{"id":"5f9d0030-9055-4317-97b8-0fcf11aa1d8a","arxiv_id":"2510.07259","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Five cosmological simulations predict that the faint ultraviolet excess seen by the Condor telescope at z≈2.5 is Lyman-alpha light from the cosmic web, with measurable detection thresholds for wide-field surveys.","lead":"The paper's simulations of gas between galaxies match a faint ultraviolet glow detected by the Condor telescope, supporting the idea that this glow is Lyman-alpha light from the cosmic web at redshift 2.5. It quantifies the sensitivity needed for future telescopes to map these intergalactic filaments directly.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Claimed consistency with the Condor UV excess is asserted, not demonstrated: no observed A2 statistic or pixel comparison is shown; the paper's own §4 caveat attributes the excess to unresolved sources.","rationale":"The reader's conditional verdict and weakest assumption (unvalidated dust model controlling the bright end) are legitimate, but the more fundamental issue is that the paper never quantitatively compares simulations to the actual Condor data. Fig. 6 only establishes that simulated maps can produce a detectable departure from Gaussian noise at certain thresholds; it does not test whether the real masked difference image exhibits that departure. This gap is highlighted by the manuscript's own admission that the excess may be unresolved sources rather than diffuse cosmic-web gas. I keep the verdict CONDITIONAL/UNCHANGED because the sensitivity forecasts and cross-simulation spread remain useful independent of the discovery claim; however, the 'consistency' claim should be explicitly downgraded until a direct observed-versus-simulated comparison is supplied. This is a different emphasis from the reader's weakest_assumption, hence 'partial' agreement.","tokens_in":30059,"tokens_out":6195,"duration_ms":57897,"concrete_test":"Take the masked Condor narrow-band minus luminance difference image from Lanzetta et al. (2024) and compute the Anderson-Darling statistic between the observed pixel CDF and (a) a pure Gaussian noise model estimated from the image, and (b) the same Gaussian plus each simulation's dust-model map (§2.7), using identical pixel scale, mask, and noise treatment. If the observed A2 is consistent with noise alone, or if adding unresolved point-source populations—not diffuse gas—is required to match the observed CDF, the central consistency claim fails. As a secondary check, rerun Fig. 6 with the LMC dust curve (Eq. 18) to see whether the 5σ threshold shifts by more than a factor of ~2.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that simulated Lyα cosmic-web emission is consistent with the Condor UV excess—rests on Fig. 6, but that figure only measures how simulated maps plus Gaussian noise deviate from pure Gaussian noise (Anderson-Darling statistic, §3.3.2). It never computes A2 for the actual masked Condor narrow-band-minus-luminance image, nor does it compare observed and simulated pixel histograms. §3.3.1 lists procedural differences but stops at 'we are nevertheless able to compare the two meaningfully' without performing the comparison. Consequently, the 5σ threshold at ~8×10^-17 erg s^-1 cm^-2 arcsec^-2 is a property of the simulations, not evidence of consistency with the observation. The manuscript's own concluding paragraph weakens the discovery claim: 'Fig.6 suggests that Condor has not detected the diffuse component, but more likely the statistical effect of unresolved sources.' If the observed excess is unresolved source shot noise, it does not validate diffuse Lyα emission. The dust escape model (§2.7, Eqs. 15–18) is a related but secondary vulnerability: it controls the simulated bright end, and the culling alternative (§2.8) gives 'dramatically different' PDFs, so without an observed benchmark the choice of dust treatment cannot be checked.