{"id":"6bc2b5cd-1f2e-434b-834a-0d88d3a2edfa","arxiv_id":"2512.14486","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Two z>8 galaxies have bright [OIII] 88 µm oxygen lines, and one of them shows broad wings suggesting a powerful ionized outflow with a mass-loading factor of about 3.","lead":"Using ALMA, astronomers detected the [OIII] 88-micron oxygen line in two galaxies from the first 600 million years of the universe, including the brightest such line known at z>8. One galaxy shows very broad line wings that may be a powerful outflow of ionized gas, and no dust was seen in any of the four targets.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Outflow claim hinges on an unresolved peak-pixel broad component; the fiducial aperture gives only ΔBIC=2.7 and the host is a merger, so the detection is not yet established.","rationale":"The reader's weakest-assumption identification is precisely the load-bearing point: the broad high-velocity wings are only strongly preferred in the peak-pixel spectrum, while the fiducial aperture yields weak evidence; the source is unresolved and the host is a merger/AGN pair. The paper itself acknowledges these caveats in Section 4.3.5, and the dirty-cube check is genuine supporting evidence against an imaging artifact. Still, the outflow interpretation is not uniquely forced. A uv-plane fit would settle whether the broad component survives independently of image-domain extraction choices and would directly test the merger alternative. Because the reader's condition is exactly this concern and the proposed response (higher-resolution follow-up or reanalysis) is appropriate, the CONDITIONAL verdict should stand unchanged. The two [OIII]88 detections and the UHZ1 non-detection are robust and advance the field; the concern is limited to the outflow claim, which is already honestly caveated in the text. No new fatal flaw is identified.","tokens_in":35891,"tokens_out":4643,"duration_ms":421276,"concrete_test":"Fit the calibrated ALMA visibilities directly (uv-plane) with three models: (1) single Gaussian, (2) narrow+broad Gaussian (the outflow model), and (3) two narrow Gaussians with free velocity offsets (the merger model), using the same priors and the same two SPWs as the paper. Compare ΔBIC. If the narrow+broad model does not beat the single Gaussian by ΔBIC>10 in the uv-plane, or if the two-narrow model is within ΔBIC<10 of the narrow+broad model, the claimed outflow is not established by the current data and the verdict should remain conditional pending higher-resolution ALMA imaging.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central novel claim—a powerful ionized outflow in UNCOVER-10646 with Mdot_out=128 Msun/yr and eta=2.9—depends entirely on the existence of a broad [OIII]88 component. The paper's own Appendix D shows that the 'very strong' ΔBIC≈20 is obtained only for the peak-pixel spectrum (Section 4.3.1, Figure 6). The fiducial 0.8″-radius aperture extraction, which Section 2.3 defines as the measurement method, yields ΔBIC=2.7, i.e. only 'positive evidence' on the Raftery scale (Figure 13). The source is not resolved: the Briggs-weighted imfit gives a beam-deconvolved size of (<0.36″)×(<0.22″) (Appendix D). Moreover, UNCOVER-10646 is a likely late-stage merger of two AGN (Section 2.1.1; Section 4.3.5). The paper argues against the merger interpretation because the profile appears symmetric and there is no evidence for a third component, but it cannot exclude the possibility that the wings arise from a superposition of unresolved kinematically distinct components or from correlated noise that is enhanced in the peak-pixel extraction. The authors honestly caveat this in Section 4.3.5, and they check that the wings are present in the dirty cube, which argues against a CLEAN artifact. However, the abstract and summary present the outflow as a discovery ('uniquely reveals'), which goes beyond what the data currently support. If the broad component is not real, the outflow mass, rate, and mass-loading factor are not physically meaningful.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents the first systematic ALMA Band 7 study of [OIII] 88 μm emission and dust continuum in four gravitationally lensed, JWST-selected galaxies at z = 8.5–10.3. [OIII] 88 is robustly detected in UNCOVER-10646 at z = 8.5080 ± 0.0011 (15σ) and in DHZ1 at z = 9.3113 ± 0.0006 (6σ), with intrinsic luminosities L ≈ (1.1–1.6)×10^9 L_sun that place them on the local [OIII]-SFR relation; neither of the two z > 10 targets is detected, yielding deep limits. The central novel claim is that the high-S/N detection in UNCOVER-10646 shows a broad line component (FWHM ≈ 1366 km/s) interpreted as a powerful ionized outflow, with inferred Mdot_out ≈ 128 M_sun/yr and mass-loading factor η ≈ 2.9. The authors extensively caveat that this outflow signature is statistically strong only in a peak-pixel extraction (ΔBIC ≈ 20) and weak in the fiducial 0.8 arcsec aperture (ΔBIC = 2.7), and that the source is an unresolved, likely merging pair of AGN.","tokens_in":36301,"tokens_out":4259,"duration_ms":38554,"significance":"If the outflow interpretation is confirmed, this would be