{"id":"90e6cf8b-d595-4695-96a6-e773fcb5cf38","arxiv_id":"1908.03199","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"HATLAS J084933-W, an unlensed hyperluminous infrared galaxy at z=2.41, hosts a broad-line Type-1 AGN with M_bh ~ 2e9 M_sun and X-ray luminosity ~1.4e45 erg/s, yet shows no corresponding SED excess.","lead":"New observations of a hyperluminous starburst galaxy at redshift 2.4 reveal a supermassive black hole, about 2 billion times the Sun's mass, actively accreting gas. Its energy output rivals the galaxy's intense star formation, yet it is almost invisible in the galaxy's ultraviolet and optical light, puzzling astronomers.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Type-1 AGN detection is secure, but the claim that its bolometric luminosity rivals the starburst rests on an untested alpha_OX conversion that the observed SED does not independently anchor.","rationale":"The paper's detection of a Type-1 AGN in HATLAS J084933-W is well supported: broad H-alpha with FWHM ~9700 km/s is a classic broad-line-region signature, the X-ray luminosity of 1.4 x 10^45 erg/s far exceeds plausible star-formation limits, and the X-ray source positionally coincides with component W. The X-ray spectral slope is typical of quasars. I do not see a credible challenge to the core AGN detection. The reader's weakest_assumption correctly identifies the alpha_OX-based bolometric luminosity as the most fragile link. This quantity enters the headline claim that the accretion luminosity rivals that of the starburst, and the three explanatory scenarios in Section 4 are built around that parity. The conversion is a two-step scaling relation calibrated on unobscured quasars, with ~0.3 dex intrinsic scatter. For a source whose SED shows no UV/optical AGN excess, applying the standard relation is least secure: either the AGN continuum is hidden by dust, making the observed optical photometry useless for checking alpha_OX, or the relation does not apply, making L_bol overestimated. Footnote 8 explicitly acknowledges the alternative alpha_OX ~ 1.1 obtained from the observed optical SED, yielding a lower bolometric luminosity, so the paper is aware of the degeneracy. The stated factor-of-two uncertainty comes only from scatter in the scaling relations and does not include this systematic model uncertainty. The 2-sigma N_H detection further weakens the 'little intrinsic absorption' narrative, but that is secondary to the bolometric concern. These concerns do not invalidate the paper; they make the quantitative interpretation more model-dependent than the text suggests. The proposed Balmer-decrement measurement is a decisive, feasible check that would break the degeneracy. Consequently, the reader's ACCEPT verdict stands unchanged.","tokens_in":19298,"tokens_out":5355,"duration_ms":64415,"concrete_test":"Measure the Balmer decrement of the broad emission lines: obtain deep rest-frame optical spectroscopy covering H-beta (observer-frame H band) for HATLAS J084933-W. The observed broad H-alpha/H-beta ratio, corrected with a standard extinction law, yields A_V directly. If A_V is large (>2-3 mag), the alpha_OX-based L_bol can remain viable because the 250 nm AGN continuum is obscured; if A_V is small (<1 mag), then the 250 nm flux predicted by the alpha_OX relation should be visible in the HSC/optical photometry, and its absence would disfavor the standard alpha_OX conversion and lower the inferred L_bol. Either outcome would discriminate between the cavity scenario of Section 4.1 and the alternative geometries, and would also test the consistency of the X-ray-derived absorption column.