{"id":"1b8da8e9-92c9-40fc-8b3d-b13d036de9db","arxiv_id":"2412.11966","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"A maximum-likelihood search of Pierre Auger data finds a 3.3 sigma (global) association between a cosmic-ray multiplet and the Sombrero Galaxy, supporting AGN jet and lobe acceleration of UHECRs.","lead":"Using 17 years of Pierre Auger Observatory data, the authors find a cluster of about 25 ultrahigh-energy cosmic rays whose arrival directions follow an energy-dependent pattern pointing at the Sombrero Galaxy. The association reaches 3.3 sigma globally, meaning it is suggestive evidence, not yet a discovery, that this nearby AGN with large radio lobes accelerates the highest-energy particles known.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Fisher concentration parameter in Eq. (4) is dimensionally wrong: kappa = 2/sqrt(sigma_ran^2 + sigma_ang^2) should be about 1/sigma^2 with sigma in radians, so the published source likelihood is nearly isotropic and the quoted TS/significance cannot be reproduced as written.","rationale":"The reader's weakest assumption (pure composition) is a reasonable physics uncertainty, but the manuscript contains a more concrete internal defect: the Fisher concentration parameter in Eq. (4) is written with incompatible dimensions. Eq. (3) defines sigma_ran as an rms deflection angle in degrees; the text then sets kappa = 2/sqrt(sigma_ran^2 + sigma_ang^2), which has units of inverse degrees and yields kappa ~ 0.7 for the best-fit A_ran = 2.7 deg. Such a small kappa makes the source term f_src almost uniform, so it cannot produce the sharp, energy-ordered multiplet claimed in Figure 1 or the TS = 37.6. The correct Fisher relation is kappa ~ 1/sigma^2 with sigma in radians; with sigma_ran = 2.7 deg and sigma_ang = 1 deg, the combined variance is about 0.0022 rad^2, giving kappa ~ 450. This is a roughly 600-fold difference, large enough to change every fitted parameter and significance. If this is only a typographical error, the paper still needs a corrected equation and ideally code release; as written, the method section is not self-consistent. This is why I recommend UNVERDICTED rather than REJECT: the reported astrophysical association might survive a corrected formula, but the present manuscript does not support it. The energy-error asymmetry in the significance estimation (data TS computed on reconstructed energies while background simulations include energy smearing) is an additional concern but secondary to the kappa defect. The CRPropa simulation in Figure 3 provides qualitative support for a magnetic-deflection pattern, but it does not validate the analytic likelihood used for the statistical claim.","tokens_in":14558,"tokens_out":10211,"duration_ms":91102,"concrete_test":"Independently re-derive Eq. (4): for a Fisher distribution, kappa = 1/sigma^2 with sigma in radians. Recompute the 6-parameter fit and TS_M104 for the PAO data using the corrected concentration, then rerun the background simulations. If the corrected global significance falls below 3 sigma or TS_M104 drops by more than about 5, the central claim is not supported by the published method. If the authors provide the likelihood code, run the same fit with the code's kappa; if the code uses kappa = 1/sigma^2, the paper's Methods need correction but the science may survive. Also run the published kappa formula on simulated isotropic skies; if it produces TS ~ 37.6 as often as claimed, the formula is internally consistent but the physical energy-ordered interpretation is unsupported.","verdict_should_be":"UNVERDICTED","load_bearing_attack":"The central likelihood (Eq. 4) depends on the Fisher concentration kappa defined in Methods as kappa = 2/sqrt(sigma_ran^2 + sigma_ang^2), with sigma_ran and sigma_ang quoted in degrees. For the Sombrero fit (A_ran = 2.7 deg, sigma_ang = 1 deg), this gives kappa ~ 0.71 at 100 EeV. A Fisher distribution with kappa ~ 0.7 is nearly isotropic (mean cos(theta) ~ coth(kappa) - 1/kappa ~ 0.09), so f_src would barely distinguish source events from background, and the tight energy-ordered multiplet claimed in Figure 1 could not produce the reported TS = 37.6. The correct small-angle relation is kappa = 1/sigma^2 with sigma in radians, giving kappa ~ 450 for the same parameters. Thus either the Methods formula is wrong and all reported TS values, best-fit n_s, source coordinates, and significances are not reproducible from the paper, or the code used a different formula and the paper misdescribes the method. Since every likelihood ratio in Eqs. (5)-(9) flows through Eq. (4), this is the most load-bearing issue in the manuscript; the composition assumption and energy-error treatment are secondary.