{"id":"8666961d-8919-4db7-94e5-e88671bbaad8","arxiv_id":"2504.13253","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Measurements of z~2.4 ultraviolet background light from Ly-alpha absorption troughs in 400,000 stacked HETDEX spectra give 12.9 plus or minus 3.7 nW m^-2 sr^-1 at 4134 Angstroms, with strong dependence on galaxy density.","lead":"Astronomers stacked 400,000 Lyman-alpha emitting galaxies from the HETDEX survey and found that dark absorption features around the emission line match absorption of background ultraviolet light by hydrogen gas around the galaxies. If the model is right, the depth of these features gives a new way to measure the extragalactic background light at redshift 2.4 and shows it varies strongly with galaxy density.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Reported EBL value assumes a fully covering, saturated absorber; the paper states this gas is unquantified, so the absolute normalization is not yet established.","rationale":"The paper is a serious attempt to measure the z~2.4 UV background using an interesting and potentially powerful technique. It has real supporting evidence: the troughs appear in LAE stacks but not in [O II] emitter stacks, they appear in pre-sky-subtracted spectra, and they scale with local density. The relative density trend is likely robust. However, the headline absolute intensity is the most load-bearing output, and it depends exactly on the assumption the authors explicitly flag as unquantified: that the absorbing H I is saturated and fully covers the aperture. The reader's weakest assumption identifies this same point. My stress test sharpens it by separating the covering fraction and wavelength-dependent optical depth from the fitted offset. A forward-model fit to the existing pre-sky-subtraction stacks would settle whether the offset can be interpreted as the full EBL or needs a correction factor. Since the authors already call for radiative-transfer modeling, the conditional verdict is appropriate; no verdict change is needed, but the condition should explicitly include measuring the covering fraction and optical depth before the absolute EBL value is cited.","tokens_in":13666,"tokens_out":8271,"duration_ms":88644,"concrete_test":"Fit a forward model to the pre-sky-subtraction LAE stacks shown in Figure 3: model the trough as damped Lyα absorption with a grid of column densities log N_HI = 19–22, Doppler parameters b = 30–300 km/s, and covering fractions f = 0.2–1.0, with a free incident UVB intensity. If the best-fit f is significantly below 1 or the best-fit optical depth is below unity at the wavelengths used for the offset, recompute the EBL as 12.9 / f (or as the fitted incident intensity); if the fit cannot distinguish f from 1, the reported absolute value remains conditional on the fully covering saturated assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central number 12.9 ± 3.7 nW m^-2 sr^-1 is obtained by adding a constant offset to stacked, sky-subtracted LAE spectra until the Lyα troughs are non-negative and interpreting that offset as the full incident UVB (Eq. 3). This identification requires that the absorbing H I be optically thick across the exact wavelength range used to define the trough and that the absorber cover the entire effective aperture (9.85 arcsec) used in Eq. 4. Neither condition is measured. Section 4.1 states: “We first assume that the absorption of the UVB by an LAE is saturated, as in a DLA. We note that we have not fully quantified the properties and physical extent of the absorbing gas”; Section 6 concedes that “our physical model of the absorption is not complete.” If the covering fraction is f < 1, the observed deficit is f times the incident UVB, so the reported intensity is too low by 1/f; if the trough minimum is not fully saturated, the offset is a lower limit. The density-dependent trend in Figure 5 is a relative measurement and does not calibrate this absolute factor. Thus the headline intensity is model-normalized, not directly measured.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript uses stacks of roughly 400,000 HETDEX Ly-alpha emitter spectra to measure the z~2.4 rest-frame ultraviolet background light. It argues that the negative Ly-alpha absorption troughs around the emission line are produced by saturated, DLA-like H I absorption of the full UV background, and that HETDEX's sky subtraction removes that background, so the additive offset required to raise the troughs to zero equals the observed-frame EBL. The authors report a density-weighted average of 12.9 +/- 3.7 nW m^-2 sr^-1 at a median observed wavelength of 4134 Angstroms, corresponding to a rest-frame UVB of 508 +/- 145 nW m^-2 sr^-1 at z~2.4, and find that the inferred intensity scales with local galaxy density. They compare the result with the Haardt & Madau (2012) UVB model and with direct optical EBL/COB measurements, concluding rough consistency.","tokens_in":13933,"tokens_out":4195,"duration_ms":39558,"significance":"If the physical model is validated, this would be a novel and potentially important probe of the z~2-3 ultraviolet background and its small-scale anisotropy, exploiting a unique HETDEX dataset. The paper has notable strengths: a very large sample, a careful stacking procedure, a direct pre-sky-subtraction comparison in Figure 3, a useful density-dependent trend in Figure 5 that serves as a consistency check, and an unusually candid statement of limitations. However, the absolute calibration of the