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The claim that the AGN's bolometric luminosity (1.6e47 erg/s) rivals the starburst L_IR rests on converting the 2-10 keV luminosity to a rest-frame 250 nm luminosity via the alpha_OX relation of Just et al. (2007), then applying a bolometric correction of 110x from Richards et al. (2006). This two-step scaling is calibrated on unobscured, optically selected quasars. HATLAS J084933-W is not such an object: its rest-frame UV-optical-mid-IR SED is consistent with an SMG and lacks the AGN excess seen in APM 08279+5255 and BR 1202-0725 (Fig. 5). The paper acknowledges in footnote 8 that using the observed optical magnitudes gives alpha_OX ~ 1.1 and a lower L_bol, but dismisses this because it would imply heavy UV obscuration. That is not an independent check: if the AGN UV/optical continuum is obscured, the observed magnitudes measure the host galaxy, not the AGN, and the alpha_OX-based value is untestable with current photometry. Because the three scenarios in Section 4 are framed around L_bol ~ L_IR, an error in this conversion weakens the central interpretative claim even though the AGN detection itself is secure. Additionally, the X-ray-derived N_H = (5 +/- 3) x 10^21 cm^-2 is only a 2-sigma detection (Delta chi^2 = 5.9 for N_H = 0), so 'little intrinsic absorption' is also poorly supported and feeds the same extinction accounting. This is model-dependent quantitative inference, not an internal inconsistency.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports new ALMA, XMM-Newton, and VLT/KMOS observations of the z=2.41 hyperluminous infrared galaxy HATLAS J084933.4+021443. The ALMA data confirm a fifth dusty, gas-rich protocluster member, component E. The KMOS rest-frame optical spectroscopy of the brightest component, W, reveals a very broad H-alpha line (9700 km/s FWHM), and the XMM-Newton data reveal a point-like X-ray source coincident with W. The authors conclude that W hosts a Type-1 AGN with a black hole mass of about 2e9 solar masses, a modest intrinsic column of NH ~5e21 cm^-2, and a 2-10 keV luminosity of 1.4e45 erg/s. Using standard alpha_OX and bolometric-correction scaling relations, they estimate a bolometric AGN luminosity of 1.6e47 erg/s, comparable to the starburst's L_IR, despite the absence of obvious AGN signatures in the rest-frame UV-optical-mid-IR SED. Three scenarios are proposed to explain this configuration: an AGN viewed through a cavity in the host disk, an unseen companion AGN, or an ejected supermassive black hole.","tokens_in":19829,"tokens_out":5943,"duration_ms":64777,"significance":"The detection of a broad-line Type-1 AGN in an unlensed, intrinsically hyperluminous starburst is an important observational result, and the paper makes a strong case for it using independent rest-frame optical and X-ray diagnostics. The measurement of an approximately 2e9 solar mass black hole and the apparent absence of strong X-ray absorption in such a gas-rich system are notable. The data and analysis are generally careful, and the authors are explicit about the statistical weakness of the intrinsic column density detection. The value of the paper would be increased if the bolometric-luminosity claim, which underpins the central 'conundrum', were presented with fuller treatment of the systematic uncertainties in the adopted scaling relations.","major_comments":[{"comment":"The statement that the intrinsic absorption is 'little' is based on NH = (5 +/- 3) x 10^21 cm^-2, which is only a 2-sigma detection (Delta chi^2 = 5.9 for NH = 0). The 3-sigma upper limit is 1.2e22 cm^-2. The authors do note in Section 3 that the absorption is only marginally significant, but the abstract and later discussion use the low absorption as a key part of the interpretation. The paper should consistently state that the column is poorly constrained and that the comparison with typical SMGs should be read as an order-of-magnitude statement rather than a precise measurement.","section":"Section 3, bolometric luminosity estimate"}],"minor_comments":[{"comment":"The observing programme identifier appears as '094. A - 0214(A)'; the spacing around the dash should be corrected.","section":"Section 2.1"},{"comment":"The row 'FWHM CO J=1-0 7.3 +/- 1.8 kpc' is confusing because FWHM normally denotes a velocity width in this context; please label the quantity as a spatial extent or use a different notation.","section":"Table 1"},{"comment":"The black hole mass estimate from Schulze et al. (2018) is quoted without giving the adopted virial relation or the continuum/luminosity quantity used; please provide the formula and the