{"id":"a756f525-5532-400c-baaf-9e161284ca50","arxiv_id":"2506.18993","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"Lyα photons can largely escape through a biconical outflow funnel in direct-collapse black hole seeds, making these objects JWST-detectable with a lopsided red line profile.","lead":"This paper uses Monte Carlo simulations of Lyα light to ask whether gas falling onto a direct-collapse black hole seed glows in a way telescopes can see. If the infall has a hollow, funnel-shaped outflow, most Lyα photons escape, and the glow may be detectable by JWST with a distinctive lopsided line shape.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The high escape fraction depends on an unvaried, imported funnel state; the paper never tests fesc against a clumpy or partially neutral funnel, so the JWST detectability claim is not yet robust.","rationale":"All downstream claims—JWST 10σ in 10^4 s detectability and the line-shape discriminant—scale with the Lyα escape fraction and with the transparency of the line of sight. The internal machinery is credible: the radiative transfer is standard, the two-photon destruction is included, and the code reproduces the Neufeld analytic result (Fig. 2). This locates the risk not in the solver but in the assumed input state. The funnel is the escape route, and it is imported from earlier simulations rather than derived simultaneously with the Lyα-emitting shell; the paper varies only the shell optical depth, inflow column, and opening angle, leaving the funnel density, ionization, and temperature fixed. The manuscript itself flags that these variations are not considered (§4), and it accounts for IGM absorption only on the blue side despite the red wing carrying the observable flux. The reader's weakest-assumption analysis correctly identifies the funnel state as the crux. The concern is substantial but addressable, so the existing CONDITIONAL verdict is appropriate; I do not move it to REJECT because the required checks are concrete and feasible with the authors' own tools and simulation outputs.","tokens_in":17618,"tokens_out":6621,"duration_ms":75544,"concrete_test":"Extract a snapshot from Luo et al. (2023) at the epoch when the central mass is ~10^5 M_sun and the funnel/shell structure has developed; run the same Zheng & Miralda-Escudé Monte Carlo code directly on the full three-dimensional density, temperature, velocity, and ionization fields, placing the Lyα source at the shell location with the same luminosity and destruction physics as in §2. Compare the angle-dependent fesc and the red-wing profile with the idealized Table 1 models for Nin = 10^19–10^22 cm^-2, and include the IGM damping-wing opacity on the red wing at z = 10. If the real funnel yields fesc within a factor of ~2 of 0.95 and the red peak survives the IGM, the central claim holds; if fesc drops below ~0.3 or the red peak is suppressed by the damping wing, the JWST detectability claim needs revision.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Concern: the >95% escape fraction (abstract) and the few-µJy JWST flux (§3, Eq. 7, Figs. 3–4) are produced by post-processing an idealized four-component geometry in which the biconical funnel is prescribed as nfn = 100 cm^-3, xfn = 1.0, Tfn = 10^6 K (Table 1), with a thin dense shell (τsh = 10^6–10^8) and an outer spherical inflow. These funnel parameters are imported from the authors' earlier zoom-in simulations (Luo et al. 2023; Luo & Shlosman 2024) and are then held fixed: the paper varies only Nin, τsh, and θopen, stating that the remaining parameters are kept constant and their effects disregarded (§3). The escape route is precisely the funnel, so the central claim requires the funnel to stay nearly empty and fully ionized while Lyα is being produced in or around the shell. The paper does not test what happens if the funnel is partially neutral or clumpy, and it does not run a self-consistent simulation in which Lyα production and clearing of the funnel are evolved together. The body itself concedes that the model assumes these parameters within a reasonable range guided by simulations and does not focus on their variations (§4). Because the detectable red peak appears at Δv ≈ 200–400 km s^-1 (Fig. 5) and then must traverse the outside IGM, an unquantified damping-wing opacity on the red side is a second, separate load on the same detection claim. The Monte Carlo code itself is validated against Neufeld (1990) (Fig. 2), so the weak link is the input geometry, not the transfer solver.