the earliest ionized outflow measured through [OIII] 88, with a mass-loading factor exceeding the galaxy's SFR and plausibly escaping the galaxy. The paper's strengths include robust 15σ and 6σ line detections with verified redshifts, a first systematic multi-target sample at z > 8, careful checks against cleaning artifacts (including a dirty-cube comparison), transparent presentation of the statistical evidence for both extraction choices, and comparison of the derived quantities to established local relations and external simulations. The deep upper limits for the z > 10 targets, especially UHZ1, are also valuable. However, the headline outflow claim rests on a single unresolved peak-pixel feature whose statistical significance drops to marginal in the fiducial aperture, and the source's likely merger nature provides a plausible alternative explanation. The paper is honest about these caveats, but the abstract and summary overstate the robustness of the outflow discovery.","major_comments":[{"comment":"The statistical evidence for the broad component is extraction-dependent. The 'very strong evidence' (ΔBIC ≈ 20) is obtained only for the peak-pixel spectrum (Section 4.3.1), while the fiducial 0.8''-aperture extraction defined in Section 2.3 yields ΔBIC = 2.7, i.e., only 'positive evidence' per the Raftery scale (Appendix D, Fig. 13). The authors acknowledge this in Section 4.3.5, but the abstract and summary present the outflow as 'uniquely revealed.' Because the outflow mass, rate, and mass-loading factor in Section 4.3.2 are derived entirely from the broad component, the central claim should either be reframed as a candidate outflow pending the fiducial-aperture evidence, or the paper should justify why the peak-pixel extraction is the appropriate measurement and quantify the selection effect. As it stands, the discovery claim is not supported by the paper's own fiducial measurement.","section":"Section 4.3.1 / Appendix D, Fig. 13"},{"comment":"UNCOVER-10646 is described as a likely late-stage merger of two AGN, unresolved in the ALMA observations (Appendix D: beam-deconvolved size <0.36'' × <0.22''). A superposition of kinematically distinct components from the merging galaxies can produce broad, symmetric line wings without a single physical outflow. The paper's arguments against this — the apparent symmetry of the profile and the lack of evidence for a third component — reduce the likelihood of a two-component merger signature but do not exclude it. This is load-bearing for the physical interpretation. The abstract's 'uniquely reveals ... an ionized outflow' should be softened to 'candidate ionized outflow' unless the authors present additional kinematic modeling (e.g., a two-component merger model) that can be compared to the observed profile.","section":"Section 2.1.1 / Section 4.3.5"},{"comment":"The outflow mass and rate depend crucially on quantities taken from 'Weaver et al. (in preparation)': SFR = 45.4 ± 1.7 M_sun/yr, stellar mass, metallicity 12+log(O/H) = 8.12, and the AGN identification. If this companion paper is not publicly available at the time of review, the outflow properties are not independently checkable. Please include the relevant values (or an explicit tabulation) in the present paper, or provide a public version of the companion analysis. This is not a fatal flaw, but it is necessary for reproducibility of the paper's quantitative claims.","section":"Section 4.3.2, Eq. (1)"}],"minor_comments":[{"comment":"The quoted significances of 15σ and 6σ refer to the peak pixel S/N in the moment-0 maps. Please state this explicitly in the text and figure captions, as the line-flux S/N may differ and the reader could otherwise misinterpret the detection significance.","section":"Section 3.2"},{"comment":"In the bottom panels, the two fits are shown in separate panels without repeated y-axis labels. Consider aligning the panels or adding common axis labels to facilitate direct visual comparison.","section":"Figure 6"},{"comment":"The statement that all currently detected z > 8 galaxies fall within ~0.3 dex of the local [OIII]-SFR relation would benefit from specifying whether that scatter is the intrinsic scatter of the De Looze et al. (2014) relation or the observed scatter of the plotted sample.","section":"Section 4.2"},{"comment":"The footnote appears to contain a formatting error: '1 A range of 1.2×FWHM ...' should read 'A range of 1.2×FWHM ...'.","section":"Section 2.3, footnote 1"},{"comment":"Several crucial inputs rely on 'Weaver et al. (in preparation)' and other in-press/arXiv-only references. Please update all such references to their published/arXiv versions, or note that the companion values are available in the present paper's appendix.