{"id":"5cc8cb9c-1fc5-4a7f-a3ea-2fb6948d6251","arxiv_id":"1908.01318","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"The Hercules moving group's angular momentum decreases toward the bar's major axis in Gaia DR2 data, matching the corotation-resonance prediction of a slowly rotating, dynamically old bar.","lead":"This letter uses Gaia DR2 data to show that the Hercules moving group's angular momentum changes with position around the Galaxy, as predicted if it is trapped at the bar's corotation resonance. The result favors a slowly rotating, old Galactic bar over a faster bar whose outer Lindblad resonance would leave the feature unchanged.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Uniqueness of the CR signature rests on the phase-mixed assumption; a bar younger than ~2 Gyr at the OLR can mimic the observed azimuthal Jphi slope.","rationale":"The reader and I identify the same weakest point. The paper's key claim is a smoking-gun test distinguishing CR from OLR by the azimuthal slope of the Hercules ridge. That test is only valid in the phase-mixed regime; the paper explicitly concedes that a young (<2 Gyr) bar at OLR can produce an azimuthal variation. Since the paper does not independently constrain the bar age, the data are consistent with both a dynamically old CR bar and a young OLR bar. This is an external degeneracy, not an internal inconsistency, and the authors are appropriately cautious in their wording ('in the case of a dynamically old bar'). Nevertheless, the abstract's phrasing 'distinctive prediction' and the conclusion that the data are 'as expected for the co-rotation resonance' are stronger than the evidence uniquely supports. The proposed simulation would directly test the alternative and settle whether the concern undercuts the claimed uniqueness. If the young-OLR model cannot reproduce the slope, the CR case is much stronger; if it can, the paper's conclusion should remain conditional or be softened. I therefore see no reason to change the reader's CONDITIONAL verdict.","tokens_in":8458,"tokens_out":4145,"duration_ms":41800,"concrete_test":"With the same Portail et al. (2017) potential and the same selection/analysis as in Sect. 3, build a model with Omega_b = 50 km s^-1 kpc^-1 (OLR slightly inside the Sun) and a bar that grows on timescales of ~0.5-1 Gyr, such that the disk remains far from phase-mixed (e.g., using the method of Minchev et al. 2010 or an N-body simulation with a growing bar). Measure the ridge slope in (phi, J_phi) over the azimuthal range -20 deg to 20 deg at R = R0. If the young-OLR model yields a nonzero slope comparable to the observed Hercules ridge (approximately -8 km s^-1 kpc deg^-1), the corotation identification is not unique; if the slope stays near zero, the phase-mixing concern is refuted.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central discriminant is that Hercules' angular momentum should vary strongly with azimuth at corotation but be nearly constant at the OLR (Sect. 2.2). This OLR prediction is derived from a trapped DF that is phase-mixed along pendulum angles: in Sect. 2.3 the librating DF is taken as f = <f0(Jf, Js(theta_p, Jp))>, i.e. averaged over theta_p. If the bar is younger than ~2 Gyr, orbits are not yet phase-mixed, and the paper itself notes (Sect. 3, last paragraph; Sect. 4) that an OLR origin can then produce a significant azimuthal variation of angular momentum, citing Minchev et al. (2010) and Trick et al. (2019b). Thus the observed slope of the Hercules ridge in Gaia DR2 does not uniquely select corotation over a young OLR bar; the diagnostic is conditional on the bar being dynamically old. This is acknowledged in the text but is still the load-bearing assumption on which the uniqueness claim depends.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a dynamical test to distinguish between two competing explanations of the Hercules moving group in Gaia DR2: trapping at the bar's co-rotation resonance (CR) of a dynamically old, large bar (pattern speed ~39 km/s/kpc), versus a faster bar whose outer Lindblad resonance (OLR) lies near the Sun. Using a pendulum approximation for resonant orbits (Sect. 2), the authors argue heuristically and then demonstrate in a Portail et al. (2017) based model that trapped CR stars show a strong variation of angular momentum J_phi with azimuth, whereas phase-mixed OLR-trapped stars show almost constant J_phi. They predict a ridge slope of about -8 km/s/kpc/deg in the (azimuth, J_phi) plane for CR. In Gaia DR2 with StarHorse distances, selecting