{"id":"18e0d149-e675-4e62-bfac-f80f9db5a4e0","arxiv_id":"2607.18385","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The H-alpha/[OII]/[OIII] brightness of the inner circumgalactic medium of 72 MaNGA galaxies correlates with star formation rate, not stellar mass, indicating escaping ionizing photons from star formation power the glow.","lead":"This paper maps the faint optical glow around 72 galaxies and finds it is brighter when the galaxy is forming stars faster, regardless of the galaxy's mass. The likely cause is that ionizing light leaking from star-forming regions lights up the surrounding circumgalactic gas.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 5Re boundary is not empirically secured: 2R90≈5.6Re admits HII regions at 5Re, and the BPT evidence allows hard ionization at large radius, so the SFR–Hα correlation may trace disk/scattered/shock emission rather than CGM photoionization.","rationale":"The reader's weakest_assumption correctly identifies the disk-halo boundary and contamination as the load-bearing premise. My analysis agrees and sharpens it: the 2R90 argument is internally in tension with the 5Re choice, the EW toy model covers only a biased subset, and the BPT evidence explicitly leaves hard ionization at large radii, contradicting the paper's strongest anti-shock statement. These are not newly invented concerns; they are present in the manuscript's own text. The positive contributions remain: the sample of 72 individual MaNGA galaxies is unique, the correlation is robust to SFR estimators, and the comparison with GAEA/TNG is a fair illustration of current model divergence. But the title-level claim requires excluding disk/PSF/shock origins, which the current diagnostics do not do. The proposed test—PSF subtraction and rotation-velocity masking—would settle whether the flux at 5Re is truly diffuse CGM or a superposition of disk/scattered light. Until then the conditional verdict is appropriate: the correlation is likely real but its physical interpretation is not established. I therefore retain the reader's CONDITIONAL recommendation rather than moving to accept or reject.","tokens_in":32118,"tokens_out":7718,"duration_ms":70722,"concrete_test":"Re-measure the 5Re Hα surface brightness after two additional treatments: (1) subtract a forward-modeled PSF scattering model built from the central continuum+emission maps, and (2) mask spaxels whose line-of-sight velocity follows the disk rotation curve at >2σ significance (i.e., possible disk HII regions). If the mean 5Re SB drops by >30% or the SFR–Hα slope flattens beyond the reported 0.94±0.08, the inner-CGM attribution fails. As a second check, compute 2R90 for each of the 72 galaxies and flag detections at r<2R90; if a substantial fraction of the 35 detections lie inside this radius, then by the paper's own Zaritsky & Christlein criterion those sources are formally within the empirical HII-region disk zone.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central causal claim requires that the flux at 5–10Re is CGM gas ionized by escaping star-forming photons. The paper's own boundary arguments do not exclude extended-disk or scattered-light contamination. In 'Transition to the CGM' (Methods), the adopted 5Re boundary is justified using (i) 0.05rvir≈5.7 kpc, (ii) the concentration index R90/Re=2.78 so that 2R90≈5.6Re, and (iii) an EW toy model. Argument (ii) undercuts the choice: if HII regions in extended disks can exist out to 2R90≈5.6Re, then the 5Re annulus is inside the empirical HII region limit, not outside it. The EW toy model is fitted to only 38 of 72 galaxies; the other 34 have no measurable EW profiles and low sSFR, so the model does not test high-SFR galaxies where disk contamination would be strongest. The shock-discriminating BPT evidence is limited to two galaxies, and the Methods text for the second galaxy states that line ratios at r>12 kpc are 'consistent with hard ionizing sources such as AGNs or heat shocks'—directly contradicting the §2.1 assertion that both BPT examples lie in the star-forming region. Thus the observed r=0.74 correlation between Hα SB at 5Re and SFR is exactly what an extended-disk, PSF-wing, or shock contribution would produce; the photon-budget estimate (1.9e51 s−1 vs 1.4–2.7e51 s−1) measures the recombination requirement of the observed flux, not its ionizing source. The paper itself admits in §2.1 that shocks cannot be definitively ruled out.