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper searches for ultrahigh-energy cosmic-ray (UHECR) multiplets in 17 years of Pierre Auger Phase I data above 40 EeV, using a six-parameter maximum-likelihood method that fits source coordinates, regular and random magnetic deflection parameters, and the number of source events, against an isotropic background weighted by the Auger exposure. The authors report a best-fit multiplet of about 25 events with TS = 41.1 (global significance 3.7 sigma) and, after fixing the source to the Sombrero Galaxy (M104), TS = 37.6 with local and global significances of 4.5 sigma and 3.3 sigma, respectively. They argue that M104 is the only cataloged candidate within the 95% confidence region of the fitted source position and within a conservative 39.6 Mpc horizon, and they support the interpretation with CRPropa simulations of magnetic deflections in a Galactic magnetic field model.","tokens_in":14857,"tokens_out":7204,"duration_ms":68135,"significance":"If the statistical claim holds, this would be an important step in identifying a plausible nearby AGN accelerator of the highest-energy cosmic rays: the multiplet contains a 165 EeV event, is spatially associated with a known supermassive black hole with jets and lobes, and the energy-ordered pattern is consistent with magnetic deflection expectations. The paper has clear strengths: it uses public Auger data, estimates significances with background-only Monte Carlo simulations including exposure and energy uncertainties, applies a look-elsewhere penalty, tests robustness by removing the highest-energy event, and uses an independent CRPropa simulation. However, the central likelihood formula appears dimensionally wrong as written, so the reported TS values and significances cannot currently be reproduced from the manuscript; this issue must be resolved before the statistical claim can be assessed.","major_comments":[{"comment":"The definition of the Fisher concentration parameter is dimensionally inconsistent and appears to make the source likelihood nearly isotropic. The paper states kappa = 2/sqrt(sigma_ran^2 + sigma_ang^2), with sigma_ran and sigma_ang quoted in degrees. For the Sombrero fit (A_ran = 2.7 deg, sigma_ang = 1 deg, at 100 EeV) this gives kappa ~ 0.7, for which a Fisher distribution is only weakly peaked: f_src(0)/f_src(pi) = exp(2 kappa) ~ 4, so an event at the source direction is only mildly favored over an event on the opposite side of the sky. Such a likelihood cannot produce the tight energy-ordered multiplet of Figure 1 or the reported TS = 37.6. The correct small-angle relation is kappa ~ 1/sigma^2 with sigma in radians, giving kappa ~ 450 for the same parameters. Because every likelihood ratio in Eqs. (5)-(9), and hence every quoted TS value, best-fit n_s, source position, and significance, flows through Eq. (4), this is a load-bearing issue. The authors must either correct the definition and recompute all results, or explicitly report the formula actually used in the code and show that it matches the corrected expression.","section":"Methods, Eq. (4) and surrounding text"},{"comment":"The analysis assumes a pure composition for all UHECRs above 40 EeV. This assumption is load-bearing because the energy-dependent deflection law in Eqs. (2)-(3) is linear in Z, and the apparent source position S'(E) is defined using one Z. If the composition is mixed or energy-dependent, events of different Z at the same energy have different deflections, so the fitted multiplet shape and the best-fit source coordinates can be biased. The paper acknowledges this qualitatively but does not quantify the effect. Please repeat the 6-parameter fit under alternative composition assumptions (e.g., Z=1, Z=7, and Z=26, or a two-component mixture) and report the changes in the fitted source position, n_s, TS, and the resulting significances.","section":"Methods, Eqs. (2)-(3), pure-composition assumption"},{"comment":"The claim that the Sombrero Galaxy is 'the' source rather than 'the only cataloged candidate' depends on the completeness of the adopted catalogs. The paper correctly notes that the van Velzen et al. radio-galaxy sample excludes sources below 213 mJy, and the Updated Nearby Galaxy Catalog is not a complete flux-limited sample for faint jets and lobes. Therefore the statement that no other source lies within the 95% confidence region is only as strong as the completeness limits of the catalogs used. Please state the effective completeness limits for the relevant source classes, or explicitly soften the source-identification claim to 'the only known candidate within the adopted horizon and confidence