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper post-processes five cosmological hydrodynamic simulations (IllustrisTNG, EAGLE, CROCODILE, SIMBA, Sherwood) at redshifts 2.0–2.74 to predict narrow-band Lyman-α surface brightness from the cosmic web. Emission is computed from recombination, collisional excitation, and star formation, with hydrogen species separated using published fitting functions (Rahmati et al.; Gnedin & Kravtsov). High-density gas is treated either with an SMC-based dust attenuation model (Eqs. 15–18) or by culling particles above a self-shielding density threshold (Sec. 2.8). The projected surface-brightness maps are then degraded by Gaussian noise, and the Anderson-Darling statistic is used to estimate detection thresholds. The authors report that a 5σ detection of total intergalactic/circumgalactic Lyman-α emission should be possible at fluxes brighter than ≈8×10^-17 erg s^-1 cm^-2 arcsec^-2, while the low-density cosmic-web component requires noise below ≈2×10^-19. The abstract and conclusions further claim that these predictions are consistent with the UV excess detected in the Condor Array Telescope narrow-band image reported by Lanzetta et al. (2024).","tokens_in":30323,"tokens_out":5370,"duration_ms":48444,"significance":"If the central consistency claim could be substantiated, the paper would provide a useful framework for planning wide-field narrow-band surveys of diffuse Lyman-α emission and for quantifying the sensitivity needed to map the cosmic web in emission. The multi-simulation comparison is a genuine strength: five independent hydrodynamic codes are processed through a common pipeline, and the derived neutral-hydrogen CDDFs are checked against quasar absorption measurements. The paper is also unusually candid about the limitations of the dust model and about inter-simulation scatter. However, the headline claim of consistency with the Condor detection is not demonstrated by the statistical analysis actually presented, and the conclusions contain an admission that the observed UV excess may be dominated by unresolved sources rather than diffuse Lyman-α. With a direct comparison to the observed image or a reframing as forward predictions only, the simulation methodology and threshold estimates would still be a useful contribution.","major_comments":[{"comment":"The Anderson-Darling statistic is computed only between (simulated map + Gaussian noise) and pure Gaussian noise. It is never computed for the actual masked Condor narrow-band-minus-luminance image described in §3.3.1, nor is the observed pixel PDF compared with the simulated PDFs. The statement in §3.3.1 that the two datasets can be 'compared meaningfully' is not followed by a comparison. Therefore the abstract and §4 claim that the simulation predictions are 'consistent with' the Condor UV excess is not supported by the analysis shown. Please either compute A2 (or an equivalent statistic) for the observed masked difference image, or explicitly downgrade the consistency claim to simulation-only detectability predictions.","section":"§3.3.2, Fig. 6"},{"comment":"The paper itself states that Fig. 6 suggests that Condor has not detected the diffuse component but 'more likely the statistical effect of unresolved sources.' This directly undermines the central claim that the Condor UV excess validates diffuse Lyman-α cosmic-web emission. If unresolved sources dominate the measured excess, then agreement with diffuse-emission simulations is not evidence for detection of the diffuse cosmic web. The manuscript must distinguish between (a) a claimed detection of diffuse IGM/CGM Lyman-α, and (b) a UV excess consistent with unresolved sources plus perhaps a subdominant diffuse component. The current wording conflates the two, and the conclusions should be revised to state precisely which claim is intended.","section":"§4, concluding paragraphs"},{"comment":"The dust escape model is derived from local-universe relations, and the paper concedes in §2.1 that no dust constraints exist at z≈2.5 on simulation-particle scales. This model controls the bright end of the surface-brightness distribution and hence the quoted 5σ threshold near 8×10^-17 in the left panel of Fig. 6. The alternative density-culling treatment in §2.8 yields 'dramatically different' PDFs and threshold curves (right panel of Fig. 6). Because no observed comparison anchors the choice between these treatments, the central threshold claim is sensitive to an unvalidated modeling assumption. Please quantify the dependence of the 5σ threshold on dust-model parameters (e.g., albedo, metallicity scaling, SMC vs. LMC curve) and state whether the threshold is robust across the two adopted treatments or only a property of the fiducial dust prescription.","section":"§2.7, Eqs. 15–18; §2.8; Fig. 6"},{"comment":"The factor-of-six spread in