{"id":"70f3d90b-15f3-4b8e-b90f-f88b40ab23e9","arxiv_id":"2505.22727","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"The 49-galaxy rotation curve sample shows cold gas tracing about 4 percent of the cosmic baryon fraction at all halo masses, while stellar mass scatters greatly and identifies baryon-deficient dwarfs that simulations fail to reproduce.","lead":"Astronomers measured how the visible mass of 49 disc galaxies relates to the mass of their dark matter halos, from small dwarfs to giant spirals. The results expose a wide spread in baryon retention and a possible population of dwarf galaxies with far fewer stars than current models predict.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The BDD classification rests on M200 values extrapolated from inner rotation curves; the VDM,max check inherits the same outer-profile assumption.","rationale":"The paper's central new claim, the existence of a population of baryon-deficient dwarfs, is plausible and presented with appropriate caveats. The sample is small (5 objects) and not volume-complete, but the claim is one of existence, not abundance. The most serious vulnerability is the inference of M200 for these dwarfs: the rotation curves are short relative to R200, and the adopted coreNFW profile fixes the outer slope to -3. The VDM,max cross-check is reassuring but not independent, since it is derived from the same fitted profile and peaks at radii beyond the data for these galaxies. A steeper outer profile would lower M200 and could move the BDDs back toward the expected SHMR. This is exactly the condition identified by the reader, and it can be settled by refitting with a profile that allows the outer slope to vary. The paper's other results (e.g., the gas-to-halo relation, the flaring analysis, the mass-to-light priors) are supported by the homogeneous sample and consistency with literature, and the weak-lensing comparison provides some independent support for the BDD population. Overall, the conditional verdict remains appropriate; the authors should add the profile-robustness test before the BDD population is taken at face value.","tokens_in":51453,"tokens_out":5920,"duration_ms":66470,"concrete_test":"Re-fit the five BDDs with the same rotation curves and priors but replacing the coreNFW profile with a generalized NFW whose outer slope beta is a free parameter, and also recompute VDM,max from the new posteriors. If the median log M200 of any BDD decreases by more than 0.5 dex or if log(M*/M200) comes within 1 dex of the Moster et al. (2010) prediction, the BDD classification is not robust. Additionally, check whether the outermost two measured rotation-curve points drive the high M200 by repeating the fit without them.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the existence of five baryon-deficient dwarfs (BDDs) with stellar masses 20-60 times below abundance-matching expectations (Sec. 6.1, Fig. 6). Their M200 values (log M200 ~ 10.5-11.7) come from coreNFW fits (Eq. 7) to rotation curves that extend only a few kiloparsecs, while R200 lies tens to hundreds of kiloparsecs out. The outer density profile is therefore an extrapolation. Appendix F attempts to address this by showing the BDDs also have high VDM,max, but VDM,max is computed from the same fitted coreNFW halo; for these systems the halo Vmax radius (~2.16 rs, typically >20 kpc for the BDDs) lies beyond the observed Hi extent, so VDM,max is not an independent check. If the true outer profile is steeper than NFW (e.g., a gNFW with beta > 3 or a truncated halo), the inferred M200 could be biased upward by more than the quoted uncertainties, moving these galaxies back toward the abundance-matching relation and erasing the BDD population. The analysis does not currently present a test that breaks the degeneracy between M200 and the assumed outer profile.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper assembles a curated sample of 49 nearby gas-rich disc galaxies with high-resolution Hi kinematics, NIR photometry, and standard-candle distances, and derives mass models via 3DBarolo kinematic modelling, galpynamics rotation curve decomposition, and nested-sampling Bayesian fits with self-consistently computed gas flaring. The dark matter haloes are modelled as coreNFW profiles with free mass-to-light ratios, halo mass, concentration, and core-size parameter. From these models the paper constructs the stellar, gas, and baryonic mass relations with halo mass, computes baryon retention fractions, and compares them with abundance matching, DarkLight, TNG50, and Simba. It reports two headline results: a population of five baryon-deficient dwarfs (BDDs) with stellar masses 20-60 times