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents Monte Carlo Lyα radiative transfer calculations in a prescribed four-component inflow-outflow geometry (spherical inflow, accretion disk, bi-conical outflow funnel, and expanding shell) intended to represent the central region of a direct-collapse pre-supermassive-black-hole object at z=10. The geometry is imported from the authors' earlier zoom-in cosmological simulations (Luo et al. 2023; Luo & Shlosman 2024). Varying the inflow column density Nin, shell optical depth τsh, and funnel opening angle θopen, the authors compute escape fractions, Lyα line profiles, and line diagnostics (peak velocity shift, width, asymmetry, cuspiness, equivalent width). They report that more than 95% of Lyα photons can escape through the funnel for Nin below about 10^22 cm^-2, that the resulting flux density of a few μJy is detectable by JWST NIRSpec in MOS mode at 10σ in 10^4 seconds, and that the line profiles, characterized by strong asymmetry and an extended red tail, can distinguish these objects from high-redshift LAEs and quasars.","tokens_in":18007,"tokens_out":5214,"duration_ms":57487,"significance":"If the central claim holds, the paper would provide a concrete observational strategy for identifying direct-collapse black hole seeds with JWST and for separating them from ordinary star-forming galaxies at high redshift. This would be an important step in testing the direct-collapse pathway to supermassive black holes. The Monte Carlo code is validated against the Neufeld (1990) analytic solution for spherical accretion, including the two-photon destruction channel, and the paper systematically explores a meaningful parameter space of column density, shell optical depth, and opening angle. The line-diagnostic framework is also well motivated and the comparison with observed LAE and quasar line shapes is a useful addition. However, the headline escape fraction and the resulting detectability estimate rest on a prescribed, fully ionized, low-density funnel whose properties are held fixed rather than varied or derived self-consistently, and the treatment of the intergalactic medium on the red side of the line is incomplete.","major_comments":[{"comment":"The escape route for Lyα photons is precisely the bi-conical funnel, yet the funnel parameters nfn = 100 cm^-3, xfn = 1.0, Tfn = 10^6 K are held fixed, and §3 states that 'the remaining parameters are sourced from the simulation models ... and kept constant during the calculations' and that 'their effects are disregarded.' With xfn = 1.0, the funnel is completely ionized and therefore optically thin to Lyα by construction, so a high escape fraction is effectively imposed rather than predicted by the transfer calculation. The authors should vary the funnel neutral fraction and density (including a clumpy or partially neutral funnel) and examine the effect on fesc and the JWST flux, since a modest neutral fraction or clumpiness in the funnel would increase the optical depth and could substantially reduce the escape fraction.","section":"§3 and Table 1"},{"comment":"The abstract's statement that 'the escaping fraction of Lyα radiation exceeds 95% from a z=10 pre-SMBH object' is contradicted by the paper's own results: §4 states that for Nin >~ 10^22–10^23 cm^-2 only 1–10% of photons escape, and Fig. 3 shows a drop in the peak flux by about two orders of magnitude at Nin = 10^22 cm^-2 relative to Nin = 10^19 cm^-2. The abstract and the detectability claim should be qualified to state the column-density regime (e.g., Nin <~ 10^21 cm^-2) and the corresponding restricted parameter range in which the high escape fraction applies.","section":"Abstract, §4, and Fig. 3"},{"comment":"The detectability calculation, Eq. (7), uses the full intrinsic Lyα flux, and the authors note that the blue wing is absorbed by the IGM. However, the observable red peak is located at Δv ≈ 200–400 km s^-1 (Fig. 5), and at z=10 the IGM is likely substantially neutral, so the red wing will also suffer damping-wing absorption. The paper does not compute the IGM transmission on the red side. Without an estimate of this damping-wing opacity (e.g., a Miralda-Escudé–type calculation or a numerical IGM transfer), the reported JWST flux and exposure time are upper limits rather than robust predictions.","section":"§3, Eq. (7), and §4"}],"minor_comments":[{"comment":"The table heading 'disk opening angel' should read 'disk opening angle'.","section":"Table 1"},{"comment":"In the text describing Figure 6, 'leptocurtic' should be 'leptokurtic', consistent with the footnote in §2.","section":"§3"},{"comment":"The agreement between the numerical escape fractions and the Neufeld (1990) analytic solution is described as 'excellent', but no quantitative residual or maximum deviation is given; adding a simple accuracy measure would strengthen the code-validation statement.","section":"§3 and Fig. 2"},{"comment":"The