region'.","section":"Methods, Source Catalogs, and Conclusions"}],"minor_comments":[{"comment":"The product in Eq. (7) runs from i=0 to n_tot; it should run from i=1 to n_tot.","section":"Methods, Eq. (7)"},{"comment":"There are several typographical issues: 'invariance' should likely be 'concentration parameter'; 'extragalacitc' should be 'extragalactic'; 'Jasson' should be 'Jansson'; 'Mattews' should be 'Matthews'; and 'PAOphase Idata' needs spacing.","section":"Throughout"},{"comment":"The text refers to 'Figure S3', but no supplementary figure S3 appears in the manuscript; please correct the reference.","section":"Author contributions"},{"comment":"The similarity between the observed and simulated multiplets is assessed only visually. A quantitative comparison (e.g., a two-sample test on angular distances from the source, or a likelihood ratio between the observed data and the simulated pattern) would make the supporting argument more convincing.","section":"Figure 3 and simulation comparison"},{"comment":"For the Sombrero rows in Table 2 the source coordinates are fixed at the known galaxy position, unlike the fitted coordinates in the best-fit column; this should be indicated explicitly to avoid confusion.","section":"Table 2"}],"recommendation":"major_revision","confidential_remarks":"The main issue is reproducibility of the likelihood calculation, not the astrophysical idea or the data handling. If Eq. (4) is a typographical error and the code uses kappa = 1/sigma^2 with sigma in radians, the rest of the analysis could be sound; if the code actually uses the printed formula, the reported significances are not meaningful. The revision should therefore center on correcting or clarifying the concentration parameter and rerunning the significance estimates. I would not recommend rejection at this stage because the error is potentially fixable within the manuscript's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi X,\n\nThe bottom line: this paper is not reproducible as written, and the problem is in the central likelihood. Eq. (4) defines the Fisher concentration as κ = 2/√(σ_ran²+σ_ang²), with σ in degrees. For their best-fit A_ran=2.7°, σ_ang=1°, that gives κ≈0.7, a nearly isotropic distribution. A Fisher with κ≈0.7 cannot produce a tight energy-ordered multiplet or a TS of 37.6. The correct small-angle relation is κ≈1/σ² with σ in radians, which gives κ≈400. So either the code used a different formula and the paper misdescribes the method, or the entire significance calculation is void. This is a load-bearing issue, not a footnote typo.\n\nThat said, the paper does several things well. It uses public PAO Phase I data, performs a full-sky 6-parameter maximum-likelihood search, evaluates significances with Monte Carlo background simulations, and checks that removing the 165 EeV event leaves TS≈34. The source-candidate search is careful, and the Sombrero’s jets/lobes make it a physically plausible accelerator. The extension of He et al. (2016) to include background and a global search is a reasonable step.\n\nThe secondary soft spots the reader flagged are real but minor by comparison: the pure-composition assumption is untested, and the catalog completeness within 39.6 Mpc is assumed. Those wouldn’t be fatal if the likelihood were right. But the κ issue must be resolved first. The authors need to release the likelihood code and show exactly how κ enters the fit, ideally with a corrected formula and a reproduced significance.\n\nMy verdict: this needs a serious referee, but the referee should spend most of their time on Eq. (4). If the code matches the text, the paper’s main claim collapses. If the text is wrong but the code is right, a corrected manuscript could be worth another look.","headline":"The reported 3.3σ Sombrero association rests on a likelihood equation with a dimensionally wrong Fisher concentration parameter; the paper needs a major correction or a code release before the result can be believed.","tokens_in":15473,"tokens_out":3641,"would_cite":false,"duration_ms":31408,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims a 25.7-event cosmic-ray multiplet above 40 EeV, found in 17 years of Pierre Auger Phase I data, is spatially associated with the Sombrero Galaxy at 4.5σ local and 3.3σ global significance, providing evidence that active…","keywords":["ultrahigh-energy cosmic rays","cosmic-ray multiplets","magnetic deflection","Pierre Auger Observatory","Sombrero Galaxy","active galactic nuclei","maximum likelihood","cosmic-ray sources"],"falsifier":"Compare the atomic-number composition of the 25 multiplet members: if they are not consistent with a single nucleus type, the assumed 1/E