{"id":"11c4ac3e-8d11-4451-a73f-32f5782fcc6b","arxiv_id":"1909.01355","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Outflow rates in 12 disk-dominated AGN imply inflow rates near 1 solar mass per year, supporting secular, merger-free fueling of supermassive black holes.","lead":"This paper measures gas outflows in 12 disk-shaped galaxies with active black holes and finds that outflows carry about 18 times more mass than the black holes consume. The result supports the idea that everyday galactic processes, not galaxy collisions, can fuel supermassive black hole growth.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Disk sample is the 12 brightest [OIII] outflows selected from 101 disk AGN; the 4.2σ higher accretion than B17 may be a selection effect, and the paper never tests it against the full parent sample.","rationale":"The paper's central quantitative contribution is the outflow and inflow rates, but the title and the 'dominance' conclusion lean on the comparison with B17 (Section 5.3). The disk sample is explicitly the 12 brightest [OIII] outflows from the parent sample, so the comparison is not between representative secular and merger populations. Because [OIII] luminosity tracks AGN power, the 4.2 sigma accretion excess could be a selection effect. The paper acknowledges sample incompleteness in Section 5.2 but does not correct or test for this bias in the comparison. The reader's weakest assumption, residual [OIII] from star formation, is real and the paper itself calls it the greatest limitation, but it primarily affects the absolute outflow/inflow rates, not the comparative claim that secular systems accrete faster. The outflow-rate deficit is less affected by the brightest-outflow selection because that selection would bias disk rates upward, so the deficit is conservative. The mdot comparison, however, is directly threatened. The concrete test is cheap because the parent-sample mdot values already exist in SSL17. Since the current evidence is suggestive but the selection issue is untested, the CONDITIONAL verdict remains appropriate, hence UNCHANGED.","tokens_in":26895,"tokens_out":11886,"duration_ms":127799,"concrete_test":"Recompute the KS comparison against B17 using the full SSL17 parent sample of 101 disk-dominated AGN, or at least the 58 with spectroscopically detected outflows, instead of the 12 brightest [OIII] sources. Use the same black hole accretion-rate values from SSL17, and if possible match redshift and L_bol ranges to B17. If the 4.2 sigma excess drops below roughly 2 sigma or disappears entirely, the accretion-rate component of the central claim is a selection artifact rather than evidence for secular fueling; if the excess persists, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central 'dominance' claim rests on the comparison with Bae et al. (2017, B17) in Section 5.3, but the disk sample is not representative: Section 2.1 states that the 12 targets were selected as 'the 12 brightest galaxies in the blueshifted [OIII] 5007A spectral component' from the 58 outflows detected in the parent SSL17 sample. Since [OIII] luminosity correlates with AGN bolometric luminosity, this selection plausibly enriches the sample for high-L_bol, high-mdot systems. The 4.2 sigma accretion-rate excess over B17 could therefore reflect selecting the bright tail of the secular population rather than a true secular-vs-merger difference. The bootstrap test in Section 5.3 only propagates mdot uncertainties; it does not address this selection effect. The paper has mdot values for all 101 SSL17 disk-dominated AGN and for the 58 with detected outflows, but never compares these larger samples to B17. The outflow-rate deficit is less vulnerable to this particular selection because picking bright outflows would bias disk rates upward, so the 2.6 sigma deficit may survive or strengthen; it is the accretion-rate comparison that is most at risk. A simple and direct test using the already-available parent-sample mdot values would settle whether the paper's comparative claim is physical or an artifact of sample construction.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents narrow-band [OIII] 5007 Å imaging of 12 disk-dominated (presumed merger-free) AGN selected as the brightest blueshifted [OIII] components among 58 outflow detections in the Simmons, Smethurst and Lintott (2017) sample, plus 7 disk-dominated AGN without detected spectral outflows. The authors measure outflow rates from residual [OIII] flux after continuum and PSF subtraction, derive a mean outflow rate of 0.95 ± 0.14 M⊙ yr−1, compare it with SSL17 black hole accretion rates (mean 0.054 ± 0.039 M⊙ yr−1), and infer a mean lower limit on the inflow rate of ~1.01 ± 0.14 M⊙ yr−1 via Equation (4). They compare the disk sample with Bae et al. (2017) merger-dominated AGN and report significantly higher accretion rates (4.2σ) and lower outflow rates (2.6σ), interpreting this as evidence that secular processes can dominate SMBH growth.","tokens_in":27181,"tokens_out":7191,"duration_ms":66204,"significance":"If