{"id":"89e93cab-945b-4791-97fc-89e550d488f4","arxiv_id":"2608.16665","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The power source of Lyman-alpha halos around high-redshift galaxies switches at about one virial radius from host star formation to environmental regulation.","lead":"Using ultra-deep MUSE observations, the authors mapped Lyman-alpha light around 21 faint galaxies at redshift 3 to 4 out to 50 kiloparsecs. They report that the inner glow tracks each galaxy's star formation rate while the outer glow tracks the surrounding galaxy density, with a transition near the virial radius.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Environmental correlation at r>15 kpc is not yet robust: only one of four outer bins is formally significant, the core-bright-only check is asserted without a censored-data analysis, and a few outer-bright objects may drive the trend.","rationale":"The paper has real strengths: individual ultradeep profiles, careful masking, comparison with the W18 stack, and an honest statement of limitations. The two-domain interpretation is plausible, and the SFR correlation does break around 15–20 kpc. However, the environmental half of the claim is the load-bearing pillar, and it is the weakest statistically. The paper's own Fig. 6 shows only one significant bin; the 'similarity' across bins is not a statistical test. The core-bright-only robustness is asserted in Sect. 5 but not presented, and with 16 objects mostly below S/N=2 in the outer annuli, ordinary regression on measured values, including upper limits treated as detections, can mislead. Selection on S/N15–50kpc can also amplify environment trends, and a few objects may drive the signal. These issues are fixable with censored regression and jackknife tests, but until then the conclusion should remain conditional, as the reader already decided.","tokens_in":33258,"tokens_out":5897,"duration_ms":64564,"concrete_test":"Recompute the Fig. 6 regressions for r>15 kpc using only the 16 core-bright LAEs, treating S/N15–50kpc<2 measurements as upper limits with a censored-data estimator (e.g., Akritas–Theil–Sen or a Buckley–James regression), and repeat after excluding ID 8284 and ID 106. If the core-bright-only slopes are consistent with zero, or if any outer bin loses significance after one-object removal, the environmental-regulation claim is not independently established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that outer Lyα halos are environment-regulated rests on Fig. 6. The paper itself states (Sect. 4.2.2) that the positive SB–δ+1 correlation is formally significant only for the 15–20 kpc annulus (p=0.018); the outer bins give p=0.149, 0.066, and 0.23. Pearson coefficients ignore per-point measurement errors, and the six radial bins are not independent, so the 'taken together' argument is subjective. The stated robustness check that the same trends appear in the core-bright subsample is not shown as a figure or table; only four objects in the combined sample have S/N15–50kpc>2 (Table 1), so the core-bright-only analysis is dominated by non-detections that need survival-analysis treatment rather than ordinary regression. This matters because halo-bright selection is made on the dependent variable: if high-δ environments boost outer SB through the same undetected-neighbour population, the selection preferentially admits high-δ objects and can amplify an apparent trend. Finally, individual outliers (notably ID 8284, and ID 106) may drive the outer-bin correlations; removing them could erase the signal. Until these checks are done, the environmental-regulation (and hence the diffuse-gas) conclusion is conditional.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies extended Lyα surface brightness (SB) profiles of 21 low-luminosity LAEs at 3<z<4 in the MUSE Extremely Deep Field, measuring individual profiles out to 50 kpc (about 2.5 rvir). The sample comprises 16 core-bright and 9 halo-bright objects, selected by central and annular signal-to-noise thresholds. The authors find that the inner SB (r<15-20 kpc) correlates with the host galaxy's SFR and not with environmental density δ+1, while the outer SB (r>15-20 kpc) shows the opposite behavior: no correlation with SFR and a positive correlation with δ+1. They compare the observed profiles to the expected integrated contribution from undetected faint LAEs (SBunb) and to UV-background fluorescence predictions, concluding that neither mechanism can explain the environmental trend, and therefore attribute the outer emission to diffuse circumgalactic gas regulated by the environment. The central claim is the existence of two