statement that for Nin >~ 10^22–10^23 cm^-2 only 1–10% of the radiation escapes is not directly evident from Fig. 3, where the Nin = 10^22 case appears to have a peak flux about 1% of the lowest-Nin case; please clarify which parameter combination produces the 10% end of the quoted range.","section":"§4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a useful proof-of-concept for Lyα escape in anisotropic direct-collapse geometries, and the code validation is a real strength. The central issue is that the high escape fraction is effectively an input assumption via the fully ionized, low-density funnel, and the abstract overstates the body's own parameter-dependent results. The requested robustness tests (varying funnel neutral fraction/density and adding IGM damping-wing transmission) appear feasible within the scope of the paper, so I do not recommend rejection. If those tests show a strong sensitivity of fesc to the funnel state, the detectability claim will need to be substantially toned down."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper asks whether Lyα photons can escape from direct-collapse pre-SMBH objects and be detected by JWST. The new contribution is not the general idea that anisotropic geometry aids Lyα escape—that is standard in the LAE literature—but the specific application to the inflow-outflow geometry from the authors' zoom-in simulations, with quantitative escape fractions, line profiles, and a proposed red-tail discriminant. That is a genuine step forward, and the Monte Carlo transfer code is validated against the Neufeld slab solution.\n\nWhat the paper does well: it lays out a four-component geometry, varies the sensible parameters (inflow column density, shell optical depth, funnel opening angle), and shows that the line shape—especially the red peak shift and asymmetry—separates these objects from ordinary LAEs and high-z quasars. The findability estimate with NIRSpec MOS is a concrete, testable prediction. If such a funnel exists, the calculation is credible.\n\nThe soft spots are real, though not fatal. The abstract says the escaping fraction 'exceeds 95%' with no qualification, but the body shows that this holds only for Nin ≲ 10^22 cm^-2; at Nin ~ 10^22 the flux drops by a factor ~100, and §4 concedes 1–10% escape at higher columns. The abstract should carry the same caveat.\n\nMore substantively, the funnel is assumed, not computed. The density (100 cm^-3), full ionization, and 10^6 K temperature are imported from the authors' earlier simulations and held fixed. The escape route is precisely that funnel, so the central claim depends on the funnel staying empty and ionized while Lyα is produced. The paper does not test a clumpy or partially neutral funnel. It also ignores the IGM damping-wing opacity on the red wing at z=10, which could significantly absorb the red peak at Δv ~ 200–400 km/s. These are load-bearing for the JWST detectability claim, and both are addressable.\n\nAlso, the paper is honest about the fixed parameters: §4 states that they are assumed within a reasonable range and their variations are not explored. So this is not a hidden flaw, just an open one.\n\nWho this is for: anyone working on SMBH seed formation, Lyα radiative transfer, or JWST searches for high-z sources. It deserves a serious referee, not a desk reject. I would send it to review with a request that the abstract be tightened and that the authors add a test of funnel persistence (or at least a clumpy/neutral funnel variant) and a rough damping-wing estimate.\n\nRecommendation: engage with it, and require revision before acceptance. If the authors add those two checks, this becomes a solid contribution.","headline":"A useful, well-executed model-prediction paper whose abstract overstates the escape fraction and whose central claim rests on an unvaried funnel geometry; still deserves review and likely publication after revision.","tokens_in":18500,"tokens_out":4163,"would_cite":true,"duration_ms":37977,"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":"Lyα photons can escape direct-collapse black hole seeds through an ionized funnel, making them JWST-detectable.","keywords":["direct collapse","supermassive black hole seeds","Lyα emission","Lyα radiative transfer","biconical outflow","JWST NIRSpec","high-redshift galaxies","two-photon emission"],"falsifier":"Run a self-consistent radiation-hydrodynamic zoom-in simulation of direct collapse at z=10 that follows both the gas dynamics and the Lyα photon production, and measure the escape fraction without imposing a pre-cleared funnel; if the self-consistently maintained funnel has a neutral fraction above a few percent or a clumpy interior, the escape fraction will drop far below 