the central measurements are correct, the paper provides one of the first direct estimates of the gas supply rates required to sustain AGN activity in disk-dominated systems and a concrete observational contrast between secularly and merger-fed AGN. The new narrow-band imaging data are an independent contribution, and the authors are careful to mark contaminated or non-detected sources as limits in Table 2 and to acknowledge the main systematics (PSF subtraction and star-formation contamination) in Section 5.4. However, the statistical comparison that underpins the 'dominance' claim is weakened by sample selection and by method heterogeneity, and the mean outflow rate is derived without a stated treatment of the upper and lower limits; these issues need to be addressed before the comparative conclusions are robust.","major_comments":[{"comment":"The disk-dom-outflow sample is defined as the 12 brightest blueshifted [OIII] 5007 Å components among the 58 outflow detections in the SSL17 sample (§2.1). Because [OIII] luminosity is a tracer of AGN bolometric luminosity, this selection will plausibly bias the sample toward high Lbol and hence high mdot = Lbol/(ηc²). The 4.2σ difference in mdot between this sample and B17 (§5.3, Fig. 10) may therefore be a selection effect rather than a physical difference between secularly and merger-fed systems. The bootstrap test in the §5.3 footnote propagates only the mdot measurement uncertainties; it does not address this selection. Since the mdot values for all 101 SSL17 galaxies and for the 58 with detected outflows are already available, I request a direct comparison of these parent-sample distributions with B17; if the parent sample shows the same elevation, the claim is robust, and if not, the central comparative conclusion (item iv in Section 6) is unsupported.","section":"§2.1, §5.3"},{"comment":"The mean outflow rate of 0.95 ± 0.14 M⊙ yr−1 is presented without stating how the five non-detections and limits in Table 2 are handled: Neville, Hermione and Cho are lower limits, while Snape and Crabbe are upper limits. Treating these bracketing values as point measurements biases both the mean and its standard error in a sample of only 12 objects. This is load-bearing because the factor-of-18 comparison with mdot and the inflow-rate estimate (Eq. 4) are derived from this mean. Please repeat the analysis using a survival-analysis estimator or a sensitivity test in which limits are set to their extreme values, and report the median and its uncertainty as well.","section":"§4, Table 2"},{"comment":"For all sources except Hermione, Neville and Cho, the flux remaining after continuum and PSF subtraction is assumed to be entirely outflow-ionized gas. The authors themselves identify star-formation contamination as the 'greatest limitation' (Section 5.4). Because Mgas and Mdot_outflow scale linearly with L[OIII] (Eq. 1), unrecognized star-formation ionization would directly inflate the mean outflow rate and the inferred inflow rate. I ask for a quantitative upper bound on this contamination, for example using the gas masses measured in the disk-dom-none sample (Section 4) as templates, and a statement of how the factor-of-18 result changes under a conservative subtraction of this contamination.","section":"§3.1, §5.4"},{"comment":"The B17 comparison compounds several methodological differences: Type 2 vs Type 1 AGN, M-sigma vs virial H-alpha black hole masses, [OIII]-based vs WISE-based bolometric luminosities, IFS vs narrow-band imaging, and B17's lack of quoted uncertainties. The two-sample KS tests in Fig. 10 treat each sample's scalar values as exact and directly comparable, so the reported p-values (0.00003 and 0.009) do not include systematic uncertainties. At minimum, the authors should quantify how each known systematic (e.g., the η correction already discussed, and the different Lbol calibrations) shifts the distributions, and ideally restrict the comparison to a matched subset in Lbol or MBH before concluding that the mdot and outflow-rate differences are driven by merger vs secular history.","section":"§5.3, Fig. 10"}],"minor_comments":[{"comment":"There is an unmatched closing parenthesis in 'SMBHs $0.054\\pm0.039~\\rm{M}_{\\odot}~\\rm{yr}^{-1}$)'.","section":"Abstract"},{"comment":"The velocity units in Equation (2) are written as 'km yr−1'; the conventional unit for outflow velocities is km s−1, and the conversion from rmax/v to years should be stated explicitly.","section":"§3.2, Eq. (2)"},{"comment":"The value of the clumping factor C is not specified in the text; since Mgas is proportional to C, please state explicitly that C = 1 (or the adopted value) and indicate how its uncertainty enters the error budget.","section":"§3.2, Eq. (1)"},{"comment":"The notation for limits such as '> 0.07 ± 0.22' and '< 0.007 ± 0.004' is ambiguous; please clarify whether the quoted uncertainty refers to the limit value itself and whether these uncertainties were used in computing the sample mean.","section":"Table 