a 400 pc annulus around the Sun, they plot the mean radial velocity in (phi, J_phi) and claim that the Hercules ridge follows the CR slope, while the OLR prediction (zero slope) is disfavored. They explicitly acknowledge that an OLR origin could reproduce the trend if the bar is younger than ~2 Gyr, i.e. if the trapped distribution is not yet phase-mixed.","tokens_in":8647,"tokens_out":4465,"duration_ms":49149,"significance":"If the result is robust, it would be a valuable step toward resolving the long-standing degeneracy between CR and OLR explanations of Hercules and would support a slow, large bar with a corotation radius beyond the Sun. The analytic pendulum heuristic is elegant and transparent, and the prediction is derived from an independent dynamical model (Portail et al. 2017; M19) rather than fitted to the Hercules ridge, so the test is not circular. The paper is honest about its limitations, including the acknowledged young-bar OLR degeneracy. However, the central comparison between model and data is purely visual, with no error bars or statistical measure, and the diagnostic power is conditional on the bar being dynamically old, which is not independently established here. These issues, while not fatal, mean that the paper's central claim is not yet fully supported.","major_comments":[{"comment":"The main observational evidence is a visual alignment of the Hercules ridge with an overplotted line of slope -8 km/s/kpc/deg, while the OLR zero-slope line is shown for contrast. There is no quantitative fit of the ridge slope, no error bars on the ridge location, and no significance statement that the zero-slope hypothesis is excluded. Because the entire Letter rests on this model-data comparison, the claim that 'the Hercules angular momentum changes significantly with azimuth' (Abstract) needs at least a simple quantitative assessment, such as fitting the ridge in (phi, J_phi) in both model and data and reporting the slope and its uncertainty from a bootstrap or similar procedure.","section":"Sect. 3, Fig. 2"},{"comment":"Systematic uncertainties in the transformation from observed parallax/proper-motion/RVS quantities to Galactocentric (R, phi, v_R, J_phi) are not quantified. The assumed values of R0 = 8.2 kpc, the solar peculiar motion, and v_c = 233.1 km/s enter directly into J_phi and phi; StarHorse distance uncertainties and the selection function of the RVS sample can bias the mean v_R in each bin. The authors should either propagate these uncertainties into the ridge location or provide a robustness check (e.g., varying R0 and v_c within current uncertainties) to show that the observed slope is not an artifact of these choices.","section":"Sect. 3"},{"comment":"The paper's abstract describes the CR signature as 'a distinctive prediction of such a model,' but the paper itself notes that an OLR origin with a bar younger than ~2 Gyr can produce a significant azimuthal variation of angular momentum, citing Minchev et al. (2010) and Trick et al. (2019b). Since the age of the bar is not constrained in this work, the test is conditional rather than distinctive. The conclusions in Sect. 4 are appropriately hedged ('reinforce the case... in the case of a dynamically old bar'), but the framing in the Introduction and Abstract overstates the discriminating power. I recommend softening the 'distinctive prediction' language or explicitly stating in the Abstract that the test discriminates only under the assumption of a dynamically old bar.","section":"Sect. 2.3 and Sect. 4"}],"minor_comments":[{"comment":"There is an inconsistency in the velocity grid bin size: the text states 'bin-size Delta v = 5 km/s' while the Fig. 1 caption writes 'Delta v = 10 km/s'. Please correct one of them.","section":"Sect. 2.3 and Fig. 1 caption"},{"comment":"Typo: 'whith' should be 'with' in the sentence '...will have evolved in magnitude by ~pi/8 = 22.5 deg, whith almost constant theta_s and J_s.'","section":"Sect. 2.2"},{"comment":"The units of the slope are given inconsistently: the caption says 'slope of -8 km/s/deg' while the text says '-8 km/s/kpc/deg'. Since the horizontal axis is azimuth in degrees and the vertical axis is J_phi in km/s*kpc, the correct unit is km/s*kpc/deg (or km/s/deg if J_phi is already divided by R0). Please unify.","section":"Fig. 1 caption"},{"comment":"The number of stars after quality cuts (6,350,087) and after the annulus cut (1,535,484) is given, but the initial RVS sample size (~7 million) is only mentioned in passing. A small clarification of the selection