predicted diffuse IGM temperatures is presented as an inter-simulation result, but the paper also notes that all five simulations overestimate the observationally inferred Lyman-α forest temperature. This is more than a diagnostic aside: the temperature enters the emissivity calculations directly, so the bright-end predictions inherit this systematic scatter. A quantitative statement of how much of the simulation-to-simulation spread in Fig. 6 is driven by temperature differences (as opposed to density/star-formation prescriptions) would strengthen the paper and help the reader judge whether the claimed consistency threshold is robust.","section":"§3.2.1, Table 2"}],"minor_comments":[{"comment":"Section 1 states that the Condor narrow-band filter probes z=2.24754±0.0030, while §3.3.1 and §4 use z=2.4754. For λ_Lyα=121.567 nm and λ_filter=422.5 nm, z≈2.4754. The first value appears to be a typo and should be corrected.","section":"§1; §3.3.1"},{"comment":"The main text says 'the intensity scaling is kept the same for all panels,' but the figure caption says 'each simulation has its own colour map... numerical display ranges are not the same.' Please reconcile these statements.","section":"Fig. 3 text vs caption"},{"comment":"The summation notation in Eq. (17) is undefined: the sums over Z_i and Z_i,0 are not specified (over which elements? per particle? solar abundances?). Please define all indices and reference abundances explicitly.","section":"Eq. (17)"},{"comment":"The horizontal '5 sd.' threshold is attributed to SciPy but the corresponding Anderson-Darling critical value or p-value conversion is not given. Please state the numerical threshold used so the claim is reproducible.","section":"Fig. 6"},{"comment":"Small typos include 'Anderson-Darling statistical text' (§3.3.2), 'observational and stimulation data' (§3.1), and the bibliographic entry 'Lanzetta et al K. M., 2024', which should be formatted consistently.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper's most prominent claim—consistency with the Condor UV excess—is not actually tested with the Condor image, and the conclusions concede unresolved sources as a likely explanation. I would like the editor to ensure that the revised manuscript either includes a direct comparison between simulated and observed pixel statistics or clearly rephrases the contribution as forward predictions. The multi-simulation methodology is worth salvaging, but the current framing overstates the observational support."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"You should know two things. First, the genuinely useful part of this paper is a systematic forecast of cosmic-web Lyα surface brightness from five independent hydrodynamic simulations, with a uniform semi-analytic emission pipeline and explicit detectability thresholds. Second, the paper's central claim — that the Condor UV excess is consistent with simulated Lyα emission — is asserted, not demonstrated. The abstract states it as fact, but the analysis never actually compares the simulations to the observed image.\n\nThe new content is the five-simulation cross-comparison: IllustrisTNG, EAGLE, SIMBA, CROCODILE, and Sherwood, post-processed with recombination, collisional excitation, star formation, the Rahmati et al. HI-fraction fitting functions, and a Laursen et al. SMC dust escape model. The CDDF check against quasar absorption data is a nice sanity test, and the paper is honest about the large systematic spread — factor-of-six IGM temperatures, SIMBA's anomalous low-metallicity high-density population, and the \"dramatically different\" PDFs produced by the alternative high-density culling treatment. The sensitivity thresholds (roughly 8×10^-17 erg s^-1 cm^-2 arcsec^-2 for total Lyα, and ~2×10^-19 for the diffuse low-density component in the most optimistic case) are useful planning numbers for wide-field narrow-band surveys.\n\nThe soft spot is load-bearing. Fig. 6 shows the Anderson-Darling statistic for simulated maps with added Gaussian noise versus pure Gaussian noise. It never computes A2 for the actual masked Condor narrow-band-minus-luminance image, and never compares observed and simulated pixel histograms. Section 3.3.1 lists procedural differences and stops at \"we are nevertheless able to compare the two meaningfully,\" without performing the comparison. More damaging, the paper's own conclusion says \"Fig.6 suggests that Condor has not detected the diffuse component, but more likely the statistical effect of unresolved sources.