodoro & Fraternali 2015"},{"why":"Provides the M82 H I wind observations whose downgraded position-velocity diagrams test the wind interpretation.","marker":"Martini et al. 2018"},{"why":"Defines the local baryonic Tully-Fisher relation and its scatter, against which GMASS 0953 is placed.","marker":"Lelli et al. 2016b"},{"why":"Supplies the conical-outflow geometry and emissivity assumptions used to derive the ionized outflow mass and rate.","marker":"Cano-Díaz et al. 2012"},{"why":"Provides the reference method and high-redshift sample of AGN-driven [O III] outflows that the paper follows.","marker":"Carniani et al. 2015"},{"why":"Defines the outflow maximum velocity $v_{\\rm of}={\\rm FWHM}/2+|\\Delta v|$ used in the outflow-rate calculation.","marker":"Rupke et al. 2005"}],"fun_headline_variants":["Massive z≈2 galaxy spins as one multi-phase disc","Ionized and molecular gas trace same rotating disc at z~2","Central rotation peak in z~2 galaxy: single disc","GMASS 0953: multi-phase disc with rising central rotation","AGN host at z≈2 shows rotation curve peaking at center"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the observed [O III] and H$\\alpha$ velocity gradients come from circular rotation in a tilted disc with inclination fixed to 75 degrees and position angle 94 degrees, rather than from a galactic-scale wind; at the available spatial resolution the two interpretations are not fully separable.","fun_headline_variants_meta":{"raw":{"variants":["Massive z≈2 galaxy spins as one multi-phase disc","Ionized and molecular gas trace same rotating disc at z~2","Central rotation peak in z~2 galaxy: single disc","GMASS 0953: multi-phase disc with rising central rotation","AGN host at z≈2 shows rotation curve peaking at center"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000707,"raw_usage":{"total_tokens":3372,"prompt_tokens":1317,"completion_tokens":2055,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":933,"completion_tokens_details":{"reasoning_tokens":1965}},"tokens_in":933,"tokens_out":2055,"duration_ms":16494,"temperature":1.0,"reasoning_tokens":1965,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:15:17.196545+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete test would be sub-kpc-resolution integral-field or ALMA observations of the inner few kiloparsecs: a rotating disc should show the classic spider-diagram velocity field, with line-of-sight velocities following $V_{\\rm rot}\\,\\sin i\\,\\cos\\theta$ around the fixed major axis, whereas an outflow cone should show a biconical pattern with the highest velocities along the minor axis and no matching high-velocity signature in the CO disc.","supporting_citations":[],"review_version":1}