{"id":"036d1bf1-974d-4266-8228-3a58208bab9e","arxiv_id":"2412.17714","paper_version":2,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A Monte Carlo population model predicts that AGN-disk GRBs are rarely detectable, and detectable events are concentrated at low redshift, high black hole mass, and large disk radius.","lead":"This paper simulates gamma-ray bursts that occur inside the dense gas disks around supermassive black holes. It finds that only a few percent would be visible if their light has to diffuse out, and those would come mostly from the outer parts of very massive disks.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Undiffused detection fractions are computed for θobs=0 and Eq. (10) lacks the Ω/4π beaming factor present in Eq. (9), so the headline 40–50% probability is an on-axis conditional number, not a per-event detection probability.","rationale":"The paper is a thoughtful first forward model, and it benefits from the use of published pAGN disk profiles and existing emission codes for high-density GRBs. The reader's identified assumption about progenitor placement (P(R) ∝ Σ, P(M) ∝ MΦ) is real and explicitly acknowledged by the authors; if migration traps dominate, the radial mix of detectable events could shift. I do not dispute that concern, but the single most load-bearing issue for the central claim is the viewing-angle normalization. The text sets θobs = 0 for all events, and the attenuation law for the undiffused case has no beaming factor, while the diffused law does. This makes the headline 40–50% probability not comparable with the diffused few-percent numbers and not directly usable as a population detection probability. A random observer sees only the beaming fraction of on-axis-detectable bursts; with θ_j = 5° the undiffused detectability drops to roughly 0.2%, which is below the diffused values reported for TQM. The mass/radius/redshift trends are based on the same on-axis selection, so they may persist, but their normalization and the relative prominence of the two scenarios would change. This warrants keeping the CONDITIONAL verdict, with the condition extended to include an explicit viewing-angle marginalization or a clear on-axis qualifier and beaming-corrected rates.","tokens_in":20253,"tokens_out":11179,"duration_ms":114462,"concrete_test":"Recompute the undiffused detection fractions from the same Monte Carlo draws after marginalizing over viewing angle: draw cosθ uniformly in [0,1] for each event, count a detection only when θ < θ_j (the assumed jet half-opening angle, 5° in the afterglow model) and the on-axis flux exceeds the Fermi/GBM threshold, and compare with Fig. 9 and Figs. 14–15. If the undiffused fraction drops by 1 − cosθ_j ≈ 0.004, the headline 40–50% is an on-axis upper limit; an analytical cross-check is to compare the uncorrected fraction multiplied by Ω/4π with the diffused fractions.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 3.1 step (v) fixes the viewing angle at θobs=0 for every simulated burst, and the detection fractions in Figs. 9–15 are built from fluxes computed with Eq. (10) for the undiffused channel. Equation (10), L_att = L0 exp(-τ), is an isotropic-equivalent on-axis luminosity: it contains no factor Ω/4π. Equation (9), by contrast, explicitly includes Ω/4π in L_diff. Since Eq. (11) takes max(L_diff, L_att), the undiffused detection fractions are conditional on the jet pointing directly at the observer, whereas the diffused fractions include the geometric probability of viewing the beamed jet. For a random orientation and a jet half-opening angle of 5°, the undiffused '40–50%' should be multiplied by (1−cosθ_j) ≈ 0.004, giving ≈ 0.2%, an order of magnitude below the diffused 2–3% for TQM. The paper does not state this 'on-axis' qualifier in the abstract or in the presentation of the detection probabilities. The qualitative trend toward low redshift, high SMBH mass, and outer radii may survive, but the central quantitative statement about the undiffused detection probability, and the relative detectability of the two scenarios, is not yet a population-normalized prediction.