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper fits the journal's scope and the bulk of the observational analysis is careful. The main issue is the framing of the outflow as a discovery when the paper's own fiducial aperture yields only weak statistical evidence (ΔBIC = 2.7) and the source is an unresolved merger candidate. The authors are honest in the caveats section, but the abstract and summary do not reflect that caution. I recommend major revision to either strengthen the outflow claim with additional modeling/justification of the peak-pixel extraction, or reframe the outflow as a candidate requiring confirmation. The line detections and SFR-relation results themselves are solid and could be published after these revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: the two [OIII]88 detections are robust and useful; the outflow in UNCOVER-10646 is plausible but not established, and the abstract oversells it.\n\nThe genuinely new items are the ALMA observations: UNCOVER-10646 at z=8.5080 (15σ) becomes the most luminous [OIII]88 emitter known at z>8, DHZ1 at z=9.3113 (6σ) the most luminous at z>9, and the UHZ1 limit at z=10.07 is one of the deepest at that epoch. The dust continuum non-detections are also useful. The paper is careful with the line measurements — realistic uncertainties, explicit checks that the high-velocity wings appear in the dirty cube, and an honest treatment of the uncertain redshift of UNCOVER-37126. The [OIII]88-SFR comparison uses external relations and makes no obvious circular move.\n\nThe soft spot is the outflow. The double-Gaussian fit is preferred by ΔBIC≈20 only in the peak-pixel spectrum; in the 0.8″ aperture spectrum — the paper's own fiducial extraction — ΔBIC=2.7. The authors argue a compact outflow would be diluted in a larger aperture, which is reasonable, but it leaves the detection dependent on the extraction choice. The source is unresolved and is a likely late-stage merger of two AGN, so unresolved kinematics could mimic a broad component. The case against that is argued, not settled. The outflow mass and mass-loading factor rely on an assumed density, radius, and metallicity from unpublished work, so the quoted numbers should be treated as illustrative. The body of the paper is candid about these issues; the abstract's 'uniquely reveals' is not.\n\nThis is a paper for people working on high-redshift ISM and ALMA follow-up of JWST galaxies. The detections and upper limits stand on their own, and the outflow is a worthwhile candidate for follow-up. I'd send it to peer review — the data are new and the analysis is mostly good — but I'd ask the referee to require that the abstract be aligned with the caveats, and to make clear the outflow is not yet a detection. The detections alone justify publication.\n\nRecommendation: accept after major revision (or at least an abstract fix), not reject.","headline":"Two solid new [OIII]88 detections at z>8, plus a plausible but unproven outflow whose statistical support depends on the extraction.","tokens_in":37038,"tokens_out":4537,"would_cite":true,"duration_ms":41128,"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 first systematic study of [OIII] 88µm at z>8 finds two luminous oxygen lines and a candidate powerful ionized outflow in UNCOVER-10646, with a mass outflow rate that exceeds the galaxy's star formation rate.","keywords":["[OIII] 88 µm","high-redshift galaxies","Epoch of Reionization","ionized outflows","ALMA","star formation","gravitational lensing","mass loading factor"],"falsifier":"A high-resolution ALMA observation that spatially resolves [OIII]88 in UNCOVER-10646 and finds the high-velocity emission either disappears or is not spatially coincident with a compact outflow; or JWST/NIRSpec high-resolution spectroscopy showing no broad component in [OIII]5007; either would overturn the outflow interpretation.","tokens_in":35783,"feed_emoji":"🌌","tokens_out":5737,"duration_ms":45963,"temperature":0.7,"pith_summary":"The paper tries to establish that the far-infrared [OIII] 88µm line can be used systematically to find and characterize galaxies in the first 600 million years of cosmic history, and that at least one such galaxy hosts a powerful ionized outflow. Using ALMA to target four UV-luminous, gravitationally lensed galaxies at z=8.5–10.3, it detects the line in two sources, UNCOVER-10646 at z=8.51 and DHZ1 at z=9.31, and shows that their intrinsic luminosities (roughly 1–2×10^9 Lsun) follow the local [OIII]-SFR relation for dwarf galaxies. The central new claim is that the [OIII]88 line in UNCOVER-10646 is better described by a narrow plus a broad component, with the broad component having FWHM≈1370 km/s and carrying about twice the flux of the narrow one. Interpreted as a compact ionized outflow, this implies a mass outflow rate of about 130 solar masses per year and a mass loading factor of about three, meaning the galaxy expels ionized gas faster than it forms stars, at a speed that likely exceeds its escape velocity. If correct, this is the earliest ionized outflow yet measured through this line, showing that feedback-driven gas ejection was already underway within 600 million years of the Big Bang.","feed_headline":"Earliest ionized outflow detected in the first 600 million years","feed_subtitle":"A z=8.5 galaxy ejects ionized gas at triple its star-formation rate, likely escaping into intergalactic space.","key_machinery":"The central object is the [OIII] 88µm far-infrared fine-structure line, observed with ALMA at ~30–50 km/s spectral resolution; it traces the low-density ionized ISM and is