95%. Alternatively, a JWST NIRSpec MOS survey of a statistically meaningful sample of direct-collapse candidate halos at z≈10 that finds no Lyα emitters with the predicted asymmetric red-tailed profiles at fluxes above ~1 µJy in $10^{4}$ second exposures would contradict the detectability claim.","tokens_in":17400,"feed_emoji":"🔭","tokens_out":3396,"duration_ms":35281,"temperature":0.7,"pith_summary":"This paper argues that Lyα photons, long thought to be trapped and destroyed in the spherical collapse picture of supermassive black hole seed formation, can actually escape when the collapse is anisotropic. Using a Monte Carlo Lyα radiative transfer code on the inflow-outflow geometry taken from earlier zoom-in cosmological simulations, the authors find that a nearly empty, fully ionized biconical funnel plus a thin expanding shell allows more than 95% of Lyα photons to escape from a z=10 pre-SMBH object. The escaping flux, a few microjansky, would be detectable by JWST NIRSpec in multi-object spectroscopy mode within about $10^{4}$ seconds at 10σ. The resulting line profiles show strong asymmetry and an extended red tail, which the authors propose as a way to observationally distinguish direct-collapse seeds from Lyman-$\\alpha$ emitters and quasars. If correct, this gives a concrete observational route to finding the seeds of supermassive black holes before they grow into quasars.","feed_headline":"Lyα escape route makes black hole seeds JWST-visible","feed_subtitle":"At z=10, a biconical funnel lets over 95% of Lyα photons out, enough for a 10-sigma NIRSpec detection in ~10^4 seconds.","key_machinery":"The load-bearing element is the anisotropic inflow-outflow geometry: a biconical funnel of hot, fully ionized, low-density gas (n ≈ 100 $cm^{-3}$, T ≈ $10^{6}$ K, x ≈ 1) carved by an outflow, surrounded by a thin expanding shell with Lyα optical depth up to $10^{8}$, and an outer spherical inflow with column density $10^{19}$–$10^{22}$ $cm^{-2}$. This funnel provides a low-optical-depth escape route for Lyα photons, while the expanding shell's velocity field suppresses the blue wing and enhances the red wing. The paper combines this geometry with a Monte Carlo radiative transfer algorithm that tracks resonant scattering, Doppler shifts, and the destruction probability into two-photon emission, using the escape fraction formula f_esc = (1 − p_dest)^{N_sca}.","core_discovery":"The central claim is that in the anisotropic inflow-outflow geometry inferred from prior zoom-in simulations of direct collapse, Lyα photons can escape the central region with a fraction exceeding 95% when the column density of the spherical inflow outside the shell is below about $10^{22}$ $cm^{-2}$. The geometry consists of a central $10^{5}$ solar mass object with effective temperature 6×$10^{4}$ K, a rotationally supported disk, a low-density hot funnel cleared by radiation or magnetic forces, and a dense radiatively driven expanding shell. The Monte Carlo transfer, which includes destruction into two-photon emission, produces double-peaked Lyα lines whose blue wing is absorbed by the intergalactic medium, leaving a redshifted red peak with velocity shifts of roughly 150–1200 km/s, FWHM of 200–2000 km/s, and a pronounced red tail. The escaping flux of a few microjansky is predicted to be detectable by JWST NIRSpec MOS at z=10 in $10^{4}$ seconds at 10σ, and the line shape diagnostics—peak shift, asymmetry, cuspiness, and equivalent width—are proposed as discriminators against LAEs and high-redshift quasars.","pith_inferences":["If the funnel is clumpy, partially neutral, or transient during the Lyα production phase, the escape fraction could drop far below 95%; a self-consistent radiation-hydro or MHD simulation that produces both the Lyα photons and the funnel simultaneously would settle this directly.","The paper neglects damping-wing absorption of the red line wing by the partially neutral intergalactic medium at z=10; including a realistic IGM transmission model could reduce the predicted observed flux and push the required exposure time beyond 10^4 seconds.","The strong inclination dependence of the peak flux—a factor of four reduction from face-on to edge-on for narrow funnels—implies that surveys will be biased toward face-on direct-collapse seeds, and the red tail may be more visible in edge-on systems where it is broadened.","The proposed line-shape diagnostic could be applied to existing JWST spectra of high-redshift compact objects or 'little red dots' to search for candidate direct-collapse seeds, even before a dedicated survey is designed."],"forward_implications":["JWST NIRSpec MOS observations at z≈10 could detect direct-collapse pre-SMBH objects with exposure times of roughly 10^4 