{"id":"79ad3a8f-9637-4c92-8712-ab8cbe6076b5","arxiv_id":"1908.07937","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":1,"one_line_summary":"The paper claims a frequency-independent Planck temperature asymmetry in M104 consistent with a rotating cold baryonic halo, but the statistical case is not robust.","lead":"Planck data toward the Sombrero galaxy M104 show a microwave temperature asymmetry that the authors attribute to a rotating, dark baryonic halo. The claimed 0.2% chance probability relies on multiplying weakly significant, non-independent measurements, so the detection is not yet established.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Claimed <0.2% random probability is built from four nested/cumulative rings treated as independent plus a post hoc variant choice and unaccounted look-elsewhere; the significance does not survive a global trial correction.","rationale":"The paper's headline is a detection claim: a frequency-independent temperature asymmetry up to about 65 microK toward M104, interpreted as Doppler rotation of a baryonic dark halo, with a chance probability below 0.2%. The load-bearing element of that claim is the probability estimate. The frequency independence and the null average of the control fields are useful supporting checks, and I credit them as evidence that the measured asymmetry is not obviously instrumental. However, they cannot rescue the significance estimate if the trials are not handled correctly. The paper explicitly multiplies four per-ring probabilities 'assuming independent probabilities,' but the radial bins are nested, so this assumption is not justified and is likely false. The two variants, the four radii, and the fact that several galaxies were previously examined in the same series create a look-elsewhere surface whose size is not accounted for. The control fields are also not independent samples at the quoted spacing because adjacent fields are separated by less than the CMB correlation length. A Monte Carlo or a recomputation with disjoint annuli and pre-specified orientation would directly quantify the global false-positive rate. Since the reader flagged the same issue and the rejection rests on it, I see no reason to change the verdict: the paper as written does not establish the 0.2% significance and therefore does not establish the rotating baryonic dark halo claim.","tokens_in":9240,"tokens_out":5660,"duration_ms":58023,"concrete_test":"Recompute the Variant-1 asymmetry for the 360 control fields and for M104 using the same Planck maps, but define the four regions as disjoint annuli (0–0.2, 0.2–0.4, 0.4–0.6, and 0.6–1.0 degrees) and select the orientation after trying all four 90-degree rotations. Take the global p-value as the fraction of control fields whose maximum asymmetry over annuli and rotations exceeds the M104 maximum at any radius; if that global p-value is of order 1% or larger, the claimed <0.2% significance is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (abstract and Section 2.2) is that the north–south asymmetry in M104 has probability less than about 0.2% of being a random CMB fluctuation. This number rests on multiplying four per-ring probabilities stated as 0.30, 0.31, 0.11, and 0.26. The text itself says 'assuming independent probabilities,' but the radial regions R0.2, R0.4, R0.6, and R1.0 are nested apertures: the asymmetry is measured within each radius, so the four measurements are strongly correlated, not independent. The orientation used for Variant 1 (A1+A4 versus A2+A3) was not fixed a priori; the paper also examines Variant 2 and finds a less regular signal, and the same team previously searched many other galaxies (M31, Cen A, M82, M33, M81), so the effective number of trials is considerably larger than one. In addition, the 360 control fields are spaced by 1 degree in Galactic longitude at latitude b ≈ 51 degrees, which corresponds to angular separations of only about 0.6 degrees, comparable to the CMB correlation length; they therefore do not provide 360 independent null realizations. Even if the measured 25–65 microK asymmetry is real, and the frequency independence and control-field plots are supportive, the published p-value is not a valid global probability; with a conservative trial count the detection significance falls below the claimed level.