below abundance-matching expectations, and a c200-M200 relation with systematically low concentrations below M200 ~ 10^11 Msun and at the highest masses.","tokens_in":51744,"tokens_out":5374,"duration_ms":66573,"significance":"If the BDD population is real, it would substantially widen the observed scatter in baryon retention efficiency at fixed halo mass and would challenge current galaxy formation models and abundance-matching assumptions. The paper's strengths are its homogeneous and high-quality data treatment: beam smearing is handled with 3DBarolo, gas disc flaring is solved self-consistently in vertical hydrostatic equilibrium, distances are checked with TRGB/Cepheid measurements, and the kinematic profiles are made public. The comparison with independent weak-lensing measurements of the SHMR is a valuable cross-check. However, both headline claims depend on halo parameters inferred from a specific assumed dark matter profile, and the concentration claim is made with the DJ19 relation imposed as a Gaussian prior; these dependencies need to be explicitly tested before the conclusions can be fully accepted.","major_comments":[{"comment":"The VDM,max check in Appendix F does not break the outer-profile degeneracy on which the BDD classification rests. For the five BDDs, the observed rotation curves extend only a few kiloparsecs while R200 is tens to hundreds of kiloparsecs away, and VDM,max is computed from the same fitted coreNFW haloes; for these systems the radius at which VDM,max occurs (~2.16 rs) lies beyond the observed Hi extent, so VDM,max inherits the same extrapolation as M200. If the true outer profile is steeper than NFW (e.g., a gNFW with beta > 3 or a truncated halo), the inferred M200 could be biased upward by more than the quoted uncertainties, moving these galaxies back toward the abundance-matching relation. I request an explicit test with an alternative outer profile family, or an equivalent analysis that does not assume the same coreNFW extrapolation, and a demonstration of whether the five BDDs survive that test.","section":"Appendix F and Sec. 6.1"},{"comment":"The conclusion that c200 is systematically lower than the DJ19 relation at 10^10 < M200/Msun < 10^11 is drawn from posteriors obtained with a Gaussian prior centred exactly on the DJ19 c200-M200 relation with sigma = 0.16 dex. The paper itself states in Sec. 4.3 that for some galaxies the concentration remains unconstrained with a flat prior, which is the stated motivation for imposing this prior. With such a narrow prior, the low-concentration values are pulled toward the prior mean, so the 'substructure' in Fig. 10 is a residual from that pull rather than an independent measurement. Please re-fit the low-mass subsample with a flat or substantially wider c200 prior and report how many galaxies remain below the DJ19 relation by more than 1 sigma; this is necessary to support the claim of systematically low concentrations.","section":"Sec. 6.4 and Sec. 4.3"}],"minor_comments":[{"comment":"The text says the selection cuts result in 16 massive spiral galaxies, but only 15 galaxies are listed (NGC 0253 through NGC 5055); please correct the number or add the missing galaxy.","section":"Sec. 2.1"},{"comment":"In the paragraph on adiabatic contraction, 'the gap galaxies do not have high c200 values' appears to be a typo for 'these galaxies'; as written, the phrase is unclear.","section":"Appendix F"},{"comment":"The label 'Moster + 10' in the bottom panels is ambiguous; it should read 'Moster et al. (2010)' to match the reference style used elsewhere.","section":"Fig. 6"},{"comment":"The base of the logarithm in the luminosity-mass-to-light relation is not stated; given the use of log(Mstar/Msun) elsewhere, base 10 is presumably intended and should be stated explicitly.","section":"Eq. (C.1) and surrounding text"},{"comment":"The abstract describes BDDs as having stellar masses ~1-1.5 orders of magnitude lower than expected, while Sec. 6.1 quotes a factor of 20-60 (i.e., 1.3-1.8 dex); these statements should be harmonised.","section":"Sec. 6.1 and abstract"}],"recommendation":"major_revision","confidential_remarks":"The paper is a strong observational contribution and the BDD claim is potentially important, but the main new conclusions rest on halo parameters that depend on the assumed outer dark matter profile and on the imposed c200 prior. The revision should include the alternative profile test and the flat-prior concentration check described in the major comments. I do not see this as a rejectable circularity, but the current evidence is not yet sufficient for the strength of the claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Know this: read this paper for the sample and the scaling relations, not for the headline claims as stated. The compilation of 49 gas-rich discs with homogeneous kinematics, 22 new dwarf models, and publicly released data products is a real contribution. The self-consistent treatment of gas flaring and the care with beam smearing and distances are above average for this literature.