seconds, provided the inflow column density stays below about 10^21 cm^-2.","The predicted Lyα line shape—a strongly asymmetric single peak with an extended red tail—gives observers a spectroscopic discriminant to separate direct-collapse seeds from ordinary Lyman-alpha emitters and high-redshift quasars.","If the inflow column density exceeds about 10^22–10^23 cm^-2, the escape fraction drops to only 1–10%, so only objects with cleared funnels will be visible in Lyα, biasing any survey toward the most evolved seeds.","The two-photon continuum flux increases by about three orders of magnitude with inflow column density, offering an independent broadband signature of Lyα destruction in these systems.","The model predicts a negative correlation between Lyα peak velocity shift and rest-frame equivalent width, a trend also seen in LAEs and high-z quasars, meaning this relation alone cannot distinguish the populations."],"supporting_citations":[{"why":"Supplies the Monte Carlo Lyα radiative transfer code used for all scattering and escape calculations.","marker":"Z. Zheng & J. Miralda-Escudé (2002)"},{"why":"Provides the analytical escape probability for Lyα photons in optically thick media, used to validate the code.","marker":"D. A. Neufeld (1990)"},{"why":"The zoom-in cosmological simulation that produced the disk, biconical outflow, funnel, and expanding shell geometry adopted here.","marker":"Y. Luo et al. (2023)"},{"why":"Extends the geometry to include magnetic forces and reinforces the funnel formation scenario.","marker":"Y. Luo & I. Shlosman (2024)"},{"why":"Represents the spherical-collapse baseline where Lyα is trapped and destroyed, the contrast case for the anisotropic escape claim.","marker":"Q. Ge & J. H. Wise (2017)"},{"why":"Supplies the estimate that roughly 40% of atomic cooling radiation emerges as Lyα, justifying the assumed Lyα luminosity.","marker":"M. Dijkstra et al. (2014)"},{"why":"Defines the line diagnostics (peak shift, half-width, asymmetry, cuspiness) used to characterize the predicted profiles.","marker":"R. T. Emmering et al. (1992)"}],"fun_headline_variants":["Lyα escape through biconical funnel reveals black hole seeds","Black hole seeds detectable via Lyα funnel and red tail","Pre-SMBH objects as Lyα emitters: JWST can catch them","Lyα funnel lets 95% escape, making black hole seeds JWST-visible","Direct-collapse black hole seeds appear as Lyα emitters"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result rests on the existence and persistence of a nearly empty, fully ionized biconical funnel with a thin dense shell during the Lyα production phase, a geometry adopted from earlier simulations rather than produced self-consistently in the same calculation that emits the Lyα photons.","fun_headline_variants_meta":{"raw":{"variants":["Lyα escape through biconical funnel reveals black hole seeds","Black hole seeds detectable via Lyα funnel and red tail","Pre-SMBH objects as Lyα emitters: JWST can catch them","Lyα funnel lets 95% escape, making black hole seeds JWST-visible","Direct-collapse black hole seeds appear as Lyα emitters"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000442,"raw_usage":{"total_tokens":2336,"prompt_tokens":1137,"completion_tokens":1199,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":753,"completion_tokens_details":{"reasoning_tokens":1105}},"tokens_in":753,"tokens_out":1199,"duration_ms":10869,"temperature":1.0,"reasoning_tokens":1105,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:40:17.836849+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run a self-consistent radiation-hydrodynamic zoom-in simulation of direct collapse at z=10 that follows both the gas dynamics and the Lyα photon production, and measure the escape fraction without imposing a pre-cleared funnel; if the self-consistently maintained funnel has a neutral fraction above a few percent or a clumpy interior, the escape fraction will drop far below 95%. Alternatively, a JWST NIRSpec MOS survey of a statistically meaningful sample of direct-collapse candidate halos at z≈10 that finds no Lyα emitters with the predicted asymmetric red-tailed profiles at fluxes above ~1 µJy in $10^{4}$ second exposures would contradict the detectability claim.","supporting_citations":[{"cited_title":"2023, ApJ, 955, 99, doi: 10.3847/1538-4357/acefb9","cited_arxiv_id":null,"evidence_quote":"The zoom-in cosmological simulation that produced the disk, biconical outflow, funnel, and expanding shell geometry adopted here."},{"cited_title":"2024, ApJ, 976, 85, doi: 10.3847/1538-4357/ad7fec","cited_arxiv_id":null,"evidence_quote":"Extends the geometry to include magnetic forces and reinforces the funnel formation scenario."}],"review_version":1}