distinct Lyα halo domains with a turnover near ~1 rvir ≈ 20 kpc.","tokens_in":33524,"tokens_out":5068,"duration_ms":51569,"significance":"If the central claim holds, this is an important step beyond stacking analyses: it demonstrates, on an object-by-object basis, that the inner and outer CGM of faint LAEs respond to different physical drivers, with the transition at the virial radius. The paper's strengths include the use of uniquely deep MXDF data with careful PSF modeling, noise treatment, and an honest discussion of uncertainties. The SBunb and UVB fluorescence predictions are computed from externally published parameters rather than fitted to the data, and the analysis is largely reproducible from public catalogs. However, the environmental-regulation conclusion rests on a correlation that is formally significant in only one of four outer radial bins and on a robustness claim that is asserted but not demonstrated. The statistical and selection-effect questions are therefore load-bearing, and the paper needs additional analyses before the two-domain interpretation can be considered established.","major_comments":[{"comment":"The central claim of environmental regulation in the outer halo is supported by a Pearson correlation that is formally significant only in the 15-20 kpc annulus (p=0.018), with p=0.149, 0.066, and 0.23 in the three outer annuli. The statement that the four panels 'taken together' indicate a real effect is not backed by any combined statistical test. I request a joint test across the four outer bins (e.g., a permutation test on a combined statistic, a mixed-effects model, or a meta-analytic combination of the four p-values) that accounts for the non-independence of annuli within each object, and a multiple-comparison-aware interpretation.","section":"§4.2.2, Fig. 6"},{"comment":"The halo-bright sample is selected by S/N15-50kpc>2, which is a selection on the dependent variable used for the environmental correlation. The paper states in §5 that 'the same analysis on core-bright sample only reveals the same trends,' but this analysis is not shown. Table 1 gives S/N15-50kpc values for the 16 core-bright objects: only four (2726, 4842, 400, 6297) exceed 2, so the core-bright-only sample is dominated by non-detections or low-S/N measurements in the outer annuli, where ordinary Pearson regression on the reported SB values is not appropriate without censored-data methods. Please present the core-bright-only version of Fig. 6 using survival analysis (e.g., a generalized Kendall tau or ASURV-style estimator) and also report the correlations with the four overlapping halo-bright objects removed.","section":"§2.2.2, §5, Table 1"},{"comment":"The claim that undetected faint LAEs are 'insufficient to drive the observed correlation' with environment is not directly tested. Fig. 5 shows that at 37-50 kpc many measured SB values are consistent with SBunb within 1σ, so the two-halo term is not negligible in the outermost bin. Since SBunb itself depends on the clustering boost ζ and therefore on environment, the positive SB vs δ+1 trend in Fig. 6 could be partly or wholly a reflection of the undetected-neighbor contribution. I request a plot and correlation test of the residual SBobs - SBunb (or SBobs/SBunb) versus δ+1 in the four outer bins, to demonstrate that the environmental correlation survives subtraction of the expected discrete-source contribution.","section":"§4.2.1 and §5.2, Fig. 5"},{"comment":"The outlier sensitivity of the environmental correlation is not quantified. The paper itself identifies ID 8284 (and, to a lesser degree, ID 106) as objects whose outer emission is likely boosted by close companions or complex structure. The 37-50 kpc panel has only p=0.23 even with ID 8284 present; removing it or ID 106 could erase the trend in that bin. Please provide a jackknife or bootstrap analysis over objects, and report the correlation coefficients and p-values with IDs 8284 and 106 excluded, to demonstrate that the outer-halo correlations are not driven by one or two systems.","section":"§5.2, Fig. 6"}],"minor_comments":[{"comment":"There are several typos and formatting issues: 'agress' in §3.2.2, 'derivarions' in §6, 'probablility' in the Fig. 6 caption, 'negligble' in §4.3.1, 'pa star formation' in §5.2, and the Table 1 note 'S/N 15−10kpc' which should read 'S/N 15−50kpc'.","section":"Throughout"},{"comment":"ID 106 appears twice with different central-component parameters; please clarify in a footnote that the two rows correspond to the two-component and elongated-halo fits described in §3.2.2, or merge them for clarity.","section":"Table 