headline intensity rests on untested assumptions about absorber saturation and covering fraction, and no forward model or independent calibration is presented. The result is therefore best understood at present as a model-normalized estimate rather than a direct measurement, so the central claim needs additional support before publication.","major_comments":[{"comment":"The central identification of the additive offset with the full EBL requires that the absorbing H I be saturated and that it fully cover the effective aperture, but Section 4.1 explicitly states that 'we have not fully quantified the properties and physical extent of the absorbing gas' and Section 6 concedes that 'our physical model of the absorption is not complete.' If the covering fraction is f < 1, the observed deficit equals f times the incident UVB, so the reported 12.9 nW m^-2 sr^-1 is too low by a factor 1/f; if the trough minimum is not optically thick, the offset is only a lower limit. The density trend in Figure 5 is a relative measurement and cannot calibrate this absolute factor. Please provide an external constraint on the covering fraction and optical depth, or explicitly present the absolute intensity as a lower limit rather than a measured value.","section":"§4.1, Eq. (3)"},{"comment":"The assumption that the co-added LAE spectra have no detectable stellar continuum is load-bearing for the absolute offset. The paper states that shifting the full-sample stack to zero requires +0.063 x 10^-17 erg s^-1 cm^-2 Angstrom^-1, whereas shifting to the detected continuum requires +0.084 x 10^-17, about 33% larger. Figure 2 shows detectable continuum in the large stack, so the assertion that the z~2.4 stacks used for the measurement have no continuum must be quantified for those exact stacks, and the associated systematic uncertainty must be propagated into the reported 12.9 +/- 3.7 nW m^-2 sr^-1.","section":"§4.1, Figure 2"},{"comment":"No forward model or independent calibration is presented that connects the measured trough depth to the incident UVB intensity. The interpretation is imported from Paper I, and Eq. (3) makes the EBL equal to the additive offset by construction, so the physical meaning of the offset is not independently tested in this manuscript. A radiative-transfer or mock-stacking model of the expected trough shape and amplitude, or a cross-check using known background sources behind H I absorbers, is needed to separate the EBL intensity from the absorber properties and to validate the absolute normalization.","section":"§4.1, §6"}],"minor_comments":[{"comment":"There are typographical errors: Section 3.1 contains 'are are not purely' and Section 5 contains 'our value only does not include'; also the abstract and Section 4.1 use 'as a' inconsistently in a few places.","section":"§3.1, §5"},{"comment":"Please define the quantity 9.85 arcsec explicitly: is it a diameter, a radius, or the effective area solid angle of the aperture? The unit conversion in Eq. (4) would be clearer if the solid angle of the effective aperture were written out and the intermediate units shown.","section":"Eq. (4)"},{"comment":"The uncertainty estimates on the EBL values in the right panel of Figure 5 are not described; please state how the 1-sigma errors were computed (e.g., bootstrap, jackknife, or propagation of the stack noise) and whether they include systematic contributions from the offset determination.","section":"Figure 5, right panel"},{"comment":"The choice of luminosity-function normalization bins (0-0.3, 0.95-1, 1.5-1.6) should be justified in more detail; it is unclear why the average-density bin does not include the range 1.0-1.5 and how uncertainties in the field-to-field LF normalization affect the bin assignment.","section":"§4.2"}],"recommendation":"major_revision","confidential_remarks":"The authors are commendably explicit about the assumptions and limitations, but the headline absolute measurement is not yet robust because it reduces to an additive offset under untested saturation and covering-fraction assumptions. The paper would be much stronger if the authors reframed the result as a lower limit or added quantitative constraints on the absorbing gas. The heavy reliance on Paper I for the physical interpretation may also warrant an independent check of that model during review. This is a promising measurement but not yet suitable for publication in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the paper is worth reading and refereeing, but the headline EBL intensity should not be taken at face value. The idea is genuinely new—using the negative Lyα troughs in stacked HETDEX LAE spectra as a probe of the z~2.4 UVB, where the trough depth is set by oversubtracted background flux. The stacking is careful, the troughs are real (Paper I did the heavy lifting of ruling out instrumental causes), and the environmental trend—deeper troughs in overdense fields—is a good consistency check that fits the absorption scenario.\n\nThe soft spot is exactly the one the authors flag in Section 6: the physical model of the absorber is not complete. Turning trough depth into an absolute EBL intensity requires (a) saturated, DLA-like absorption, (b) full covering of the ~9.85\" effective aperture, and (c) no stellar continuum in the trough region. None of these is measured. The paper states the gas properties are 'not fully quantified,' which is honest but it undermines the absolute number. If the covering fraction is below one, the reported 12.9 nW/m²/sr is too low by 1/f; if the absorption is not fully saturated, the offset is a lower limit. So the headline value is a model-normalized offset, not a direct measurement.