systematic uncertainty, given the extreme 9700 km/s line width.","section":"Section 3"},{"comment":"The sentence about the blue [N II] line having 'a fixed wavelength offset and flux ratio (3.06-1x)' is awkwardly worded; it should say that the red-to-blue [N II] flux ratio is fixed at 3.06.","section":"Figure 1 caption"},{"comment":"The statement that the AGN would be 'roughly 5 mag fainter than the prediction for the AGN before obscuration' would benefit from stating the reference magnitude or wavelength used for the prediction.","section":"Section 4.1"}],"recommendation":"major_revision","confidential_remarks":"The core AGN detection is solid and the paper is well within the scope of the journal. My main concern is the framing of the bolometric luminosity as 'rivals the starburst' when it depends on scaling relations that are not independently anchored for this source. I believe the paper can be made suitable for publication by reframing the claim as model-dependent and adding a robustness discussion, rather than by additional observations."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real news is the broad H-alpha and X-ray detection: HATLAS J084933-W hosts a Type-1 AGN. That part is solid. The KMOS spectrum shows a ~9700 km/s broad component, the X-ray source is positionally coincident with W, and the X-ray luminosity is far above what star formation can produce. The fifth protocluster member, E, is also a straightforward, useful addition. I would not quibble with the AGN detection or the black hole mass estimate.\n\nThe soft spot is the headline that the AGN bolometric luminosity rivals the starburst. That number comes from converting the 2-10 keV luminosity to rest-frame 250 nm via the Just et al. alpha_OX relation, then applying a Richards et al. bolometric correction. Both are calibrated on unobscured, optically selected quasars. This object's UV-to-mid-IR SED is consistent with an SMG and shows no AGN excess. The paper acknowledges in footnote 8 that using the observed optical magnitude would give alpha_OX ~ 1.1 and a lower L_bol, but dismisses that because it would imply heavy UV obscuration. That is not an independent check. If the AGN continuum is obscured, the observed magnitudes measure the host galaxy, and the alpha_OX-based value is untestable with current photometry. So L_bol ~ L_IR is a model-dependent estimate, not a measurement. Also, the intrinsic N_H = (5 +/- 3) x 10^21 cm^-2 is only a 2-sigma detection; \"little intrinsic absorption\" is weaker than the abstract implies.\n\nThe three scenarios in Section 4 are clearly labeled speculation, which is fine, but the first scenario leans on the same bolometric-luminosity estimate. None of this undermines the core detection. The paper is a single-object study with no new framework, but it is a clean example of a Type-1 AGN that is invisible in the rest-frame UV-optical-mid-IR SED, and that is worth having on the record. The SED comparison with SMGs and with APM 08279+5255 and BR 1202-0725 is informative. The external scaling relations are used as benchmarks, not fit outputs, so there is no circularity problem.\n\nBottom line: this deserves a serious referee. The main quantitative claim should be presented with its caveats in the abstract, and the N_H significance should be stated honestly, but the AGN detection and the protocluster member are solid. I would send it to review.","headline":"The broad-line AGN detection in HATLAS J084933-W is solid and worth reporting; the bolometric-luminosity comparison with the starburst is a scaling-relation estimate that the SED does not independently anchor.","tokens_in":20274,"tokens_out":2204,"would_cite":false,"duration_ms":23940,"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":"Black hole in a dusty starburst rivals the starburst's power","keywords":["hyperluminous infrared galaxies","active galactic nuclei","broad-line region","supermassive black holes","X-ray spectroscopy","protoclusters","submillimetre galaxies","HATLAS J084933"],"falsifier":"Measure the broad Balmer decrement by detecting Hβ in deeper rest-frame optical spectroscopy; if the extinction