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper analyzes Planck temperature maps towards the Sombrero galaxy (M104) and reports a north-south temperature asymmetry of up to ~65 microK within 0.6 degrees, which the authors claim is frequency-independent and consistent with the Doppler effect from a rotating baryonic dark halo. They quote a cumulative probability of less than 0.2% for the asymmetry being a random CMB fluctuation, based on per-ring probabilities in four nested apertures, and use the amplitude of the asymmetry to derive a dynamical mass of M104 out to ~100 kpc under an assumed cloud optical depth. The paper concludes that the detection confirms the authors' method for mapping dark halos in nearby edge-on galaxies.","tokens_in":9554,"tokens_out":6836,"duration_ms":60410,"significance":"If the claims were correct, this would be a novel probe of baryonic dark matter in galactic halos and an important extension of the kinetic Sunyaev-Zeldovich technique to galaxy scales. The paper uses publicly available Planck data, describes a control-field comparison, and is transparent about the independence assumption in the probability calculation, which is commendable. However, the statistical significance is not established as presented: the per-ring probabilities are not independent, the choice of the north-south variant is post hoc, and the control fields are not independent realizations. The mass estimate further depends on an assumed optical depth from the authors' own previous work. These issues undermine the central claims.","major_comments":[{"comment":"The cumulative probability of 1.8×10^-3 is obtained by multiplying the per-ring probabilities 0.30, 0.31, 0.11, and 0.26 for the nested apertures R0.2, R0.4, R0.6, and R1.0. Because these apertures are concentric and enclose one another, the same CMB fluctuation contributes to all four measurements; the rings are strongly correlated, not independent, as the text itself acknowledges with 'assuming independent probabilities.' A valid joint probability requires the covariance between rings or a simulation-based null test; without it, the quoted <0.2% is not a meaningful significance for the asymmetry.","section":"§2.2, probability calculation"},{"comment":"The paper tests two variants, Variant 1 (A1+A4 vs A2+A3) and Variant 2 (A1+A2 vs A3+A4), and reports the stronger signal for Variant 1. Since Variant 1 was not specified a priori, this is a post hoc selection that introduces a further trial factor. The same group has previously searched for microwave asymmetries toward M31, Cen A, M82, M33, and M81, so the effective number of independent trials is at least several times larger than one. Even if the per-ring probabilities were independent, the global probability of finding an asymmetry of the reported magnitude somewhere in this family of analyses is much larger than 0.2%.","section":"§2.2, Variant selection and look-elsewhere"},{"comment":"The 360 control fields are spaced by 1 degree in Galactic longitude at b≈51°, which corresponds to an angular separation of about 0.6 degrees, comparable to the CMB correlation length. These fields are therefore not independent realizations of the null hypothesis; the effective number of independent control regions is far smaller than 360. The statement that the control fields show an asymmetry 'consistent with zero' is based on the plotted means and errors, but the error bars themselves are likely correlated. A quantitative null test using a properly decorrelated set of control fields, or a simulation-based estimate of the covariance, is needed.","section":"§2.2, control fields"},{"comment":"The dynamical mass in Eq. (1) depends on the inverse square of the effective cloud optical depth τ_eff, for which the paper adopts 'a few ×10^-3' from Tahir et al. (2019), i.e., from the authors' own previous work. The quoted value M_dyn ≈ 3×10^12 M☉ is therefore not a model-independent measurement but a re-parameterization of the assumed τ_eff. Since the same unverified cloud model underlies both the mass estimate and the interpretation of the asymmetry, the agreement with earlier mass determinations is not an independent confirmation. The authors should present the mass as conditional on τ_eff and discuss the sensitivity of the conclusion to this parameter.","section":"§3, Eq. (1), mass estimate"},{"comment":"The claim of frequency independence of the asymmetry is supported only by visual inspection of Figs. 3 and 4. The four Planck bands have different beam sizes and noise levels, so a quantitative test (e.g., a chi-square consistency check across frequencies, or a fit with a common amplitude) is necessary to exclude frequency-dependent foreground contamination. The abstract's statement that the asymmetry is 'frequency-independent' is currently an assertion rather than a demonstrated result.","section":"§2.2, frequency independence"}],"minor_comments":[{"comment":"The reference to Gurzadyan (2018) as an unpublished Master's essay is not a standard citable source; the negative results for M63, M64, M65, M66 should be documented in a published