\n\nThe comparison with Haardt & Madau (2012) and the COB measurements is also loose—they compare a z~2.4-only contribution to total COB values with a rough scaling, and 'consistent within an order of magnitude' does a lot of work. The density-dependent trend in Figure 5 is more robust because it is relative, but even there the LF normalization comes from an unpublished paper, which blocks independent checking.\n\nTo be useful as a measurement, the paper needs a radiative-transfer model of the absorber, a covering fraction estimate, and a public release of code and data. That said, it deserves a serious referee; the flaws are addressable with revision. I would not cite the absolute value yet, but the environmental trend is worth watching.\n\nRecommendation: send to peer review, but with the expectation of major revisions and a clear request to either soften the claims or provide the forward model.","headline":"A genuinely new probe of the z~2.4 UVB, but the headline EBL intensity is an assumed offset rather than a measured value; the density trend is the more solid result.","tokens_in":14515,"tokens_out":2901,"would_cite":false,"duration_ms":25091,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The negative Ly-alpha troughs in stacked HETDEX LAE spectra measure the z~2.4 ultraviolet background light: 12.9 ± 3.7 nW m⁻² sr⁻¹ at 4134 Å.","keywords":["Lyman-alpha galaxies","extragalactic background light","ultraviolet background","spectral stacking","HETDEX","damped Ly-alpha systems","sky subtraction","high-redshift galaxies"],"falsifier":"Stack a sample of LAEs with measured H I column densities (from damping-wing fits or 21-cm absorption) and check whether the additive offset required to make the troughs non-negative is independent of column density: if the offset varies with column density, the absorption is not saturated and the reported EBL intensity is not the true background. A second check is to build empty-fiber stacks with the identical sky-subtraction pipeline; negative troughs of comparable depth in those stacks would show the signal is a subtraction artifact rather than background absorption.","tokens_in":13454,"feed_emoji":"🔭","tokens_out":11431,"duration_ms":89515,"temperature":0.7,"pith_summary":"This paper claims that the negative Ly-$\\alpha$ absorption troughs seen in stacks of roughly 400,000 HETDEX Lyman-$\\alpha$ emitter spectra are not artifacts: they record ultraviolet background light from behind each galaxy being absorbed by neutral hydrogen in the galaxy's own halo. Because HETDEX's sky subtraction removes the same background light from the off-source sky, the troughs become over-subtracted, and the additive offset needed to lift them to zero equals the incident UV background intensity. The authors measure $12.9 \\pm 3.7$ nW m$^{-2}$ sr$^{-1}$ at a median observed wavelength of 4134 Å, corresponding to a rest-frame UV background of $508 \\pm 145$ nW m$^{-2}$ sr$^{-1}$ at $z \\sim 2.4$, and find that this intensity rises with local galaxy density. If correct, the method turns a sky-subtraction residual into a probe of the extragalactic background light and its variation across the sky.","feed_headline":"Over-subtracted sky in 400,000 spectra pegs the z≈2.4 UV background","feed_subtitle":"Over-subtracted sky in stacked spectra reveals the z≈2.4 ultraviolet background and how it varies with galaxy density.","key_machinery":"The central object is the negative Ly-alpha absorption trough in stacked LAE spectra, produced by saturated H I absorption of the UV background in a damped Ly-alpha system (a high-column-density neutral hydrogen absorber that fully removes the line). The load-bearing identity is that the additive flux offset required to bring the troughs to zero equals the over-subtracted background light, i.e., the EBL. The machinery that makes this measurable is HETDEX's local sky subtraction: a per-shot 'empty' fiber spectrum is constructed from the remaining fibers and subtracted, and after stacking with a weighted biweight statistic the residual trough depth becomes a surface brightness via an aperture-area unit conversion.","core_discovery":"In the DLA-like model the paper adopts, H I gas in and around a foreground Lyman-$\\alpha$ emitter absorbs the diffuse ultraviolet background at Ly-$\\alpha$, so the observed spectrum before sky subtraction contains the background, the galaxy's own emission, and the absorbed part removed. The sky model built from off-source fibers contains the same background plus foreground sky; subtracting it leaves the Ly-$\\alpha$ emission with a negative trough whose depth equals the absorbed background. The paper therefore determines the flux offset in the observed frame that must be added to each spectrum to make the troughs in the stack non-negative, and calls that offset the $z \\sim 2.4$ ultraviolet contribution to the extragalactic background light. Averaging over density-weighted fields gives $12.9 \\pm 3.7$ nW m$^{-2}$ sr$^{-1}$ at 4134 Å, with individual density bins ranging from about 5 to 18 nW m$^{-2}$ sr$^{-1}$; the corresponding rest-frame UVB is $508 \\pm 145$ nW m$^{-2}$ sr$^{-1}$.","pith_inferences":["If the saturated-absorption assumption is correct, the required trough offset should be independent of the absorbing gas column density; splitting the stack by H I column or by impact parameter