2"},{"comment":"The text says the environmental correlation is 'formally significant at more than 5% confidence only for the 15-20 kpc annulus'; this is an honest statement, but the subsequent 'taken together' reasoning should be replaced or supplemented by the quantitative combined test requested above.","section":"§4.2.2"},{"comment":"The text describes R as 'photon production per unit time, surface and solid angle'; this is dimensionally unclear. Please rephrase to 'per unit time' or provide the full expression with the proper factors (R/4πr²).","section":"Appendix D, Eq. (D.2)"},{"comment":"The stated sensitivity of SBunb to the faint-end slope α (factor ~3 for α=−1.84) is important; consider adding a short paragraph in §5.2 quantifying how an α−=1.84 would change the residual correlation test proposed in the major comments, rather than only dismissing the steeper slope as 'unlikely'.","section":"Appendix C"}],"recommendation":"major_revision","confidential_remarks":"The paper is a good fit for A&A and uses a valuable, unique dataset. The main concern is statistical robustness rather than methodological soundness: the central environmental correlation is weak in individual bins, the core-bright robustness check is not shown with censoring, and the claim that discrete neighbors are excluded needs a residual-based test. With the requested additional statistics, the paper could become a solid contribution. The extensive reliance on related work from the same group (HA23, Pharo et al. in prep) is appropriate given the direct methodological continuity; no novelty-disclosure concern arises."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Read this one before citing it. It is the first object-by-object measurement of Lyα halo surface brightness beyond roughly one virial radius for intrinsically faint LAEs, and that novelty is real. The data work is careful: pseudo-narrowband extraction, galfit with HST priors, explicit masking, public MXDF data, and a stack that matches Wisotzki et al. 2018. The inner-halo SFR correlation is solid (p≈0.004–0.016 in the first three bins) and clearly dies by 20–27 kpc; the environmental correlation reverses sign around 15–20 kpc. I think the two-domain picture, with a turnover near 1 rvir, is plausible and worth taking seriously.\n\nThe soft spots are exactly where the stress-test note lands. In Fig. 6 only the 15–20 kpc bin reaches p=0.018; the outer three bins give p=0.149, 0.066, and 0.23. Pearson coefficients ignore the measurement errors, and the bins are not independent. The 'they all look similar, taken together' argument is suggestive, not a significance test. The halo-bright sample is selected on the dependent variable (S/N>2 in 15–50 kpc), and the claim in Sect. 5 that core-bright-only shows the same trends is asserted without a figure or table. With only four objects in the union having S/N>2 in that annulus, the core-bright check is mostly upper limits and needs survival analysis or censored regression, not ordinary Pearson. Outliers matter too: ID 8284, and to a lesser extent ID 106, can pull the outer bins; I do not see an analysis with them removed.\n\nThe diffuse-gas conclusion is more fragile than the abstract implies. It leans on the Lyα luminosity function faint-end slope and the clustering boost; the authors use externally published parameters and discuss the main uncertainty, which is honest, but the margin is not huge. The UVB fluorescence upper limits are acknowledged as roughly 1σ, which is fine. I also see no circularity problem: the predictions are not fitted to the observed profiles.\n\nWho should read it: people working on CGM physics, Lyα halos, and MUSE deep-field science. The new empirical result—individual outer-halo diversity and the correlation reversal—should not get lost. But the environmental-regulation claim needs either better statistics (more bins, censored-data methods, outlier removal) or a much more muted abstract. Send it to referees, and ask for those tests.","headline":"First object-by-object look at faint LAE outer halos, with a plausible two-domain picture, but the environmental side rests on one significant bin and a selection subtlety—worth refereeing, with serious revisions.","tokens_in":34135,"tokens_out":2857,"would_cite":true,"duration_ms":30505,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Lyα halos switch from star-formation powering to environmental regulation near one virial radius.","keywords":["Lyman-alpha halos","circumgalactic medium","intergalactic medium","surface brightness profiles","galaxy environment","high-redshift galaxies","virial radius","two-halo term"],"falsifier":"One