steps would help reproducibility.","section":"Sect. 3"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well within the scope of A&A Letters and represents a timely contribution to the Hercules debate. The main concern is the qualitative nature of the model-data comparison; adding a simple quantitative ridge fit and an uncertainty estimate would substantially strengthen the paper. The reliance on the authors' own previous model (Portail et al. 2017; M19) is not circular because the model is not fit to the Hercules feature, but it does mean the test is not fully independent. No concerns about citation practice or novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe thing to know: this is a short, honest letter that proposes a new azimuthal diagnostic for the origin of the Hercules moving group. The core claim is that if Hercules is trapped at the bar's corotation, its angular momentum should shift with Galactic azimuth (slope ~ -8 km/s/kpc/deg), while an OLR origin would give a nearly flat ridge. The Gaia DR2 ridge looks more like the CR prediction. That is a real, falsifiable prediction, and the qualitative match is worth taking seriously.\n\nWhat's actually new: the slope prediction in (phi, Jphi) space for CR versus OLR. Prior work (Friske & Schönrich, Trick et al.) looked at azimuthal variations but not this sharp discriminant. The heuristic pendulum reasoning in Sect. 2.2 is clear and physically transparent. The model calculation follows M19 and is not circular: the predicted slope comes from an independent Galaxy model (Portail et al. 2017), not from fitting the Hercules ridge. The paper also flags the secondary lower-Jphi component and the horn, showing they are not sweeping things under the rug.\n\nSoft spots, in order of importance. First, the data-model comparison is visual. There are no error bars, no quantitative fit, no assessment of how distance uncertainties (StarHorse) or assumed solar parameters would move the ridge. The paper says only the slope matters, but even the slope is drawn by eye on a binned map. A simple robustness check—varying the solar azimuth offset or distance scale, for instance—would strengthen the claim considerably. Second, the young-bar OLR degeneracy is real and acknowledged: if the bar is younger than ~2 Gyr and the trapped orbits are not phase-mixed, an OLR origin can also produce azimuthal variation. The paper's conclusion is therefore conditional on the bar being dynamically old. That is not fatal, but it means the abstract slightly overstates the uniqueness; the actual statement is 'consistent with CR if the bar is old.' Third, the azimuthal coverage is narrow (-20 to +20 deg), and the ridge could be affected by selection effects or spiral structure; the paper mentions this as future work.\n\nBottom line: a solid, readable letter that makes a genuine contribution—a new observable discriminant and a clean physical explanation. The central argument holds up as a plausibility argument, not a proof. It deserves peer review, and a more quantitative comparison would make it stronger. If you work on bar dynamics or moving groups, I'd bring it to the reading group and cite it. I'd have referees ask for a quantitative fit and an explicit treatment of the young-bar OLR case, but I would not desk reject.","headline":"A short, honest letter proposing a genuinely new azimuthal diagnostic for Hercules, but the data-model comparison is visual and the young-bar OLR degeneracy means the conclusion is conditional, not unique.","tokens_in":9224,"tokens_out":2371,"would_cite":true,"duration_ms":24320,"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":"Hercules' angular momentum decreases with azimuth, as expected for stars trapped at the co-rotation resonance of a slow, dynamically old Galactic bar.","keywords":["Galactic bar","Hercules moving group","co-rotation resonance","outer Lindblad resonance","stellar kinematics","Gaia DR2","action-angle variables","resonance trapping"],"falsifier":"Simulate a self-consistent bar younger than about 2 Gyr with pattern speed near $55\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}$ and check whether its outer-Lindblad ridge in the (azimuth, angular momentum) plane reproduces the observed $-8\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}\\,\\mathrm{deg}^{-1}$ slope; if it does, the claimed uniqueness of the co-rotation interpretation fails. On the data side, extending the Hercules ridge over a much wider azimuth range and finding a clear deviation from the predicted monotonic decline