\n\nWhat is actually new: the homogeneous multi-decade sample, the new 3DBarolo kinematic models, and the demonstration that the scatter in stellar and baryonic fractions at fixed halo mass is wider than TNG50, Simba, or DarkLight produce. The Mgas-M200 relation sitting at roughly 4% of the cosmic baryon fraction is a clean result that should be useful for a while.\n\nThe soft spots are concentrated in the two abstract-level claims. First, the c200-M200 'systematic deviation' below 1e11 is derived from posteriors that use the DJ19 relation as a Gaussian prior; all points sit within 2 sigma of that prior, and no group significance is computed. That is not enough to call it a systematic deviation. Second, the baryon-deficient dwarfs rest on M200 values extrapolated from inner rotation curves out to R200. Appendix F's VDM,max check is not independent for these systems because their Vmax radius lies beyond the Hi extent, so it inherits the same outer-profile assumption. The weak-lensing comparisons are suggestive but not individually constraining. The paper does label the BDDs tentative in the conclusions, which is the right tone; the abstract is too confident.\n\nMinor things: the sample is not volume-complete and the BDD threshold is post hoc. Both are acknowledged and neither is fatal, but they should be kept in mind when quoting the number of BDDs. The math and data handling are solid, and the citation pattern is appropriate—self-citations appear where the pipeline is genuinely reused.\n\nBottom line: a careful, honest paper that deserves a serious referee and would benefit from softened abstract claims and a robustness test with a steeper outer halo profile. I would cite it for the data and the scaling relations.","headline":"A genuinely useful sample paper whose two flashiest claims—baryon-deficient dwarfs and low-mass c200 deviations—are real possibilities but softer than the abstract makes them sound.","tokens_in":52320,"tokens_out":4136,"would_cite":true,"duration_ms":51502,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["98.62.-g","98.62.Dm","98.62.Gq"],"model":"deepseek-v4-flash","headline":"Rotation curves of 49 disc galaxies reveal five 'baryon-deficient dwarfs' with stellar masses 20–60 times below the abundance-matching expectation for their haloes, a population current models do not produce.","keywords":["galaxy-halo connection","dwarf galaxies","rotation curve decomposition","dark matter haloes","baryon retention fraction","abundance matching","coreNFW halo profile","Hi kinematics"],"falsifier":"Deep, wide-field Hi or CO mapping of the five baryon-deficient dwarfs (DDO 190, LVHIS 017, LVHIS 072, LVHIS 080, UGC 8508) to extend their rotation curves well beyond the current few-kiloparsec coverage. If the outer rotation curves turn over or flatten so that the fitted halo mass drops by the roughly 1 dex needed to reach the abundance-matching track, the BDD population dissolves into ordinary scatter; if the high circular speeds persist out to radii where the halo is directly probed, the population is confirmed.","tokens_in":51246,"feed_emoji":"🌌","tokens_out":22573,"duration_ms":206567,"temperature":0.7,"pith_summary":"Using rotation curves of 49 gas-rich disc galaxies spanning nearly six orders of magnitude in stellar mass, this paper builds a dynamical census of how baryons sit inside dark matter haloes. Its central finding is a population of five 'baryon-deficient dwarfs' (BDDs): galaxies whose stellar masses are roughly 20 to 60 times lower than abundance matching predicts at their inferred halo masses, with their cold gas likewise depleted relative to halo mass. Cold gas, by contrast, is tightly regulated: gas-rich galaxies hold about 4 per cent of the cosmic baryon fraction in gas across the whole halo mass range, while stellar content varies enormously at fixed halo mass, with some massive spirals retaining essentially all of their baryons. The paper argues the BDDs are unlikely to be artefacts, citing a check against an extrapolation-free halo speed and independent weak-lensing measurements, while noting that their halo masses carry large uncertainties. Because abundance matching, the DarkLight model, and the TNG50 and Simba simulations do not reproduce the BDDs or the full spread in baryon retention, the paper