decisive check is to build an environment-unbiased sample of core-bright LAEs with individually detected 20-50 kpc emission at the sensitivity of the present survey and re-measure the SB-$\\delta+1$ correlation: if it does not appear there, the claimed environmental regulation fails. A second complementary test is radially resolved Lyα line spectroscopy: outer-halo emission that showed the same outflow kinematics as the central line, with no blueshift, would rule out the inflow/cooling interpretation that the paper offers for the outer domain.","tokens_in":33034,"feed_emoji":"🌌","tokens_out":9053,"duration_ms":79416,"temperature":0.7,"pith_summary":"This paper uses ultradeep integral-field spectroscopy to measure individual Lyman-alpha halos around 21 intrinsically faint galaxies at redshift 3 to 4 out to 50 kpc, and argues that the halos split into two distinct radial domains at about one virial radius (~20 kpc). Inside that radius the surface brightness is driven by the host's star formation rate and is oblivious to the surrounding environment; outside it the correlation flips, with no SFR dependence but a positive correlation with the local density of Lyman-alpha emitters. The paper further argues that neither individually undetected neighboring galaxies nor ultraviolet-background fluorescence can supply the observed outer emission, so the outer halo must be diffuse gas whose amount or excitation is environmentally regulated. If right, this locates the CGM/IGM transition at a concrete radius and redirects theoretical work on outer halos toward accretion and cooling radiation rather than aggregated satellites.","feed_headline":"Lyα halos switch power source at one virial radius","feed_subtitle":"Inner halo follows star formation; outer halo tracks the surrounding galaxy density instead.","key_machinery":"The load-bearing object is the azimuthally averaged Lyα surface-brightness profile split into six radial apertures from 0 to 50 kpc, with a fixed outer annulus 15-50 kpc used to define 'halo-bright' objects. The argument runs on the contrast between two correlation sequences across those apertures: SB versus SFR and SB versus environmental density $\\delta+1$, estimated from a kernel-smoothed redshift distribution of LAEs. The turnover at ~1 $r_{\\mathrm{vir}}$ is what carries the two-domain conclusion. The quantitative exclusion of discrete neighbors uses the predicted integrated SB of undetected LAEs, $\\mathrm{SB}_{\\mathrm{unb}}$, built from the Lyα luminosity function, the survey's detection selection function, and a halo-occupation clustering boost, evaluated object by object; the exclusion of fluorescence uses the optically-thick-cloud Lyα SB relation with a modern UVB photoionization rate and unity covering fraction as an upper limit.","core_discovery":"The central claim is that the radial surface brightness profile of a Lyα halo encodes a regime change rather than a single exponential plus a smooth 'two-halo' addition. For radii below roughly 15-20 kpc the observed SB correlates with the SFR of the central galaxy (Pearson r about 0.56-0.63) and shows no correlation with environmental overdensity $\\delta+1$; beyond ~20 kpc the SFR correlation is lost (r about 0.06, -0.04, -0.34 in the outer three annuli, none significant) while a positive correlation with $\\delta+1$ emerges in every outer annulus. Since three independent mass estimates place the typical virial radius at ~20 kpc, the authors identify the turnover at ~1 $r_{\\mathrm{vir}}$. They then test two external power sources quantitatively: the integrated light from undetected faint LAEs is predicted to be several times below the measured SB inside 37 kpc and only approaches it in the outermost bin, and UVB fluorescence at maximum covering factor contributes at most about 1-$\\sigma$. The paper's conclusion is that the outer Lyα halo is mostly diffuse cool gas beyond the virial radius, regulated by the environment, not a superposition of unresolved galaxies.","pith_inferences":["The turnover radius may scale with halo mass; if so, studying halos in units of $r/r_{\\mathrm{vir}}$ rather than kiloparsecs should unify low- and high-mass LAE populations.","A testable extension would compare halo-bright objects with a volume-limited, environment-blind sample to determine how much of the reported correlation is selection-driven.","If the outer halo is accretion or cooling radiation, its ratio to the inner SFR-powered halo should increase with halo mass, predicting brighter outer halos around more massive emitters such as Lyman-break galaxies and AGN.","The lack of an internal correlation inside 15 kpc suggests