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a Monte Carlo population synthesis of long and short GRBs occurring in the disks of AGNs, combining a cosmological SMBH mass function, two disk models (Sirko-Goodman and Thompson-Quataert-Murray, implemented via pAGN), and high-density GRB emission models for the prompt and afterglow phases. It considers two extreme propagation scenarios: 'undiffused', in which the jet escapes through a low-opacity funnel, and 'diffused', in which radiation is Thomson-scattered and absorbed by the disk. The output is a set of detection probabilities in gamma-ray, X-ray, optical, and radio bands against representative instrument thresholds (Fermi, Chandra, HST, VLA), together with distributions of peak luminosity, afterglow flux, source location in the SMBH mass-radius plane, and T90 stretching. The central qualitative findings are that diffused bursts are observable in only a few percent of cases, preferentially from low redshifts, high SMBH masses, and outer disk radii, while undiffused bursts have much higher on-axis detection probabilities (~40-50%); the T90 distributions are stretched so that short GRBs appear long and long GRBs appear very long.","tokens_in":20582,"tokens_out":11577,"duration_ms":114036,"significance":"If the orientation issue identified below is addressed, the paper would provide a useful first end-to-end population model for AGN-disk GRBs, with falsifiable trends (T90 stretching, mass/radius/redshift selection, and band-dependent detectability) that can be tested with current and future facilities. The work is a forward model that does not fit to its own target predictions, and it makes helpful use of publicly available pAGN disk profiles and of previously published high-density emission calculations. The decision to present probability distributions rather than absolute rates is appropriate given the large uncertainty in the normalization of the AGN stellar population, and the explicit two-scenario treatment (undiffused vs. diffused) frames the problem usefully. The qualitative distinction between a rare, outer-disk, high-mass, low-redshift detectable population in the diffused case and a more accessible population in the undiffused case is physically plausible and worth communicating, pending the quantitative corrections discussed below.","major_comments":[{"comment":"The undiffused detection fractions quoted in the abstract (∼40-50%) are on-axis conditional probabilities, not population-averaged detection probabilities. The Monte Carlo fixes the viewing angle to θobs = 0 for every realization, and Eq. (10), L_att = L0 exp(-τ), contains no solid-angle factor, so these events are only seen if the jet happens to point at the observer. Averaging over a random orientation of the disk/jet axis with the top-hat jet half-opening angle of 5° assumed in Sec. 2.2.2 reduces the undiffused per-event probability by (1 - cos 5°) ≈ 3.8×10^-3, i.e., from ≈40-50% to ≈0.2%. The diffused channel, by contrast, already contains the Ω/4π factor in Eq. (9). The paper therefore does not currently provide a consistent population-normalized comparison of the two scenarios; the undiffused numbers should be relabeled as conditional or, preferably, the simulation should be rerun with θobs drawn from an isotropic distribution.","section":"Sec. 3.1(v), Eqs. (9)-(11), Figs. 9, 14, 15"},{"comment":"The assumed radial and mass distributions of progenitors, P(R) ∝ Σ(R) and P(M) ∝ MΦ(M,z), are load-bearing for the quantitative fractions in Figs. 9-15. The paper explicitly calls this a zeroth-order approximation and notes that migration traps could alter the radial distribution, but it does not explore the sensitivity of the reported percentages to plausible alternatives. A simple test (e.g., a uniform-in-log-radius prior or a migration-trap-concentrated prior) or an analytic scaling of the detection fraction with the prior would be needed to know whether the 'few percent' and '40-50%' numbers are robust. Without it, these numbers are predictions for one specific stellar-distribution model, not for the AGN-disk GRB population as a whole.","section":"Sec. 3.1(iii), Figs. 9-15"},{"comment":"The prompt-emission calculations fix Eiso = 10^53 erg for LGRBs and 10^51 erg for SGRBs, with Γ∞ = 100, for every realization. Observed GRB isotropic energies span several orders of magnitude, and the detection fraction is a strong function of the distance at which this fixed luminosity falls below the detector threshold. The quoted quantitative probabilities are therefore conditional on a single representative engine. Sampling Eiso from an observed distribution, or at least showing how the detection fractions vary with Eiso, would make the population predictions meaningful.","section":"Secs. 2.2.1 and 3.2, Fig. 3, Figs. 9-15"}],"minor_comments":[{"comment":"The caption reads 'LGBRs' but should read 'LGRBs'.","section":"Fig. 7 caption"},{"comment":"The phrase 'massive star collap' should be 'massive star collapse'.","section":"Sec. 3.1(v)"},{"comment":"The