implied by Hα/Hβ is much larger than the X-ray column density of about 5×$10^{21}$ $cm^{-2}$, the AGN is seen through dusty gas and the clear-cavity picture fails, and if broad Hβ is absent entirely, the interpretation of the broad Hα as a Type-1 AGN would need re-examination.","tokens_in":19134,"feed_emoji":"🔭","tokens_out":7063,"duration_ms":70237,"temperature":0.7,"pith_summary":"At redshift 2.41, the hyperluminous infrared galaxy HATLAS J084933-W has long looked like a pure starburst across its ultraviolet-to-radio spectral energy distribution. This paper reports near-infrared spectroscopy and X-ray observations that reveal a broad Hα line and a luminous, lightly absorbed X-ray source, which together show that the galaxy hosts a Type-1 active galactic nucleus (AGN), meaning its broad-line region is directly visible. The authors estimate a black hole mass of about $2\\times10^9$ solar masses and an AGN bolometric luminosity of about $1.6\\times10^{47}$ erg s$^{-1}$, comparable to the starburst's infrared luminosity. The puzzle is that this powerful accretion engine is invisible in the rest-frame ultraviolet, optical, and mid-infrared; the paper outlines three scenarios to explain where the AGN light is emerging.","feed_headline":"Black hole in dusty starburst rivals the starburst's power","feed_subtitle":"X-ray and broad H-alpha reveal a two-billion-solar-mass AGN that is invisible in the galaxy's light.","key_machinery":"The load-bearing diagnostic is the broad Hα emission line, whose width and flux yield the black hole mass through the virial scaling relation, combined with the X-ray power-law spectrum, whose normalisation and slope feed the $\\alpha_{\\rm OX}$ relation (the correlation between rest-frame 250 nm and 2 keV luminosity) and a bolometric correction that turns the 2--10 keV luminosity into a total accretion luminosity. The panchromatic SED comparison against median submillimetre galaxies and two dusty Type-1 quasars provides the contradiction: the inferred AGN power has no obvious rest-frame ultraviolet, optical, or mid-infrared counterpart.","core_discovery":"The paper establishes, from the profile of the Hα emission line and from the X-ray spectrum, that component W of the HATLAS J084933 protocluster contains a broad-line Type-1 AGN. The broad Hα line has a FWHM of roughly 9,700 km s$^{-1}$ and implies $M_{\\rm bh}\\approx 2\\times10^9\\ M_\\odot$; the X-ray spectrum is an absorbed power law with photon index $\\alpha=0.8\\pm0.1$, intrinsic column $N_{\\rm H}\\approx 5\\times10^{21}$ cm$^{-2}$, and a 2--10 keV luminosity $L_X=1.4\\times10^{45}$ erg s$^{-1}$. Using the $\\alpha_{\\rm OX}$ relation and a bolometric correction, the authors derive a bolometric AGN luminosity of $1.6\\times10^{47}$ erg s$^{-1}$ ($\\log L_{\\rm bol}/L_\\odot\\approx13.62$), matching the starburst's $\\log L_{\\rm IR}/L_\\odot=13.52$. Yet the panchromatic SED shows at most a mild blue excess relative to typical submillimetre galaxies and is orders of magnitude fainter than known dusty Type-1 quasars across rest-frame 0.15--30 $\\mu$m. The paper argues that the accretion power must be emerging somewhere that is not obvious, and proposes three scenarios: a central cavity excavated by the AGN giving a clear line of sight; a second, unseen AGN in a companion galaxy; or a supermassive black hole ejected by asymmetric gravitational radiation, with the resulting lack of feedback explaining the extreme starburst.","pith_inferences":["If the $\\alpha_{\\rm OX}$-based bolometric luminosity is too high by even the factor-of-two scatter, the AGN would still be luminous but would no longer rival the starburst, weakening the central tension.","A direct test of the cavity scenario would be spatially resolved spectroscopy of the ionised gas: a northeast ionisation cone and a deficit of molecular gas along the line of sight would support it.","The ejected-AGN scenario predicts that the X-ray source and the broad Hα centroid should be measurably offset from the CO dynamical centre; current data cannot resolve this, but sharper X-ray or near-infrared integral-field observations could.","Because the