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Using MaNGA integral-field data for 72 low-redshift galaxies whose IFU footprints extend to at least 5 Re, the authors construct H-alpha, [O II], and [O III] surface-brightness radial profiles and measure line emission in annuli at 5, 7.5, and 10 Re. They report a break at 1-2 Re, an SFR-dependent slope beyond the break, and strong correlations between the 5-10 Re H-alpha (and [O II]/[O III]) surface brightness and galaxy SFR/sSFR (Pearson r=0.74, p=3.54e-7 at 5 Re), with no significant correlation with stellar mass. A photon-budget argument is used to argue that escaping ionizing photons from central star formation (1-2% escape fraction) can account for the CGM flux. The paper also compares cool-gas surface densities in GAEA and TNG simulations with the observed correlations.","tokens_in":32523,"tokens_out":6722,"duration_ms":57863,"significance":"If robust, this is a valuable result: a large, uniform IFU sample mapping the transition between the disk and the inner CGM would establish that star formation, rather than stellar mass or UV background, regulates the cool ionized gas at 5-10 Re, and would provide a new observational constraint for subgrid feedback models. The paper uses publicly available MaNGA data, provides reproducibility details, and tests the SFR correlation with independent MPA-JHU and GSWLC catalogs. The GAEA/TNG comparison, although qualitative, is a useful demonstration of model dispersion. However, this significance is conditional on the disk-halo boundary and contamination controls; the causal claim currently rests on inference and is explicitly qualified by the authors' own shock caveat.","major_comments":[{"comment":"The 5Re disk-halo boundary is the load-bearing assumption that separates CGM emission from disk emission. The three justifications do not secure it. The concentration argument gives 2R90 about 5.6Re, so the annulus 4.5-5.5Re is inside the radius to which Zaritsky & Christlein (2007) find HII regions, not outside it. The EW toy model is fitted to only 38 of 72 galaxies; the remaining 34 are low-sSFR systems without measurable EW profiles, so the model does not constrain the high-SFR galaxies where extended-disk or PSF-wing contamination would be strongest. No PSF-wing, scattering, or inclination test is provided. Since any of these contamination channels would also produce an f_Halpha-SFR correlation, the Fig. 2 result alone cannot establish the CGM origin. Please add a per-galaxy break-radius analysis, PSF-wing subtraction tests, and an investigation of whether the 5Re flux depends on in","section":"Methods: Transition to the CGM"},{"comment":"The BPT discriminant is internally inconsistent. Section 2.1 states that the BPT diagrams for two galaxies lie in the star-forming region and uses this as the third argument against shocks. Methods, however, says for the second galaxy that line ratios at r>12 kpc are 'consistent with hard ionizing sources such as AGNs or heat shocks.' Because only two galaxies have radially resolved BPT data, this contradiction removes the only direct ionization diagnostic beyond 5Re. The subsequent concession that shocks 'cannot be definitively ruled out' is appropriate, but the abstract's causal claim ('Star formation powers...') is stronger than the evidence allows; it needs either additional ionization diagnostics at 5-10Re or a more carefully bounded conclusion.","section":"Section 2.1 and Methods (second example galaxy)"},{"comment":"The photon-budget argument shows consistency, not causation. The 1.9e51 s^-1 figure is the LyC recombination rate required by the observed Halpha flux under Case B; the 1.4-2.7e51 s^-1 from SFR times (1-2%) shows that central star formation can supply enough photons. Shocks, cooling radiation, or scattered central disk photons would also satisfy the same photon budget. The text seems to acknowledge this in the shock-discussion paragraph, but then concludes that star formation 'fully account[s] for the observed emission line fluxes.' Please rescope the conclusion or provide a quantitative test that distinguishes the candidate mechanisms (e.g., He I/He II ratios, [O I]/Halpha, or a comparison of the predicted and observed slope of f_Halpha