\" That is not validation of diffuse Lyα emission. The dust model is secondary but still important: it is unvalidated at z~2.5 and controls the bright end that sets the 5σ threshold, and the culling alternative gives far different results. Also, the observation and theory come from overlapping authors, so this is not independent confirmation.\n\nCredit where due: the paper is unusually transparent about its limitations, and the theoretical machinery is based on standard, previously published fitting functions rather than new untested physics. That honesty is appreciated. But it does not rescue a headline that outruns the analysis.\n\nThis paper is for anyone planning low-surface-brightness Lyα surveys, and for theorists interested in semi-analytic emission modeling. It deserves a serious referee because the forecast methodology is relevant and the observational comparison could in principle be made quantitative. My recommendation: engage, but require the authors to either compute the observed A2 and pixel comparison or explicitly downgrade the consistency claim to speculation. The discovery interpretation should not be accepted on the current evidence.","headline":"A useful multi-simulation Lyα forecast whose claimed validation of the Condor excess is asserted, not demonstrated — and the paper's own final caveat admits unresolved sources.","tokens_in":30978,"tokens_out":4517,"would_cite":true,"duration_ms":37739,"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":"The paper argues that the ultraviolet excess seen in a deep, wide-field narrow-band image taken by the Condor Array Telescope is Lyman-α emission from the cosmic web at redshift z≈2.5, a claim supported by synthetic maps built from five ind","keywords":["Lyman-alpha emission","cosmic web","intergalactic medium","narrow-band imaging","surface brightness","hydrodynamic simulations","dust attenuation","Condor Array Telescope"],"falsifier":"A narrow-band image of the same COSMOS field taken with a filter offset by a few ångströms (so Lyα at z=2.4754 falls outside but [OII] at z=0.1332 remains inside) should show the excess vanish if it is Lyα; if it persists, the signal is not the cosmic-web glow. Alternatively, a future 5σ detection of the diffuse component at noise ~2×10^-19 erg s^-1 cm^-2 arcsec^-2 would confirm the most optimistic simulation.","tokens_in":29912,"feed_emoji":"🔭","tokens_out":6646,"duration_ms":59485,"temperature":0.7,"pith_summary":"The authors set out to show that the faint UV excess detected in the deep Condor narrow-band image is the long-sought Lyman-α glow of the cosmic web, not a foreground contaminant or artifact. They construct synthetic Lyman-α surface-brightness maps from five hydrodynamic simulations, modeling recombination, collisional excitation, star formation, and dust attenuation, and find that the simulated brightness distributions match the statistical shape of the observed excess. If correct, this would be the first wide-field, narrow-band detection of the cosmic web in emission at z≈2.5, opening a new observational window onto the gas that underpins large-scale structure.","feed_headline":"Simulations tie Condor UV excess to cosmic web gas","feed_subtitle":"Five hydro models match the faint glow at z≈2.5; the diffuse component still needs 100x deeper imaging.","key_machinery":"The central machinery is a set of synthetic Lyman-α surface-brightness maps produced by post-processing five full-physics hydrodynamic simulations (IllustrisTNG, EAGLE, CROCODILE, SIMBA, Sherwood) onto a common grid, using semi-analytic separation of hydrogen species and emissivities from recombination, collisional excitation, and star formation. Dust attenuation is treated particle-by-particle via a slab-geometry escape fraction with an SMC extinction curve and a dust-to-baryon scaling. The Anderson-Darling statistic, which weights the tails of the brightness distribution, is then applied to noisy simulated maps to quantify detection thresholds against pure Gaussian noise.","core_discovery":"On the paper's own terms, the central discovery is that the surface-brightness distribution of narrow-band Lyman-α emission predicted by hydrodynamic simulations of the z≈2.5 cosmic web is statistically consistent with the UV excess in the Condor