that environment acts on the gas reservoir rather than on ongoing star formation, which could explain part of the scatter in Lyα escape fractions without invoking radiative-transfer complexity."],"forward_implications":["Outer-halo surface brightness beyond ~1 $r_{\\mathrm{vir}}$ is not a star-formation indicator; it should be interpreted as an environmental tracer.","Stacked profiles that flatten near 30-50 kpc should be read as the onset of the environmentally regulated outer domain rather than as domination by unresolved satellites.","Radially resolved spectra should show systematically different Lyα kinematics, such as bluer emission, outside ~1 $r_{\\mathrm{vir}}$ if the outer halo is inflow- or cooling-driven.","Environmental density becomes a predictor of outer-halo extent, allowing surveys to map gas reservoirs around low-mass galaxies without detecting the gas directly.","Simulations of the CGM/IGM interface now have a fixed physical radius, ~1 $r_{\\mathrm{vir}}$, against which to calibrate the transition from stellar-powered to environment-regulated Lyα emission."],"supporting_citations":[{"why":"Supplies the ultradeep integral-field data, the catalog of Lyα emitters, and the SFR/stellar-mass estimates used for every measurement.","marker":"Bacon et al. 2023"},{"why":"Provides the stacked Lyα surface-brightness profile of faint LAEs to which the sample's median profile is compared.","marker":"Wisotzki et al. 2018"},{"why":"Supplies the formalism and clustering-boost factor used to predict the integrated contribution of undetected neighboring LAEs to the outer SB.","marker":"Herrero Alonso et al. 2023b"},{"why":"Establishes the UVB fluorescence surface-brightness relation and covering-fraction assumptions used to bound the fluorescence contribution.","marker":"Gallego et al. 2021"},{"why":"Provides the updated metagalactic UV background photoionization rates used in the fluorescence calculation.","marker":"Puchwein et al. 2019"},{"why":"Defines the faint-end slope of the Lyα luminosity function adopted for the undetected-LAE prediction.","marker":"Tornotti et al. 2025"},{"why":"Supplies typical Lyα halo scale lengths and virial-radius estimates for low-luminosity LAEs at these redshifts.","marker":"Leclercq et al. 2017"},{"why":"Reports stacked profiles from the same deep field showing the same transition near 1 rvir, providing independent support for the turnover radius.","marker":"Guo et al. 2024b"}],"fun_headline_variants":["Lyα halos flip from star formation to environment at one virial radius","At one virial radius, Lyα halos change their power source","Lyα halos: inner glow tracks stars, outer glow tracks density","Beyond a virial radius, Lyα halo light follows the environment","Lyα halos split at virial radius: stars vs surroundings"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the halo-bright selection: objects were admitted by their signal-to-noise in the 15-50 kpc annulus, and the environmental correlation is measured on that sample; if selecting on outer surface brightness also selects overdense sightlines, the turnover at ~1 $r_{\\mathrm{vir}}$ could be partly a selection effect rather than purely a physical regime change.","fun_headline_variants_meta":{"raw":{"variants":["Lyα halos flip from star formation to environment at one virial radius","At one virial radius, Lyα halos change their power source","Lyα halos: inner glow tracks stars, outer glow tracks density","Beyond a virial radius, Lyα halo light follows the environment","Lyα halos split at virial radius: stars vs surroundings"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000846,"raw_usage":{"total_tokens":3799,"prompt_tokens":1179,"completion_tokens":2620,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":795,"completion_tokens_details":{"reasoning_tokens":2525}},"tokens_in":795,"tokens_out":2620,"duration_ms":17613,"temperature":1.0,"reasoning_tokens":2525,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-27T20:32:52.848116+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"One decisive check is to build an environment-unbiased sample of core-bright LAEs with individually detected 20-50 kpc emission at the sensitivity of the present survey and re-measure the SB-$\\delta+1$ correlation: if it does not appear there, the claimed environmental regulation fails. A second complementary test is radially resolved Lyα line spectroscopy: outer-halo emission that showed the same outflow kinematics as the central line, with no blueshift, would rule out the inflow/cooling interpretation that the paper offers for the outer domain.","supporting_citations":[],"review_version":1}