versus SFR under each mechanism).","section":"Section 2.1 photon budget"},{"comment":"The x-axis SFRs partially share a tracer with the y-axis because the MaNGA DAP SFRs include Halpha-based estimates. The GSWLC SED-based SFR test is a good check, but the paper reports only that parameters are consistent within 1 sigma, with b1 deviating at 1.5-2.5 sigma. Please provide the GSWLC/MPA-JHU-only slopes and correlation coefficients for all three radii, and ideally show the Figure 2 panels with these alternative SFRs. This would make the claimed independence of the correlation from the shared tracer quantitatively transparent.","section":"Section 2.2, Table 1 and DAP SFRs"}],"minor_comments":[{"comment":"There is a typo 'asympotic' for 'asymptotic' in the display near Eq. (2). The Figure 1 caption refers to 'mean stacked radial profiles' while the text discusses both mean and median stacks; please clarify which quantity is plotted.","section":"Equation (2) and Figure 1 caption"},{"comment":"The adopted Hbeta/Halpha=0.3 for IRAS 08339+6517 is rough. Please state explicitly how much the object's placement in Figure 2 changes for a plausible range of this ratio (e.g., 0.3-0.5), given that the literature comparison is used as supporting evidence.","section":"Section 2.1, IRAS 08339+6517"},{"comment":"The sentence 'There is no evidence for disk emission line sources outside of about 5Re' is too strong. The cited Zaritsky & Christlein (2007) sample is small and does not span the full mass/SFR range of the present sample; a softer wording such as 'in the cited galaxy sample' would be more accurate.","section":"Methods: Transition to the CGM"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is well within the scope of the journal. I see no citation-pattern issues; the authors' own prior stacking papers are heavily cited, but this is justified by the direct lineage. The main risk is that the causal title and abstract outrun the evidence. I recommend major revision rather than reject because the empirical correlation is valuable and the boundary/contamination issue is addressable with targeted analyses."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take on arXiv:2607.18385. The new thing is real: 72 individual, normal MaNGA galaxies mapped to 5-10 Re, showing that Hα, [OII], [OIII] surface brightness at these radii correlates with SFR and sSFR (Pearson r=0.74 for Hα at 5Re) and not with stellar mass. That is a fresh, credible observational result and a useful advance over the earlier stacking work. The robustness check with GSWLC and MPA-JHU SFRs helps; the inclusion of two external galaxies (MUSE HDFS, IRAS 08339) is a nice touch; and the GAEA/TNG comparison is a sensible way to put the measurement in context, even if it is only a gas-density proxy.\n\nThe paper is honest about many of its own limitations — shocks are explicitly not ruled out, the O32 escape fractions carry 1-2 dex scatter, the stacked slope trend is only 3-sigma with 7-11 galaxies per bin. That candor is to its credit.\n\nWhere I'd push back is the title-level claim. The inference that escaping star-forming photons power the inner CGM rests on three legs, and each is wobbly. First, the disk-halo boundary at 5Re: the paper uses R90/Re=2.78, so 2R90≈5.6Re, then says HII regions are confined within 2R90 and therefore 5Re is outside the disk. That is backwards — if HII regions exist out to 5.6Re, then the 4.5-5.5Re annulus is inside the empirical HII region limit, not outside it. The EW toy model is fit to only 38 of 72 galaxies and uses an ad hoc functional form; it gives some support, but it doesn't settle the boundary. Second, the BPT evidence: the methods section for the second example galaxy states that line ratios at r>12 kpc are 'consistent with hard ionizing sources such as AGNs or heat shocks', which contradicts the text in §2.1 claiming both BPT examples lie in the star-forming region. That needs a fix. Third, the photon budget (1.9e51 s−1 required vs 1.4-2.7e51 s−1 available for fesc=1-2%) is a consistency check, not a source identification; it shows the SFR can in principle account for the recombination photons, but doesn't show that it does.