Array Telescope image reported by Lanzetta et al. (2024). Using the Anderson-Darling statistic, the authors show that three of the four simulations that include dust predict a 5σ deviation from pure Gaussian noise for fluxes brighter than roughly 8×10^-17 erg s^-1 cm^-2 arcsec^-2, and that isolating the low-density diffuse component requires noise below about 2×10^-19 erg s^-1 cm^-2 arcsec^-2, reachable only by the most optimistic simulation (Illust","pith_inferences":["If a follow-up observation with a detuned filter (a few ångströms off the Lyα redshift) shows the excess disappearing, the cosmic-web interpretation is strongly confirmed; this is the most direct test the paper leaves implicit.","The paper's own caveat that the signal is likely unresolved sources rather than diffuse gas implies that 'cartographic mapping' of the filamentary network may first require separating a population of faint, clustered emitters before the diffuse component becomes visible.","The same semi-analytic pipeline could be adapted to other redshifts and emission lines, turning wide-field imagers into systematic probes of diffuse baryons, provided dust models can be validated at those epochs.","The markedly different surface-brightness PDFs produced by the dust-escape model versus high-density culling indicate that the claimed 5σ threshold for the total component is not yet pinned down; calibrating dust-to-gas at z~2.5 would materially sharpen the forecasts."],"forward_implications":["If the simulations are right, the Condor UV excess is Lyman-α emission from intergalactic and circumgalactic gas at z≈2.48, making it the first wide-field detection of the cosmic web in emission.","Wide-field narrow-band surveys with sensitivities around 10^-17 erg s^-1 cm^-2 arcsec^-2 or better can statistically detect the total cosmic-web Lyman-α glow over volumes of hundreds of thousands of cubic megaparsecs.","Detecting the truly diffuse, low-density component requires roughly two orders of magnitude deeper imaging (noise below ~2×10^-19 erg s^-1 cm^-2 arcsec^-2 in the most optimistic case), a target for next-generation facilities.","The broad agreement across four independent simulations, despite large differences in subgrid physics, suggests the statistical detection is not an artifact of a single baryon model, though the bright end is dust-sensitive.","Contaminating lines such as [OII] at z≈0.13 are shown to contribute at most a few tenths of a percent of the signal, so the 422 nm narrow-band excess is effectively pure Lyman-α."],"fun_headline_variants":["Simulations align with cosmic web's faint Lyman-alpha glow","Cosmic web Lyman-alpha: simulations match Condor excess","Wide-field Lyman-alpha detection backed by five simulations","Diffuse cosmic web glow needs 100x deeper imaging","Lyman-alpha glow at z≈2.5: simulations vs. Condor UV"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The local-universe dust escape model—SMC extinction with dust-to-baryon proportionality—correctly describes how Lyman-α photons escape from z≈2.5 gas, even though no dust constraints exist at those redshifts on simulation-particle scales.","fun_headline_variants_meta":{"raw":{"variants":["Simulations align with cosmic web's faint Lyman-alpha glow","Cosmic web Lyman-alpha: simulations match Condor excess","Wide-field Lyman-alpha detection backed by five simulations","Diffuse cosmic web glow needs 100x deeper imaging","Lyman-alpha glow at z≈2.5: simulations vs. Condor UV"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000828,"raw_usage":{"total_tokens":3514,"prompt_tokens":861,"completion_tokens":2653,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":605,"completion_tokens_details":{"reasoning_tokens":2578}},"tokens_in":605,"tokens_out":2653,"duration_ms":12380,"temperature":1.0,"reasoning_tokens":2578,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T10:58:37.438405+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A narrow-band image of the same COSMOS field taken with a filter offset by a few ångströms (so Lyα at z=2.4754 falls outside but [OII] at z=0.1332 remains inside) should show the excess vanish if it is Lyα; if it persists, the signal is not the cosmic-web glow. Alternatively, a future 5σ detection of the diffuse component at noise ~2×10^-19 erg s^-1 cm^-2 arcsec^-2 would confirm the most optimistic simulation.","supporting_citations":[],"review_version":1}