notation 'Δt−3' is ambiguous; it should be written as Δt_{-3} with a definition, e.g., Δt_{-3} = Δt / 10^{-3} s.","section":"Eq. (2)"},{"comment":"The text refers to observed T90 distributions but does not state whether the cosmological (1+z) time dilation has been applied to the simulated durations; a clarification or correction is needed.","section":"Sec. 3.2 and Fig. 16"},{"comment":"The caption says 'mass profile' but the quantity plotted is a mass function; the wording should be corrected.","section":"Fig. 1 caption"}],"recommendation":"major_revision","confidential_remarks":"The viewing-angle/beaming issue is the main quantitative blocker and should be fixed by a rerun or by clearly relabeling the undiffused fractions as on-axis conditional probabilities. The other two major comments are standard robustness requests for a population synthesis paper. I do not see a fundamental error that would require rejection; the modeling framework is reasonable and the paper is publishable after the requested revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First, the paper is a genuine first: a Monte Carlo population synthesis connecting AGN disk models to observable GRB detectability. It combines the Merloni-Heinz SMBH mass function, SG and TQM disk models, the HD-GRB prompt emission code of Lazzati et al. (2022), and the dense-medium afterglow code of Wang et al. (2022). The outputs — detection fractions in gamma-ray, X-ray, optical, and radio, plus T90 distributions and the mass/radius/redshift structure of the detectable population — are new. The qualitative trends are robust: detectable bursts are rare in the diffused case, they come preferentially from low redshifts, high SMBH masses, and outer disk radii, and the high-density environment stretches T90. These conclusions follow from the density and opacity scalings and are consistent across SG and TQM. The paper is also honest about its assumptions and does not fit anything to its own predictions, so the circularity burden is low.\n\nThe load-bearing soft spot is the undiffused detection probability. The 40–50% numbers are computed with θobs fixed at 0, and Eq. (10) for L_att has no Ω/4π factor, unlike Eq. (9). That makes them conditional on the jet pointing at the observer, not per-event probabilities. For a random orientation and a 5° jet half-opening angle, the geometric factor is (1 − cos 5°) ≈ 0.004, so the orientation-averaged undiffused probability is about 0.2%, an order of magnitude below the diffused TQM value of 2–3%. The abstract and the results section do not state this qualifier, so the relative detectability of the two scenarios is misrepresented. This is fixable: report both the on-axis conditional numbers and the beaming-corrected per-event numbers, or state the conditional nature up front.\n\nMinor issues: no parameter scans, fixed Eiso and microphysical parameters, and the P(R) ∝ Σ(R) progenitor placement is a zeroth-order approximation, which the authors acknowledge. No code is released, which would help but is not fatal.\n\nOverall, this paper deserves a serious referee. The beaming correction changes the quantitative headline but not the core forward-modeling framework or the qualitative results. I would send it to review with a request to address the beaming factor. I'd also bring it to a reading group — it is a good example of a population synthesis that is transparent about its assumptions, and the on-axis/off-axis distinction is a useful lesson.","headline":"First population synthesis of AGN-disk GRBs with useful predictions, but the undiffused 40–50% detection fraction is an on-axis conditional number that needs a beaming correction before the paper's central comparison holds.","tokens_in":21112,"tokens_out":5129,"would_cite":true,"duration_ms":47970,"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":"Gamma-ray bursts from AGN disks are mostly invisible: at most a few percent are detectable if their light diffuses, and the survivors are low-redshift, outer-disk events around very massive black holes.","keywords":["gamma-ray bursts","active galactic nuclei","accretion disks","high-density GRBs","cosmological population","gravitational-wave counterparts","Monte Carlo simulation","synchrotron self-absorption"],"falsifier":"Find and localize AGN-disk GRB candidates in a large sample: the model predicts detectable events overwhelmingly at low redshift (z<1-ish), around SMBHs with M>$10^{7}$.5 solar masses, and from outer disk radii, so a well-localized burst at high redshift or in a low-mass AGN - or with a bright radio afterglow, which