would disfavour the co-rotation model.","tokens_in":8270,"feed_emoji":"🌌","tokens_out":11906,"duration_ms":104176,"temperature":0.7,"pith_summary":"This paper claims that the conspicuous stellar velocity clump known as the Hercules moving group consists of stars trapped at the co-rotation resonance — the radius where stars circle the Galaxy at the same rate as the bar — of a slowly rotating, dynamically old Galactic bar, and that this origin can be distinguished from the competing fast-bar explanation by how the group's angular momentum changes with azimuth. In the co-rotation picture, the angular momentum of the Hercules ridge at the Sun's radius should fall by roughly $8\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}\\,\\mathrm{deg}^{-1}$ as the viewing azimuth approaches the bar's major axis; in a phase-mixed outer-Lindblad picture, the ridge should stay flat. The authors trace the mean radial velocity of stars from the second Gaia data release in a 400-pc annulus around the Sun as a function of angular momentum and azimuth, and find that the observed Hercules ridge follows the co-rotation slope. If this reading is right, it pins the Milky Way's bar to a pattern speed near $39\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}$ and rules out an outer-Lindblad origin for Hercules unless the bar is younger than about 2 Gyr.","feed_headline":"Gaia ridge ties Hercules to the slow bar's co-rotation","feed_subtitle":"Its angular momentum drops with azimuth exactly as predicted for co-rotation, not for a fast bar's outer resonance.","key_machinery":"The machinery is the pendulum reduction of stellar motion near a bar resonance. In slow and fast action-angle variables, with slow angle $\\theta_s = l\\theta_R + m(\\theta_\\phi - \\Omega_b t)$ and slow action $J_s = J_\\phi/m$, the near-resonant Hamiltonian becomes a pendulum whose librating solutions are the trapped orbits. At the $m=2$ co-rotation resonance ($l=0$), $\\theta_s = 2(\\theta_\\phi-\\Omega_b t)$, so the angular momentum $J_\\phi = 2J_s$ responds strongly to a change in azimuth toward the bar; at the outer Lindblad resonance ($l=1$), the radial angle varies faster than $J_s$, leaving $J_\\phi$ nearly constant. The trapped distribution function is built by averaging the unperturbed quasi-isothermal distribution over the pendulum angle, the step that assumes the bar has been present long enough for orbits to phase-mix.","core_discovery":"The central claim is that the key discriminant between different explanations of Hercules is geometric: resonant trapping zones leave different footprints in the plane of Galactocentric azimuth versus angular momentum measured at a fixed radius. For the $m=2$ co-rotation resonance of the bar, trapped orbits must change their angular momentum markedly with azimuth, producing a ridge with slope around $-8\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}\\,\\mathrm{deg}^{-1}$ in the model; for the $m=2$ outer Lindblad resonance of a phase-mixed old bar, the same ridge is essentially flat. The authors show, using the second Gaia data release with Bayesian photometric distances, that the Hercules ridge in mean radial velocity versus (azimuth, angular momentum) has the negative slope predicted by co-rotation, over azimuths roughly from $-20^\\circ$ to $+20^\\circ$. They interpret this as evidence that Hercules is built from stars co-rotating with a large, dynamically old bar, and that the only remaining route to an outer-Lindblad origin requires a bar younger than about 2 Gyr, before the trapped orbits have phase-mixed, i.e. spread evenly around their orbits.","pith_inferences":["Because the diagnostic compares slopes rather than absolute zero-points, it should transfer directly to other Galactocentric radii and to action-space ridges, allowing the bar's resonant structure to be mapped across the disc once surveys cover more azimuths.","A natural follow-up is to measure stellar ages inside the Hercules ridge: a phase-mixed co-rotation origin requires the trapped population to be old, whereas the young-bar loophole would predict a distinct age-dependent signature in the same ridge.","The acknowledged young-bar loophole implies that the test's uniqueness is conditional on bar age; combining the azimuthal slope with independent bar-age and pattern-speed measurements could tighten the conclusion further, while a wider azimuth range would test whether the slope remains linear or curves under the influence of spiral arms and satellite encounters."],"forward_implications":["A successful