concludes that galaxy formation models need more stochastic baryon retention and more diverse feedback implementations.","feed_headline":"Five dwarf galaxies hold 20–60 times fewer stars than models expect","feed_subtitle":"A 49-galaxy rotation-curve census shows baryon retention varies far more than current simulations produce.","key_machinery":"The argument is carried by rotation curve decomposition: each galaxy's observed circular speed $V_{\\rm circ}$ is reproduced as the quadrature sum of stellar disc, bulge, gas (H\\,I and H$_2$), and dark matter contributions. The halo is described by the coreNFW profile, a Navarro-Frenk-White profile modified to allow a constant-density core of size $r_c = \\eta R_e$ with free parameters $M_{200}$, concentration $c_{200}$, and $\\eta$; the gas disc's flared scale height is solved simultaneously from vertical hydrostatic equilibrium, so halo parameters, stellar mass-to-light ratios, and gas thickness come out self-consistently. The quantity that exposes the BDDs and the scatter is the baryon retention fraction $\\tilde{f}_i = M_i / (f_{\\rm bar,cosmic} M_{200})$ evaluated for the stellar, gas, and baryonic components. The comparison baselines — the abundance-matching relation, the semi-empirical DarkLight model, and the TNG50 and Simba hydrodynamical simulations — supply what 'expected' means against which the BDDs are defined.","core_discovery":"The paper's central claim is that the galaxy-halo connection in nearby disc galaxies is not a single efficiency curve but a broad, partly stochastic band. At fixed halo mass $M_{200}$, the stellar retention fraction $\\tilde{f}_\\ast = M_\\ast / (f_{\\rm bar,cosmic} M_{200})$ ranges from massive spirals that approach or exceed the cosmic baryon fraction down to five baryon-deficient dwarfs (DDO 190, LVHIS 017, LVHIS 072, LVHIS 080, UGC 8508) whose stellar masses sit $\\sim$1–1.5 orders of magnitude below the abundance-matching expectation at $\\log(M_{200}/M_\\odot) \\sim 10.5$–$11.7$. The gas channel is far more regular: $M_{\\rm gas}$ tracks $M_{200}$ with a median of about 4 per cent of the cosmic baryon fraction, constant across four orders of magnitude in halo mass. The authors argue the BDDs are not modelling artefacts: the same deficit appears when halo mass is replaced by the directly measured maximum circular speed of the halo, and weak-lensing measurements land nearby; even so, they call the evidence tentative because the relevant $M_{200}$ values carry large uncertainties. None of the comparison models — abundance matching, DarkLight, TNG50, or Simba — reproduces the BDDs or the observed diversity of baryon fractions, which the paper reads as a need for more stochastic baryon retention and more diverse feedback implementations.","pith_inferences":["An extension the paper only gestures at: if the five BDDs really sit in overdense haloes, they should betray it through independent tracers such as unusually low globular-cluster counts or sparse satellite systems, giving a non-kinematic test of the classification.","The paper leaves the origin of the concentration deficit open; a testable next step is to compare $c_{200}$ for field dwarfs and group dwarfs at fixed $M_{200}$, which would separate the low-density-environment explanation from the feedback-driven core formation explanation.","If the BDD population is real, abundance-matching-based stellar mass functions need a much wider scatter in the stellar-mass–halo-mass relation at $M_{200} \\sim 10^{10.5}$–$10^{11.7}\\,M_\\odot$ than currently assumed, with consequences for inferred galaxy formation efficiencies.","The tightness of the gas relation combined with the wide stellar scatter suggests cold gas accretion is the regulated channel while star formation is the stochastic one; a concrete test is whether the scatter in $\\tilde{f}_\\ast$ at fixed $M_{200}$ tracks star-formation main-sequence offset or accretion history proxies in the same galaxies."],"forward_implications":["If the BDDs hold up, abundance matching misses a real population: a halo of $M_{200} \\sim 10^{11}\\,M_\\odot$ can host a galaxy 20–60 times poorer in stars than the mean relation predicts, so the scatter in stellar mass at fixed halo mass is a physical signal, not noise.","The gas-to-halo mass relation being a constant 4 per cent of the cosmic baryon fraction across four decades in halo mass is a sharp empirical target that any galaxy formation model should reproduce before its stellar feedback is tuned.","Baryon retention spans the full range from baryon-poor dwarfs to spirals with essentially no missing baryons, so models that produce a single baryon retention efficiency curve cannot match the data.","Below $M_{200} \\sim 10^{11}\\,M_\\odot$ the observed halo concentrations lie systematically below dark-matter-only predictions, and TNG50's concentrations are too high, indicating over-efficient adiabatic contraction in that simulation.","Baryon-rich systems appear at both low and high halo mass, meaning some dwarfs retain a substantial share of the cosmic baryon budget, extending the 'no missing baryons' result down from the massive spiral regime."],"supporting_citations":[{"why":"The abundance-matching relation that defines the expected stellar mass at fixed halo mass; the BDDs are measured as 20–60 times below this curve.","marker":"Moster et al. (2010)"},{"why":"Introduces the coreNFW halo profile used for every dark matter fit and the supernova-driven core-size considerations behind the η prior.","marker":"Read et al. (2016a)"},{"why":"The rotation-curve decomposition method with self-consistent gas disc flaring that this paper follows and extends to 22 newly modelled dwarfs.","marker":"Mancera Piña et al. (2022a)"},{"why":"Supplies kinematic models for 11 of the dwarf galaxies and the asymmetric-drift correction procedure used to compute Vcirc.","marker":"Iorio et al. (2017)"},{"why":"Provides the high-resolution Hi kinematic models for the 16 massive spirals that anchor the high-mass end of the sample.","marker":"Di Teodoro & Peek (2021)"},{"why":"The concentration–mass relation used as the Gaussian prior on c200; the paper's concentration results are measured as deviations from it.","marker":"Diemer & Joyce (2019)"},{"why":"The DarkLight semi-empirical model used as the main low-mass comparison; it shows the correct dwarf scatter direction but produces no BDDs.","marker":"Kim et al. (2024a)"},{"why":"The TNG50 hydrodynamical simulation whose baryon fractions and concentrations are compared against the data and found lacking.","marker":"Nelson et al. (2019)"},{"why":"The Simba simulation used for comparison; it produces a few BDD-like galaxies but with roughly twice too much gas to match.","marker":"Davé et al. (2019)"},{"why":"Weak-lensing measurement whose stellar-to-halo estimates land close to the observed BDDs, the independent support the paper cites for their reality.","marker":"Thornton et al. (2024)"}],"fun_headline_variants":["Five dwarf galaxies defy simulations with far fewer stars than predicted","Simulations miss five baryon-poor dwarfs and the spread in baryon fraction","Halo mass doesn't set stellar mass: baryon retention is stochastic","Gas follows halo mass, stars don't: a diversity puzzle for simulations","Baryon-poor dwarfs expose gaps in galaxy formation models"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The halo masses of the dwarfs, including the five baryon-deficient ones, are inferred by fitting a coreNFW profile to rotation curves that extend only a few kiloparsecs and extrapolating that profile out to R200, tens to hundreds of kiloparsecs away; if the true outer halo is less massive than assumed, the BDDs would relax back toward the normal relation.","fun_headline_variants_meta":{"raw":{"variants":["Five dwarf galaxies defy simulations with far fewer stars than predicted","Simulations miss five baryon-poor dwarfs and the spread in baryon fraction","Halo mass doesn't set stellar mass: baryon retention is stochastic","Gas follows halo mass, stars don't: a diversity puzzle for simulations","Baryon-poor dwarfs expose gaps in galaxy formation models"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001357,"raw_usage":{"total_tokens":5652,"prompt_tokens":1235,"completion_tokens":4417,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":851,"completion_tokens_details":{"reasoning_tokens":4322}},"tokens_in":851,"tokens_out":4417,"duration_ms":31623,"temperature":1.0,"reasoning_tokens":4322,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:03:01.927248+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Deep, wide-field Hi or CO mapping of the five baryon-deficient dwarfs (DDO 190, LVHIS 017, LVHIS 072, LVHIS 080, UGC 8508) to extend their rotation curves well beyond the current few-kiloparsec coverage. If the outer rotation curves turn over or flatten so that the fitted halo mass drops by the roughly 1 dex needed to reach the abundance-matching track, the BDD population dissolves into ordinary scatter; if the high circular speeds persist out to radii where the halo is directly probed, the population is confirmed.","supporting_citations":[{"cited_title":"H., et al","cited_arxiv_id":null,"evidence_quote":"Weak-lensing measurement whose stellar-to-halo estimates land close to the observed BDDs, the independent support the paper cites for their reality."}],"review_version":1}