would directly test this and could turn the method into a column-density-independent EBL probe.","The observed density dependence implies that any 'average' EBL measured through LAE absorption is biased toward overdense regions; correcting for that bias could help reconcile direct cosmic optical background measurements with galaxy-count predictions.","The same negative-trough signature should appear in other large spectroscopic surveys that use local sky subtraction, allowing the over-subtraction of the ultraviolet background to be used as a cosmology probe rather than only a systematic.","The model predicts an anisotropy: LAEs with more background galaxies along the line of sight should show deeper troughs, so trough depth could be cross-correlated with background galaxy counts around each emitter."],"forward_implications":["The method gives an indirect, stacking-based measurement of the $z \\sim 2$--$3$ ultraviolet background from a large untargeted survey, complementing direct measurements that only cover small patches of sky.","The EBL at observed-frame Ly-alpha is not uniform: it scales with the local density of Lyman-alpha emitters, so line-of-sight EBL measurements carry a cosmic-variance term that depends on field density.","Because the measurement stops at each LAE's redshift, the reported value is a lower limit to the total optical EBL; using the model comparison in the paper, which attributes about 60% of the EBL to sources at $z > 2$, the implied total is roughly 67% higher.","The same trough-depth offset can be measured in different redshift bins, yielding a rough spectrum of the EBL at three observed wavelengths with hints of evolution, not just a single number."],"supporting_citations":[{"why":"It establishes the physical model of the Ly-alpha absorption troughs in stacked LAE spectra from which this paper's EBL measurement follows.","marker":"Weiss et al. (2024)"},{"why":"It provides the stacking methodology and the demonstration that the negative troughs appear in large HETDEX LAE stacks.","marker":"Davis et al. (2023b)"},{"why":"It describes HETDEX sky subtraction and calibration, the procedure whose over-subtraction of background light is the basis of the measurement.","marker":"Gebhardt et al. (2021)"},{"why":"It supplies the theoretical UVB model used to compare the measured intensities and to estimate the $z > 2$ contribution to the total EBL.","marker":"Haardt & Madau (2012)"},{"why":"It presents a direct New Horizons optical EBL measurement used as the observational comparison for the reported intensity.","marker":"Lauer et al. (2022)"},{"why":"It provides a newer direct cosmic optical background measurement used to place the result in context.","marker":"Postman et al. (2024)"},{"why":"It is the foundational demonstration that Ly-alpha absorption by foreground neutral hydrogen can probe background light along the line of sight.","marker":"Gunn & Peterson (1965)"},{"why":"It is the source of the damped Ly-alpha absorber geometry and the saturated absorption assumption adopted in the model.","marker":"Wolfe et al. (1995)"},{"why":"It provides the HETDEX LAE catalog and redshifts used to select and stack the sample.","marker":"Mentuch Cooper et al. (2023)"}],"fun_headline_variants":["Sky subtraction reveals z≈2.4 UV background intensity","Negative absorption from 400k spectra measures cosmic UV light","z≈2.4 UV background measured via Lyman-alpha sky over-subtraction","How much UV light at z≈2.4? Stacked Ly-alpha spectra say: 12.9 nW/m²/sr","Galaxy halos cast Ly-alpha shadows that probe the UV background"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The measurement rests on the assumption that the Ly-alpha absorption by each foreground galaxy's hydrogen is saturated, like a damped Ly-alpha system, so that every ultraviolet background photon at that wavelength is removed and the offset needed to fill the trough equals the full background intensity; the paper notes that the properties and physical extent of the absorbing gas are not fully quantified.","fun_headline_variants_meta":{"raw":{"variants":["Sky subtraction reveals z≈2.4 UV background intensity","Negative absorption from 400k spectra measures cosmic UV light","z≈2.4 UV background measured via Lyman-alpha sky over-subtraction","How much UV light at z≈2.4? Stacked Ly-alpha spectra say: 12.9 nW/m²/sr","Galaxy halos cast Ly-alpha shadows that probe the UV background"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000703,"raw_usage":{"total_tokens":3244,"prompt_tokens":1092,"completion_tokens":2152,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":708,"completion_tokens_details":{"reasoning_tokens":2046}},"tokens_in":708,"tokens_out":2152,"duration_ms":13497,"temperature":1.0,"reasoning_tokens":2046,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T12:12:24.382161+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Stack a sample of LAEs with measured H I column densities (from damping-wing fits or 21-cm absorption) and check whether the additive offset required to make the troughs non-negative is independent of column density: if the offset varies with column density, the absorption is not saturated and the reported EBL intensity is not the true background. A second check is to build empty-fiber stacks with the identical sky-subtraction pipeline; negative troughs of comparable depth in those stacks would show the signal is a subtraction artifact rather than background absorption.","supporting_citations":[],"review_version":1}