self-absorption should suppress - would contradict the central prediction.","tokens_in":2182,"feed_emoji":"💥","tokens_out":4438,"duration_ms":112970,"temperature":0.7,"pith_summary":"The paper turns the idea that gamma-ray bursts can be born inside active galactic nucleus disks into a testable population prediction: how many such bursts would actually be detected, at what wavelengths, and from which host disks. The authors draw a cosmological population of AGN disks from a supermassive black hole mass function, place long and short GRBs within two accretion disk models, and pass the resulting prompt and afterglow emission through either free escape or Thomson diffusion in the disk medium. In the diffused case, at most a few percent of AGN-disk GRBs would be detectable, and those would preferentially come from low redshifts, the outer disk, and black holes above $10^{7}$.5 solar masses; in the undiffused case, where the progenitor's winds have cleared a funnel, the detection probability rises to roughly 40-50 percent. If correct, AGN-disk GRBs are a rare but informative population that could calibrate the stellar and compact-object content of accretion disks and the AGN contribution to gravitational-wave mergers.","feed_headline":"Detectable AGN-disk GRBs are rare, low-z, outer-disk events","feed_subtitle":"If burst light must diffuse out, under a few percent are visible; a wind-cleared funnel raises the odds to about half.","key_machinery":"The machinery is a Monte Carlo population synthesis whose pivotal quantity is the Thomson optical depth $\\tau(R_{\\rm em},R_{\\rm GRB})$ from the burst location to the disk surface. Host AGNs are drawn from a supermassive black hole mass function; disk density and scale height come from the two disk models; each GRB is placed at a radius weighted by disk surface density; the prompt emission is computed with the high-density GRB prescription, in which the external shock forms before internal shocks when $n>6\\times10^{6}\\,E_{52}\\Gamma_{\\infty,2}^{-8}\\Delta t_{-3}\\,\\mathrm{cm^{-3}}$; and the afterglow is computed with a Monte Carlo synchrotron model including self-absorption. The emission radius is the external shock radius $R_{\\rm ES}=\\max(R_{\\rm ES}^{\\rm thin},R_{\\rm ES}^{\\rm thick})$, and the dichotomy between scenarios is set by $\\tau$: for $\\tau\\lesssim1$ radiation escapes undiffused, while for $\\tau\\gtrsim1$ it emerges on the diffusion timescale $t_{\\rm diff}\\simeq[H(R)-R_{\\rm em}]\\tau/c$ with the reduced luminosity $L_{\\rm diff}\\sim L_0(t_0/t_{\\rm diff})(\\Omega/4\\pi)$.","core_discovery":"In the paper's own terms, the central result is a selection function: the AGN disk itself decides which GRBs are visible. When the burst radiation must diffuse through the disk (the fully diffused scenario), the expected detectable fraction is at most a few percent - below one percent in the SG disk model and 2-3 percent in the TQM model for prompt gamma-rays - and the survivors are concentrated at low redshift, at outer radii $R\\sim [10^6,10^7]R_g$, and around supermassive black holes with $M\\gtrsim 10^{7.5}M_\\odot$. When radiation escapes through a low-opacity funnel (the undiffused scenario), the detection probability is 38-53 percent depending on burst type and disk model, with a noticeable additional contribution from intermediate disk radii around lower-mass black holes. In both scenarios radio afterglows are essentially invisible because synchrotron self-absorption suppresses low frequencies, while the most promising channels are prompt gamma-rays in the undiffused case and X-ray afterglows in the diffused case. The duration $T_{90}$ is stretched in the dense environment, so short GRBs can appear as long bursts and long bursts as very long ones.","pith_inferences":["The same opacity filter implies that any electromagnetic counterpart to a gravitational-wave merger inside an AGN disk would be strongly biased toward low redshift, massive SMBHs, and outer disk radii; a systematic search along those lines may be more fruitful than an all-sky blind search.","If the undiffused case is the one realized in nature, current Fermi-era data should already contain a measurable AGN-disk GRB population, and the observed rate could be inverted to place upper limits on the star-formation and merger rates inside AGN disks.","The predicted near-invisibility of radio afterglows is a sharp testable corollary for wide-field radio facilities; detecting such an afterglow would force the disk models or the self-absorption treatment to be revised.","The duration-stretching