form.","section":"References"},{"comment":"There are several typographical errors: 'non-Guassianity' should be 'non-Gaussianity'; 'FHWM' should be 'FWHM'; 'i = 830' should be 'i = 83°'; 'LiteBird' and 'Polarbear' should be 'LiteBIRD' and 'POLARBEAR'.","section":"Introduction and §2.1"},{"comment":"The units in Eq. (1) should be specified consistently (ΔT in µK, R in kpc), and τ_eff should be defined explicitly before its first use in the mass estimate.","section":"§3, Eq. (1)"},{"comment":"The paper would benefit from a figure showing the quadrant definitions (A1-A4) and the orientation of the galactic rotation axis, to allow the reader to assess the physical interpretation of the asymmetry.","section":"Fig. 2 and §2.2"}],"recommendation":"reject","confidential_remarks":"The central statistical analysis is not valid as presented, and the paper's discovery claim rests on that analysis. The authors may be encouraged to resubmit a version that uses a rigorous joint statistical test, corrects for the look-elsewhere effect, and calibrates τ_eff from independent data. As it stands, the central claim is not supported."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nYou want the punchline: this paper claims a <0.2% chance that the microwave temperature asymmetry toward M104 is a random CMB fluctuation, and that doesn't survive a careful look. The measurement itself may be real, but the statistics in the paper don't support the claim.\n\nWhat's actually new: the group applies its quadrant-asymmetry method to a new galaxy, M104, using public Planck data at 70/100/143/217 GHz plus SMICA. They see a north-south asymmetry that grows to about 65 microK within ~0.6 deg and is reasonably flat across frequencies. They also check 360 control fields and get null results, which is good practice. The paper is transparent about its procedure, and even discloses that they are 'assuming independent probabilities' when multiplying the per-ring values. That's where the trouble is.\n\nThe soft spot is central: the four per-ring probabilities (0.30, 0.31, 0.11, 0.26) come from nested apertures. The same pixels are inside multiple rings, so those numbers are strongly correlated, and multiplying them is not legitimate. The variant that shows the regular signal (Variant 1) was not fixed in advance; the paper also presents Variant 2 with a less regular pattern. Add in the fact that the group has already searched several other galaxies, and the effective number of trials is much larger than one. The 360 control fields are also not independent: spaced by 1 degree in longitude at b~51 deg, they are separated by roughly 0.6 deg, close to the CMB correlation length. So the quoted p-value is not a valid global probability. The frequency independence is asserted from visual inspection rather than a formal test. The mass estimate then plugs in an effective optical depth from the authors' own model (Tahir et al. 2019), which is a free parameter; that part is explicitly model-dependent.\n\nI want to be fair: the paper is not sloppy in the sense of hiding anything — the independence assumption is stated. But the statistics are load-bearing, and when you correct for correlated rings and look-elsewhere, the detection significance falls well below the claimed 0.2%. The measurement is an interesting data point, but the paper overinterprets it as a confirmed rotating baryonic halo.\n\nWho gets value from this: someone specifically tracking the missing-baryon budget or the CMB-halo mapping technique might want the measurement on file, but they should treat it as an upper limit at best, not a detection. I wouldn't cite it as evidence for a rotating baryonic halo. As for peer review: I'd send it to a referee if I wanted a careful statistical audit, but the core result would need to be re-derived with proper trial corrections before publication. As written, I'd reject.","headline":"A new CMB temperature asymmetry measurement toward M104, but the claimed 0.2% significance is built on correlated rings and a post hoc orientation choice, so the detection as stated does not hold up.","tokens_in":10090,"tokens_out":3877,"would_cite":false,"duration_ms":36041,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that Planck temperature maps of the Sombrero galaxy show a north-south asymmetry of up to about 65 microkelvin in the halo, frequency-independent and consistent with Doppler rotation of a baryonic halo component.","keywords":["M104 Sombrero galaxy","Planck CMB data","galactic halo rotation","Doppler effect","temperature asymmetry","baryonic dark matter","cosmic microwave background","edge-on galaxies"],"falsifier":"Recompute the Variant 1 asymmetry on the same Planck maps with the