\n\nNone of this kills the core correlation. But the title and much of the discussion treat the origin as established, and it isn't yet. The paper would be stronger with the slope trend de-emphasized, the BPT contradiction fixed, and a more explicit treatment of the boundary uncertainty — perhaps splitting the sample by R90 or using the local continuum to flag residual disk emission.\n\nWho's this for? Anyone working on the CGM, escape fractions, or MaNGA emission-line mapping. It deserves serious peer review; I'd send it out. My own verdict: the correlation is likely real and useful, but the causal claim needs reshaping and the boundary question needs more work.","headline":"New MaNGA dataset shows inner-CGM Hα correlates with SFR but the causal claim 'star formation powers it' outruns the evidence; the 5Re boundary is the load-bearing assumption and it is not yet secured.","tokens_in":33155,"tokens_out":4066,"would_cite":true,"duration_ms":34726,"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 ionizing photons that make the inner circumgalactic medium glow come from star formation in the central galaxy, not from shocks or the ultraviolet background.","keywords":["circumgalactic medium","H-alpha emission","star formation rate","Lyman continuum escape","integral field spectroscopy","galaxy halos","ionization","feedback"],"falsifier":"A direct check would be to obtain deep, high-spatial-resolution spectroscopy of a few of these galaxies to search for the signature of extended H II regions, scattered light, or PSF wings at 5–10 Re. Specifically, if one could measure the velocity field of the emission at those radii and find that it rotates coherently with the disk (rather than showing a roughly stationary or outflowing halo component), the disk-origin alternative would be favored. Alternatively, a spatially resolved measurement of the Balmer decrement (H-alpha/H-beta) at 5–10 Re could distinguish recombination from scattered","tokens_in":31944,"feed_emoji":"🌌","tokens_out":6060,"duration_ms":47480,"temperature":0.7,"pith_summary":"This study analyzes optical emission-line maps of 72 normal, low-redshift galaxies observed with integral-field spectroscopy. It finds that H-alpha, [O II], and [O III] surface brightness at radii of 5–10 effective radii—the inner circumgalactic medium—correlates strongly with the galaxy's star formation rate (Pearson r = 0.74) but not with its stellar mass (r = 0.018). The ultraviolet background alone is too faint to produce the observed flux, and shock ionization is argued to be unlikely. A photon budget shows that a 1–2% escape fraction of Lyman-continuum photons from central star-forming regions can supply the required ionizing photons. The most likely source of the inner CGM's line emission is therefore star formation in the central galaxy.","feed_headline":"H-alpha glow around galaxies tracks star-formation rate","feed_subtitle":"A photon-budget match shows 1–2% of escaping ionizing photons from the central disk can explain the inner halo's line emission.","key_machinery":"The central mechanism is the escape of Lyman-continuum photons from the host galaxy's star-forming disk into the halo. The paper identifies a disk-halo boundary at ~5 Re, justified via the 90%-light radius (mean R90/Re = 2.78, so 2R90 ~ 5.6 Re) and the observation that H II regions in nearby galaxies are confined within 2R90. Along with this boundary, the key quantitative tools are (i) the strong linear correlation between H-alpha surface brightness at 5–10 Re and SFR/sSFR, and (ii) a photon-budget comparison between the LyC production rate inferred from SFR (with an assumed 1–2% escape fraction) and the LyC photon rate required by the observed H-alpha luminosity. The authors use [O III]/[O","core_discovery":"The paper's central claim is that the optical line emission observed from the inner circumgalactic medium (the region between 5 and 10 effective radii, roughly 0.1–0.25 virial radii) of normal galaxies is powered by ionizing photons escaping from the star-forming regions in the central galaxy. This is supported by two quantitative results. First, H-alpha surface brightness at 5 Re correlates with star formation rate (r = 0.74, p = 3.54e-7) and specific star formation rate (r = 0.64), but shows no significant correlation with stellar mass (r = 0.018, p = 0.92). Second, for a star-forming galaxy with SFR ~ 1 M_sun/yr, the observed H-alpha flux at 5 Re requires ~1.9 x 10^51 Lyman-continuum phot","pith_inferences":["A testable