effect suggests that some GRBs currently classified as long with no supernova and an AGN host may be AGN-disk bursts, a classification that follow-up X-ray and optical observations can test."],"forward_implications":["A fully diffused AGN-disk GRB population would be a rare source class: prompt gamma-ray detection probabilities are below 1-3 percent, so detections should be uncommon even if AGN disks produce a sizable burst rate.","If the undiffused funnel scenario operates, roughly 40-50 percent of AGN-disk GRBs would be detectable in prompt gamma-rays, making the population accessible to current instruments.","Radio afterglows from AGN-disk GRBs are predicted to be effectively undetectable with VLA-class sensitivity, so a detected radio counterpart would be difficult to reconcile with the standard self-absorption picture.","Observed durations are stretched in dense disks: short GRBs would be misclassified as long and long GRBs as very long, so AGN-disk GRBs should be searched for among long and ultra-long bursts with AGN host associations.","The marked differences between the SG and TQM disk predictions mean that even a few detected AGN-disk GRBs could discriminate between competing AGN disk structures."],"supporting_citations":[{"why":"Supplies the redshift- and mass-dependent SMBH mass function that seeds the cosmological Monte Carlo draws.","marker":"Merloni & Heinz (2008)"},{"why":"Defines the alpha-disk framework whose parameters underlie the disk structure calculations.","marker":"Shakura & Sunyaev (1973)"},{"why":"One of the two disk models used, providing outer-disk gravitational stability and density profiles.","marker":"Sirko & Goodman (2003)"},{"why":"The second disk model, with self-regulated star formation, providing TQM density profiles.","marker":"Thompson et al. (2005)"},{"why":"Provides the numerical disk solutions used to assign density and scale height across SMBH mass and radius.","marker":"Gangardt et al. (2024)"},{"why":"Defines the high-density GRB regime and supplies the prompt emission light-curve model in dense media.","marker":"Lazzati et al. (2022)"},{"why":"Supplies the Monte Carlo afterglow model including synchrotron self-absorption in dense media.","marker":"Wang et al. (2022)"},{"why":"Establishes how the disk photosphere and emission radius depend on burst location, used for the optical depth integral.","marker":"Perna et al. (2021a)"},{"why":"Provides the Fermi/GBM duration distributions from which the engine durations of long and short GRBs are drawn.","marker":"Bhat et al. (2016)"}],"fun_headline_variants":["AGN disks hide most GRBs: few escape as visible flashes","Diffusion kills GRB visibility: only low-z, outer-disk events remain","Funnel or fade: how AGN disks select which GRBs we see","AGN-disk GRBs: rare, low-z, and diffused or funneled","Disk diffusion dims GRBs: detection odds drop to a few percent"],"cache_read_input_tokens":23168,"weakest_assumption_plain":"The load-bearing assumption is that the number of GRB progenitors in a disk is proportional to the disk mass and to the local surface density, so bursts are placed exactly where the disk is heaviest; if migration traps or other formation channels concentrate progenitors elsewhere, the predicted detectable fractions and the mass/radius distribution of detections change materially.","fun_headline_variants_meta":{"raw":{"variants":["AGN disks hide most GRBs: few escape as visible flashes","Diffusion kills GRB visibility: only low-z, outer-disk events remain","Funnel or fade: how AGN disks select which GRBs we see","AGN-disk GRBs: rare, low-z, and diffused or funneled","Disk diffusion dims GRBs: detection odds drop to a few percent"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000287,"raw_usage":{"total_tokens":1801,"prompt_tokens":1173,"completion_tokens":628,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":789,"completion_tokens_details":{"reasoning_tokens":526}},"tokens_in":789,"tokens_out":628,"duration_ms":5397,"temperature":1.0,"reasoning_tokens":526,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T05:15:30.118236+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find and localize AGN-disk GRB candidates in a large sample: the model predicts detectable events overwhelmingly at low redshift (z<1-ish), around SMBHs with M>$10^{7}$.5 solar masses, and from outer disk radii, so a well-localized burst at high redshift or in a low-mass AGN - or with a bright radio afterglow, which self-absorption should suppress - would contradict the central prediction.","supporting_citations":[],"review_version":1}