co-rotation identification fixes the bar's pattern speed near $39\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}$, placing the Sun just inside co-rotation and making the Milky Way's bar large and slow rather than small and fast.","The (azimuth, angular momentum) slope becomes a standard diagnostic: ridges that tilt with azimuth are resonance-trapped structures, while flat ridges are consistent with phase-mixed outer-Lindblad or linear-deformation features.","The argument demands a dynamically old bar, older than about 2 Gyr, so any independent evidence that the bar formed recently would undermine this particular interpretation.","If the same modelling is applied to other ridges, the 'horn' at high positive azimuth should show a steep, high-order-resonance behaviour, giving a further prediction that future data can test."],"supporting_citations":[{"why":"It supplies the Galactic bar potential and pattern speed used to compute the predicted co-rotation ridge.","marker":"Portail et al. (2017)"},{"why":"It establishes the perturbation-theory method and the assignment of Hercules to co-rotation that this paper extends to azimuthal variation.","marker":"M19"},{"why":"It provides the Gaia DR2 astrometry and radial velocities from which the observed ridge is built.","marker":"Gaia Collaboration et al. (2018)"},{"why":"It provides the Bayesian distances used to convert the sample to Galactocentric positions and velocities.","marker":"Anders et al. (2019)"},{"why":"It introduced the co-rotation-trapped orbit explanation for Hercules that the azimuthal slope is designed to test.","marker":"Pérez-Villegas et al. (2017)"},{"why":"It analysed ridges in the same azimuth–angular-momentum space and set the observational context for this measurement.","marker":"Friske & Schönrich (2019)"},{"why":"It is the source for the claim that a bar younger than about 2 Gyr can produce transient outer-Lindblad features with a similar azimuthal signature.","marker":"Trick et al. (2019b)"}],"fun_headline_variants":["Hercules' azimuthal angular momentum shift favors slow bar","Co-rotation resonance leaves mark on Hercules in Gaia DR2","Bar co-rotation explains Hercules' azimuthal momentum dip","Gaia shows Hercules stars co-rotate with slow bar","Hercules' momentum varies with azimuth, bar co-rotation wins"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes the bar's gravitational perturbation has been present for more than about 2 Gyr, so trapped stars are spread evenly around their resonant orbits; if the bar is younger, the competing fast-bar explanation can mimic the observed azimuthal trend.","fun_headline_variants_meta":{"raw":{"variants":["Hercules' azimuthal angular momentum shift favors slow bar","Co-rotation resonance leaves mark on Hercules in Gaia DR2","Bar co-rotation explains Hercules' azimuthal momentum dip","Gaia shows Hercules stars co-rotate with slow bar","Hercules' momentum varies with azimuth, bar co-rotation wins"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000846,"raw_usage":{"total_tokens":3711,"prompt_tokens":1005,"completion_tokens":2706,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":621,"completion_tokens_details":{"reasoning_tokens":2622}},"tokens_in":621,"tokens_out":2706,"duration_ms":19410,"temperature":1.0,"reasoning_tokens":2622,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:16:04.232758+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Simulate a self-consistent bar younger than about 2 Gyr with pattern speed near $55\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}$ and check whether its outer-Lindblad ridge in the (azimuth, angular momentum) plane reproduces the observed $-8\\,\\mathrm{km}\\,\\mathrm{s}^{-1}\\,\\mathrm{kpc}^{-1}\\,\\mathrm{deg}^{-1}$ slope; if it does, the claimed uniqueness of the co-rotation interpretation fails. On the data side, extending the Hercules ridge over a much wider azimuth range and finding a clear deviation from the predicted monotonic decline would disfavour the co-rotation model.","supporting_citations":[{"cited_title":"2017, MNRAS, 465, 1621 (P17)","cited_arxiv_id":null,"evidence_quote":"It supplies the Galactic bar potential and pattern speed used to compute the predicted co-rotation ridge."},{"cited_title":"More than just a wrinkle: A wave-like pattern in radial velocity vs. angular momentum from Gaia Data","cited_arxiv_id":"1902.09569","evidence_quote":"It analysed ridges in the same azimuth–angular-momentum space and set the observational context for this measurement."}],"review_version":1}