deflection scaling fails. A direct test is whether the multiplet reappears at the same sky position in the next independent dataset, for example future data from the same observatory or from the other large cosmic-ray detector; if the excess does not grow, the association would be a fluctuation.","tokens_in":14358,"feed_emoji":"🌌","tokens_out":6700,"duration_ms":56877,"temperature":0.7,"pith_summary":"The paper argues that the long-standing mystery of where ultrahigh-energy cosmic rays come from has a concrete answer in the local universe: a cluster of about 26 events with energies above 40 EeV, recorded over 17 years by the Pierre Auger Observatory, appears to be a magnetic-lens image of a single source. After fitting source position and magnetic deflection parameters simultaneously, the authors find the only plausible accelerator inside the allowed horizon is the Sombrero Galaxy, a nearby active galaxy with a billion-solar-mass black hole and radio jets and lobes. The association has a local significance of 4.5σ and a global, look-elsewhere-corrected significance of 3.3σ. If correct, this would be the first compelling identification of a specific active supermassive black hole as a source of the highest-energy particles in the Universe.","feed_headline":"Cosmic-ray cluster points to Sombrero Galaxy's black hole","feed_subtitle":"A 25.7-event multiplet above 40 EeV in 17 years of Auger data matches this nearby active galaxy at 3.3 sigma.","key_machinery":"The load-bearing object is a refined maximum-likelihood function that models each cosmic-ray event as either isotropic background or as a member of a point-source multiplet whose apparent direction is shifted by a regular magnetic deflection proportional to 100 EeV divided by energy and smeared by a random deflection with the same energy scaling, convolved with an angular scatter distribution for the random walk. The six free parameters are source right ascension and declination, the regular deflection amplitude, the random deflection amplitude, the regular-deflection direction angle, and the multiplet count; the test statistic compares this model against a background-only hypothesis. Because the method fits the deflection parameters from data rather than assuming a specific Galactic magnetic-field model, the source identification is tied to the energy ordering of the multiplet rather than to prior assumptions about field geometry. A conservative horizon distance for the 165 EeV event is used to restrict the source search to a sphere of radius 39.6 Mpc.","core_discovery":"Using a six-parameter maximum-likelihood search over the 1,387 Phase I events above 40 EeV, the paper finds a multiplet of 25.7 events whose apparent source sits near the Sombrero Galaxy, with the 165 EeV event arriving closest to the source direction and lower-energy members spread systematically farther away. The energy-dependent spread matches the expectation for a single cosmic-ray nucleus type being deflected by Galactic magnetic fields, with regular and random deflection amplitudes of about 20.7 degrees and 2.7 degrees at 100 EeV. Fixing the source to the Sombrero gives a test statistic of 37.6, a local p-value of 7×$10^{-6}$, and a global p-value of 9.8×$10^{-4}$; excluding the 165 EeV event lowers the source test statistic by only 3.6, so the association is not carried by that single extreme event. The paper concludes that active supermassive black holes can accelerate cosmic rays to energies beyond 100 EeV.","pith_inferences":["Inference: at 3.3σ global significance, the result is suggestive but not yet a claimed discovery; an independent dataset of comparable size is the natural next check.","Inference: the same likelihood machinery could be pointed at every active galaxy with kilo- or tens-of-kiloparsec radio lobes within roughly 40 Mpc, converting a single-source case into a population test.","Inference: the fitted small random deflection of about 2.7 degrees makes a testable prediction for Galactic magnetic-field models: along the line of sight to the Sombrero, the turbulent field must be weak enough to keep a 100 EeV nucleus within a few degrees.","Inference: if future composition data show the multiplet is actually a mixture of nucleus types, the pure-composition likelihood would need to be replaced by a rigidity-dependent model and the apparent source position could shift, providing a direct falsification test."],"forward_implications":["If the association holds, active galactic nuclei with powerful jets and lobes become the leading identified source class for the highest-energy cosmic rays.","The Sombrero's jet power of about 2.3×10^42 erg per second is more than sufficient: the derived