aperture placed at a fine grid of position angles and latitudes, then ask what fraction of random sky positions show a north-south contrast of $\\sim65\\,\\mu$K; if that fraction is not below $0.2\\%$, the reported cumulative probability is an artifact of orientation choice and the Doppler attribution loses its statistical basis.","tokens_in":9069,"feed_emoji":"🌀","tokens_out":13924,"duration_ms":124521,"temperature":0.7,"pith_summary":"Using Planck satellite temperature maps, this paper tries to show that the Sombrero galaxy (M104) has a halo-scale microwave asymmetry: one side of the galaxy is consistently hotter than the other by up to about $65\\,\\mu$K out to roughly $1^\\circ$ from the center. The asymmetry is nearly the same at 70, 100, 143, and 217 GHz, which the paper reads as the frequency independence expected of a Doppler shift from bulk rotation, and it is absent in 360 control fields with the same geometry. Treating the four radial rings as independent trials gives a cumulative probability below $0.2\\%$ that the north-south pattern is a random fluctuation of the cosmic microwave background. If this is right, microwave data can reveal and map rotating baryonic material in the halos of nearby edge-on galaxies, a complement to optical and radio probes.","feed_headline":"M104's halo is hotter on one side, hinting at rotation","feed_subtitle":"Planck maps show a frequency-independent asymmetry up to about 65 microkelvin, under 0.2% chance of being random.","key_machinery":"The key machinery is the quadrant temperature-asymmetry comparison. The Planck field around M104 is split into four quadrants A1–A4, and the mean temperature of one opposite pair is subtracted from the other in four concentric radial rings ($0.2^\\circ$, $0.4^\\circ$, $0.6^\\circ$, $1.0^\\circ$). Variant 1 compares the north-south quadrants (A1+A4 against A2+A3) and Variant 2 compares the east-west quadrants (A3+A4 against A1+A2); 360 control fields at the same galactic latitude, spaced one degree apart, give the random-fluctuation baseline. The Doppler identification is carried by the frequency independence of the asymmetry: a bulk rotation of emitting clouds shifts the CMB temperature by the same fractional amount at 70–217 GHz, whereas most foreground emission mechanisms vary with frequency. This isolates a rotation signal from CMB noise and foregrounds without needing a detailed emission model.","core_discovery":"The central claim is that the Variant 1 north-south temperature asymmetry toward M104—the mean temperature of quadrants A1+A4 minus A2+A3—is real, extends from about $25\\,\\mu$K at $0.2^\\circ$ to a peak of about $65\\,\\mu$K within $0.6^\\circ$, and then declines toward $1^\\circ$. The signal appears in all four Planck bands and in the foreground-reduced SMICA map, while 360 same-latitude control fields show no comparable asymmetry. The paper attributes this to the Doppler effect of a rotating halo component and quotes a cumulative probability of about $1.8\\times10^{-3}$ (below $0.2\\%$) for Variant 1, with the less regular east-west Variant 2 at about $3.1\\times10^{-2}$. Interpreted through the cold-cloud model, the asymmetry implies a lower bound on M104's dynamical mass of roughly $3\\times10^{12}\\,M_\\odot$ out to about 100 kpc, in line with independent estimates, and the more complex Variant 2 behavior is presented as evidence that the halo has several kinematic components.","pith_inferences":["The quoted cumulative probability multiplies four per-ring probabilities as if they were independent, but the rings are nested and CMB fluctuations are correlated across them; a covariance-aware calculation would probably raise the chance probability, and no statistical penalty has been applied for the fact that the north-south direction was chosen after inspecting the data.","Frequency independence by itself does not prove Doppler rotation: any foreground with a flat spectrum over 70–217 GHz and a spatial gradient across the galaxy would produce a similar asymmetry, so a 353 GHz check or cross-correlation with CO, HI, or X-ray maps would sharpen the attribution.","If M104's halo really holds rotating cold baryonic clouds at the implied mass scale, similar apertures should detect the effect in other edge-on galaxies, while the three galaxies in the earlier sample that showed no asymmetry suggest the phenomenon is not universal.","A decisive control experiment would place the aperture at many random orientations and latitudes and count how often a $\\sim65\\,\\mu$K north-south contrast appears; that distribution, not the quoted per-ring p-values, would settle whether the signal is a rare fluctuation."],"forward_implications":["A real Doppler asymmetry would