extension would be to measure the same relation in edge-on versus face-on galaxies: if escaping photons from the central disk are the source, the emission should be roughly isotropic or slightly limb-brightened, not concentrated along the disk plane, which would also help separate true CGM emission from scattered light.","The 1–2% escape fraction inferred here is global; spatially resolved spectroscopy of the inner CGM could map the escape fraction as a function of azimuth and radius, testing whether 'leaky chimneys' associated with star-forming complexes are the dominant escape channels.","If the SFR–surface-brightness correlation is confirmed at even larger radii (beyond 10 Re), it may be possible to relate CGM emission directly to the ionizing photon budget available for reionization at low redshift and to calibrate LyC photon production efficiencies in normal galaxies.","The non-detection of H-alpha in quenched galaxies, despite one simulation's prediction of cool gas at large radii, suggests that either the cool gas in such halos is not ionized (e.g., shielded from the UV background) or the simulation overproduces the cool gas reservoir; this could be tested observationally with deeper, blind surveys."],"forward_implications":["If correct, the line emission from the inner CGM of normal galaxies can be used as a direct tracer of the ionizing photon escape from the central galaxy, effectively measuring the escape fraction across a broad range of galaxy properties.","The absence of a correlation with stellar mass implies that the amount of cool, line-emitting gas in the inner CGM at these radii is regulated by current star formation activity rather than by the integrated galaxy mass.","The observed break in slope at 1–2 Re and the shallower slopes at low SFR indicate that feedback processes shape the inner CGM, with the accumulated star formation history reflected in the gas distribution.","The large discrepancy between the cool-gas–SFR predictions of two current theoretical models means that CGM emission-line surveys can serve as a discriminating test for subgrid models of star formation and feedback.","The inferred escape fractions (1–2%, up to ~35% in extreme starbursts) have implications for the ionizing photon budget of the universe and for reionization, since they quantify how many photons leak from star-forming galaxies into the intergalactic medium."],"fun_headline_variants":["Star formation drives CGM line emission","Galaxy halos glow from their own starlight","H-alpha in CGM scales with star-formation rate","Inner CGM light traces galaxy star formation","CGM line emission powered by escaping ionizing photons"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The claim depends on the assumption that the emission measured at 5–10 effective radii comes from the circumgalactic medium itself, not from the galaxy's extended star-forming disk, scattered disk light, or the wings of the point-spread function.","fun_headline_variants_meta":{"raw":{"variants":["Star formation drives CGM line emission","Galaxy halos glow from their own starlight","H-alpha in CGM scales with star-formation rate","Inner CGM light traces galaxy star formation","CGM line emission powered by escaping ionizing photons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000178,"raw_usage":{"total_tokens":1179,"prompt_tokens":834,"completion_tokens":345,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":578,"completion_tokens_details":{"reasoning_tokens":272}},"tokens_in":578,"tokens_out":345,"duration_ms":3552,"temperature":1.0,"reasoning_tokens":272,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T15:33:07.515807+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct check would be to obtain deep, high-spatial-resolution spectroscopy of a few of these galaxies to search for the signature of extended H II regions, scattered light, or PSF wings at 5–10 Re. Specifically, if one could measure the velocity field of the emission at those radii and find that it rotates coherently with the disk (rather than showing a roughly stationary or outflowing halo component), the disk-origin alternative would be favored. Alternatively, a spatially resolved measurement of the Balmer decrement (H-alpha/H-beta) at 5–10 Re could distinguish recombination from scattered","supporting_citations":[],"review_version":1}