typically brighter than [CII] at z>6. The kinematic argument is carried by a dual-Gaussian decomposition of the line profile with Bayesian model comparison (ΔBIC), applied to a peak-pixel spectrum to maximize signal-to-noise on a compact, unresolved component. Outflow properties are derived from the broad Gaussian's flux, width, and assumed size (0.3″), using the standard ionized-outflow mass and mass-outflow-rate formalism.","core_discovery":"On its own terms, the paper's central discovery is that the [OIII] 88µm line in the z=8.51 galaxy UNCOVER-10646 is not a single Gaussian: its high signal-to-noise peak-pixel spectrum is significantly better fit by a narrow plus a broad component (ΔBIC≈20), with the broad component having FWHM = 1366 km/s. Under the interpretation that this broad component is an ionized outflow, the inferred outflow mass is ~1.4×10^8 M_sun, the mass outflow rate is 128 M_sun/yr, and the mass loading factor η = 2.9, exceeding the galaxy's own star formation rate and exceeding its estimated escape velocity. The paper further establishes that the two detected sources have intrinsic [OIII]88 luminosities of (1.1–","pith_inferences":["If outflows like this are common at z>8, they could explain the puzzling UV brightness of early galaxies by clearing dust on short timescales; the tentative ~2σ extended dust continuum co-spatial with UNCOVER-10646 is a direct, testable symptom of this dust-clearing scenario.","Because the target is a late-stage merger of two AGN, the outflow may be partly AGN-driven; a clean test is to compare the broad-component flux and velocity in [OIII]88 with future high-resolution [OIII]5007 mapping of each merger component.","The large difference in ΔBIC between peak-pixel (20) and aperture (2.7) extractions is a cautionary lesson: for unresolved sources, tests for broad-line outflows should be reported for both extraction methods, since aperture dilution can mask otherwise strong evidence.","If confirmed, this sets a quantitative benchmark at z~8.5 for simulations of feedback-regulated galaxy formation, which currently predict mass loading factors of order unity to a few at these stellar masses."],"forward_implications":["The [OIII]88 line can spectroscopically confirm and kinematically characterize z>8 galaxies efficiently, even in 2–4.5 hours of ALMA time, including for sources at z>9.","Luminous [OIII]88 emitters at z>8 follow the local [OIII]-SFR relation for dwarf galaxies, reinforcing [OIII]88 as a star-formation tracer in the Epoch of Reionization.","Powerful ionized outflows with mass loading factor η≈3 exist in the first 600 million years; if real, the outflow's velocity exceeds the galaxy's escape velocity, so it will enrich the circumgalactic and intergalactic medium.","The deep [OIII]88 upper limit for the z=10.07 X-ray AGN UHZ1 implies the AGN's obscuration is not due to galaxy-wide dust; the line non-detections and stringent dust limits indicate subdued dust production in these early systems.","Higher-resolution ALMA follow-up is explicitly required to confirm and spatially resolve the outflow, and JWST/NIRSpec high-resolution spectroscopy should reveal a corresponding broad component in rest-optical [OIII] lines."],"fun_headline_variants":["Ionized outflow seen in galaxy just 600 million years after Big Bang","Galaxy's early gas outflow triple its star formation rate","First [OIII] 88µm outflow at z>8 revealed","Outflow in early galaxy ejects gas 3× faster than it makes stars","Lensed galaxy at z=8.5 shows powerful ionized outflow"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The broad, high-velocity wings in UNCOVER-10646's [OIII]88 line are a real, compact ionized outflow rather than an artifact of the peak-pixel extraction or the kinematics of the unresolved merger of two AGN; the evidence is strong only in the peak-pixel spectrum (ΔBIC≈20), the source is not spatially resolved, and the 0.8″-aperture spectrum prefers the two-component fit only marginally (ΔBIC≈2.7).","fun_headline_variants_meta":{"raw":{"variants":["Ionized outflow seen in galaxy just 600 million years after Big Bang","Galaxy's early gas outflow triple its star formation rate","First [OIII] 88µm outflow at z>8 revealed","Outflow in early galaxy ejects gas 3× faster than it makes stars","Lensed galaxy at z=8.5 shows powerful ionized outflow"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000301,"raw_usage":{"total_tokens":1736,"prompt_tokens":1072,"completion_tokens":664,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":816,"completion_tokens_details":{"reasoning_tokens":568}},"tokens_in":816,"tokens_out":664,"duration_ms":5982,"temperature":1.0,"reasoning_tokens":568,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T16:01:02.858588+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-resolution ALMA observation that spatially resolves [OIII]88 in UNCOVER-10646 and finds the high-velocity emission either disappears or is not spatially coincident with a compact outflow; or JWST/NIRSpec high-resolution spectroscopy showing no broad component in [OIII]5007; either would overturn the outflow interpretation.","supporting_citations":[],"review_version":1}