UHECR luminosity of about 1.9×10^40 erg per second is roughly one percent of the jet power, comfortably within the energy budget.","The multiplet pattern predicts that higher-energy cosmic rays from the Sombrero arrive closer to its direction, so future events above 100 EeV should cluster around the fitted source position.","The method can be applied to other nearby active galaxies to decide whether Sombrero-like objects are common ultrahigh-energy cosmic-ray accelerators.","Improved energy and composition measurements from next-generation observatories should sharpen the source position and test the pure-composition assumption directly."],"supporting_citations":[{"why":"Supplies the 1387-event Phase I dataset above 40 EeV and the exposure function.","marker":"[6]"},{"why":"Provides the original maximum-likelihood multiplet search method that this work refines with global background fitting.","marker":"[12]"},{"why":"Basis for the pure-composition assumption, showing a trend toward less mixed, heavier compositions at the highest energies.","marker":"[16]"},{"why":"Catalog from which the 165 EeV event comes and its energy estimate.","marker":"[17]"},{"why":"Gives the conservative 39.6 Mpc horizon distance used to restrict candidate sources for the 165 EeV event.","marker":"[20]"},{"why":"Catalog of nearby radio galaxies with jets or lobes used to search for candidate sources.","marker":"[25]"},{"why":"Updated nearby-galaxy catalog used to expand the search for galaxies with jets or lobes.","marker":"[26]"},{"why":"Simulates Galactic magnetic deflections to show the observed multiplet pattern can be produced from the Sombrero.","marker":"[29]"},{"why":"Documents the 10-kiloparsec radio lobes of the Sombrero, supporting its ability to accelerate cosmic rays.","marker":"[39]"},{"why":"Provides the maximum-energy estimate based on shock size, field strength, and velocity that shows the Sombrero can reach beyond 100 EeV.","marker":"[43]"}],"fun_headline_variants":["Sombrero galaxy's black hole accelerates cosmic rays","Cosmic ray cluster traces to Sombrero's black hole","Highest-energy cosmic rays linked to Sombrero galaxy","Sombrero black hole named cosmic ray accelerator"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The analysis assumes every cosmic ray above 40 EeV has the same atomic number; if the true composition is a mixture or varies with energy, the energy-scaling pattern that ties the multiplet to the Sombrero could be distorted and the fitted position could shift.","fun_headline_variants_meta":{"raw":{"variants":["Sombrero galaxy's black hole accelerates cosmic rays","Cosmic ray cluster traces to Sombrero's black hole","Highest-energy cosmic rays linked to Sombrero galaxy","Sombrero black hole named cosmic ray accelerator"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000203,"raw_usage":{"total_tokens":1395,"prompt_tokens":961,"completion_tokens":434,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":577,"completion_tokens_details":{"reasoning_tokens":370}},"tokens_in":577,"tokens_out":434,"duration_ms":4722,"temperature":1.0,"reasoning_tokens":370,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T14:24:34.976237+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compare the atomic-number composition of the 25 multiplet members: if they are not consistent with a single nucleus type, the assumed 1/E deflection scaling fails. A direct test is whether the multiplet reappears at the same sky position in the next independent dataset, for example future data from the same observatory or from the other large cosmic-ray detector; if the excess does not grow, the association would be a fluctuation.","supporting_citations":[{"cited_title":"Monte Carlo Bayesian search for the plausible source of the Telescope Array hotspot","cited_arxiv_id":null,"evidence_quote":"Provides the original maximum-likelihood multiplet search method that this work refines with global background fitting."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Basis for the pure-composition assumption, showing a trend toward less mixed, heavier compositions at the highest energies."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Catalog from which the 165 EeV event comes and its energy estimate."},{"cited_title":"CHANG-ES. XXX. 10 kpc Radio Lobes in The Sombrero Galaxy","cited_arxiv_id":"2403.16682","evidence_quote":"Documents the 10-kiloparsec radio lobes of the Sombrero, supporting its ability to accelerate cosmic rays."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the maximum-energy estimate based on shock size, field strength, and velocity that shows the Sombrero can reach beyond 100 EeV."}],"review_version":1}