mean the M104 halo contains coherently moving material out to roughly 100 kpc, well beyond the visible galaxy.","Under the cold-cloud interpretation, the implied dynamical mass of about $3\\times10^{12}\\,M_\\odot$ within 100 kpc would bring the halo mass into line with independent determinations based on tracers such as globular clusters and X-ray gas.","The same quadrant method, already applied to other nearby edge-on galaxies, would become a model-independent probe of halo kinematics that does not require knowing the emission mechanism in detail.","The frequency independence across the Planck bands and SMICA would favor Doppler rotation over dust, synchrotron, and thermal Sunyaev-Zeldovich foregrounds for the north-south signal.","The irregular east-west Variant 2 asymmetry implies the halo is not a single rotating disk, supporting the view that M104 is a composite system with multiple kinematic components."],"supporting_citations":[{"why":"Proposes the cold-baryon halo model in which rotation produces a microwave temperature asymmetry; supplies the physical interpretation tested here.","marker":"De Paolis et al. 1995c"},{"why":"Supplies the Planck 2015 CMB data products used for the maps.","marker":"Adam et al. 2016"},{"why":"Supplies the 70–217 GHz frequency maps in which the asymmetry is measured.","marker":"Ade et al. 2016"},{"why":"Provides the M104 disk rotation velocity ($376\\pm12$ km/s) anchoring the Doppler interpretation.","marker":"Jardel et al. 2011"},{"why":"Provides globular-cluster line-of-sight velocities used to test halo rotation and to identify the outermost clusters.","marker":"Bridges et al. 2007"},{"why":"Gives the effective cloud optical depth values needed to convert the measured asymmetry into a dynamical mass estimate.","marker":"Tahir et al. 2019"},{"why":"Supplies the visible mass of M104 against which the inferred dynamical mass is compared.","marker":"Tempel & Tenjes 2006"},{"why":"Establishes the quadrant temperature-asymmetry procedure on another nearby galaxy, providing the template applied to M104.","marker":"De Paolis et al. 2016"}],"fun_headline_variants":["Planck reveals spin in M104's halo","M104's dark halo rotates, Planck data suggest","Halo spin hint from M104's microwave asymmetry","Planck: M104 halo rotation odds 0.2%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the four per-ring chance probabilities can be treated as independent and that the north-south direction was fixed before inspecting the data; if the rings are correlated or the direction was chosen after the fact, the claimed <0.2% significance collapses.","fun_headline_variants_meta":{"raw":{"variants":["Planck reveals spin in M104's halo","M104's dark halo rotates, Planck data suggest","Halo spin hint from M104's microwave asymmetry","Planck: M104 halo rotation odds 0.2%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00075,"raw_usage":{"total_tokens":3346,"prompt_tokens":962,"completion_tokens":2384,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":2331}},"tokens_in":578,"tokens_out":2384,"duration_ms":16635,"temperature":1.0,"reasoning_tokens":2331,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:52:59.453150+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the Variant 1 asymmetry on the same Planck maps with the aperture placed at a fine grid of position angles and latitudes, then ask what fraction of random sky positions show a north-south contrast of $\\sim65\\,\\mu$K; if that fraction is not below $0.2\\%$, the reported cumulative probability is an artifact of orientation choice and the Doppler attribution loses its statistical basis.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Planck 2015 CMB data products used for the maps."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the 70–217 GHz frequency maps in which the asymmetry is measured."},{"cited_title":"& Drory, N","cited_arxiv_id":null,"evidence_quote":"Provides the M104 disk rotation velocity ($376\\pm12$ km/s) anchoring the Doppler interpretation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides globular-cluster line-of-sight velocities used to test halo rotation and to identify the outermost clusters."},{"cited_title":"& Nucita, A.A","cited_arxiv_id":null,"evidence_quote":"Gives the effective cloud optical depth values needed to convert the measured asymmetry into a dynamical mass estimate."},{"cited_title":"& Tenjes, P","cited_arxiv_id":null,"evidence_quote":"Supplies the visible mass of M104 against which the inferred dynamical mass is compared."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the quadrant temperature-asymmetry procedure on another nearby galaxy, providing the template applied to M104."}],"review_version":1}