X-ray emission is unresolved in the existing data, a small contribution from a second, buried AGN in the starburst cannot be excluded; deeper hard X-ray imaging would settle whether the accretion power is really single or dual."],"forward_implications":["If the conclusion holds, AGN are ubiquitous among hyperluminous infrared galaxies, not merely present in X-ray-selected examples.","A roughly $2\\times10^9\\ M_\\odot$ black hole in a galaxy forming stars at thousands of solar masses per year implies that supermassive black hole growth and extreme starburst activity can proceed simultaneously.","The mismatch between inferred AGN power and the SED calls for a revision of how bolometric AGN luminosity is estimated in dust-rich systems, or for a geometric explanation that hides the accretion light.","The newly confirmed fifth protocluster member shows that HATLAS J084933 is a structure spanning roughly 120 kpc, so its extreme luminosity is a group property rather than that of a single monolithic galaxy."],"supporting_citations":[{"why":"Supplies the αOX relation used to translate the 2-10 keV luminosity into a rest-frame 250 nm luminosity.","marker":"Just et al. (2007)"},{"why":"Provides the bolometric correction that converts the rest-frame UV luminosity into a total AGN luminosity.","marker":"Richards et al. (2006)"},{"why":"Gives the Hα-based virial black hole mass scaling used to estimate M_bh ≈ 2×10^9 M_sun.","marker":"Schulze et al. (2018)"},{"why":"Provides the L_Hα-L_X correlation used to estimate extinction toward the broad-line region.","marker":"Panessa et al. (2006)"},{"why":"Establishes the X-ray-to-IR luminosity limit for star formation, showing the observed X-ray emission is too bright to be stellar.","marker":"Alexander et al. (2005a)"},{"why":"Supplies the comparison sample of X-ray-luminous submillimetre galaxies that HATLAS J084933-W outshines.","marker":"Stach et al. (2019)"},{"why":"Previously resolved the system into multiple components and provided the L_IR and the ultraviolet-to-radio SED used here.","marker":"Ivison et al. (2013)"},{"why":"Showed that the far-IR dust is at a single temperature across the disk, used to argue the AGN does not heat the dust.","marker":"Gómez et al. (2018)"}],"fun_headline_variants":["Hyperluminous starburst hides powerful AGN","Invisible AGN rivals starburst in galaxy cluster","Starburst's secret: a 2-billion-solar-mass black hole","AGN power matches starburst, but where is it?","Massive black hole hides in hyperluminous starburst"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The claim that the AGN's accretion power rivals the starburst rests on converting the measured X-ray luminosity to a bolometric luminosity through average quasar SED scaling relations, rather than on direct detection of the AGN's ultraviolet or optical light; the paper itself notes that using the observed optical magnitude would give a lower bolometric luminosity.","fun_headline_variants_meta":{"raw":{"variants":["Hyperluminous starburst hides powerful AGN","Invisible AGN rivals starburst in galaxy cluster","Starburst's secret: a 2-billion-solar-mass black hole","AGN power matches starburst, but where is it?","Massive black hole hides in hyperluminous starburst"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000652,"raw_usage":{"total_tokens":3116,"prompt_tokens":1195,"completion_tokens":1921,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":811,"completion_tokens_details":{"reasoning_tokens":1833}},"tokens_in":811,"tokens_out":1921,"duration_ms":13684,"temperature":1.0,"reasoning_tokens":1833,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:20:46.177451+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the broad Balmer decrement by detecting Hβ in deeper rest-frame optical spectroscopy; if the extinction implied by Hα/Hβ is much larger than the X-ray column density of about 5×$10^{21}$ $cm^{-2}$, the AGN is seen through dusty gas and the clear-cavity picture fails, and if broad Hβ is absent entirely, the interpretation of the broad Hα as a Type-1 AGN would need re-examination.","supporting_citations":[],"review_version":1}