2"},{"comment":"The text states that no cuts were made on signal-to-noise ratio, but the selection of the '12 brightest' blueshifted components is effectively a flux selection; a sentence clarifying the distinction would avoid confusion.","section":"§2.1"},{"comment":"The word 'inlcuding' in the first section is a typo for 'including'.","section":"§1"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the paper is within the scope of the journal and presents useful new narrow-band imaging data. My principal concern is the selection effect in §2.1 as it bears on the headline 4.2σ accretion-rate comparison; because the parent-sample mdot values are already in hand, a fix is straightforward and should be requested. If the authors cannot provide a parent-sample comparison, I would consider the comparative claim unsubstantiated. The mixed limit treatment in §4 is a secondary but important issue that also needs to be resolved before the quantitative inflow-rate result is reliable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a useful observational paper with a central comparison that is not yet trustworthy. The new data are real: narrowband [OIII] imaging of 12 disk-dominated AGN, with outflow rates from 0.007 to 2.92 Msun/yr, and an inference that these systems need about 1 Msun/yr of inflow. That is worth having.\n\nWhat the paper does well: it is transparent about its own limitations. It flags PSF over-subtraction, star formation contamination, and it assigns upper or lower limits to five of the twelve sources rather than hiding them. It also does not overclaim the bar result, noting the bar excess is only 2.1 sigma. The interpretation in terms of spin and geometry is clearly labeled as speculative and follows Nayakshin et al. (2012), which is fine. The inflow rate is simply outflow plus accretion, so no circularity there.\n\nThe soft spots are real but not all equal. The biggest issue is sample selection. The 12 targets are the 12 brightest [OIII] outflows from the 58 detected in the parent sample of 101. Since [OIII] luminosity tracks bolometric luminosity, this selects for high accretion rate. The paper then compares those 12 to Bae et al.'s merger sample and finds a 4.2 sigma excess in accretion rate. That excess could just be the bright tail of the secular population, not a secular-vs-merger difference. The authors have the mdot values for all 101 and all 58; running the same KS test against B17 with those larger samples would settle it. They don't. That is a fixable but important gap.\n\nThe outflow-rate deficit at 2.6 sigma is less vulnerable to this selection, because picking bright outflows would bias the disk rates upward, so the deficit might actually be larger in an unbiased sample. But 2.6 sigma is not a strong claim.\n\nThe other issues are smaller. The mean outflow rate treats upper and lower limits as point values; that should be redone with a survival analysis or at least stated clearly. The B17 comparison uses different methods (IFS vs narrowband imaging, Type 2 vs Type 1) and the paper acknowledges but doesn't quantify the impact. The systematic uncertainties in ne, C, f, and metallicity are not propagated, but those affect all such measurements and are not a reason to reject.\n\nBottom line: the paper deserves a serious referee. The outflow measurements are new and the inflow rate estimate is a useful benchmark. But the headline claim about the dominance of secular growth needs the parent-sample comparison and a more careful treatment of limits before I'd believe it. I'd send it to review, and the review should push on those two points.","headline":"New outflow measurements for 12 disk-dominated AGN, but the secular-vs-merger claim is undermined by a selection effect the authors never test.","tokens_in":27691,"tokens_out":5669,"would_cite":false,"duration_ms":45643,"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":"This paper shows that disk galaxies with no merger history can fuel their supermassive black holes entirely through secular processes, with measured outflows implying inflow rates of about one solar mass per year.","keywords":["AGN outflows","supermassive black hole growth","secular evolution","disk galaxies","narrowband imaging","[O III] emission","black hole accretion","galaxy mergers"],"falsifier":"Take the same 12 galaxies with an integral-field spectrograph and map the kinematics of the residual [O III] emission: if most of the flux outside the AGN PSF has narrow line widths and [O III]/H$\\beta$ ratios typical of star formation rather than a broad, blueshifted AGN-driven component, then the measured outflow masses and derived inflow rates would not stand.","tokens_in":26696,"feed_emoji":"🌌","tokens_out":7863,"duration_ms":70277,"temperature":0.7,"pith_summary":"The paper sets out to test whether supermassive black holes in galaxies that have never undergone a major merger can still grow at the rates observed, fuelled entirely by internal processes such as bars and spiral arms. Using narrowband imaging of the [O III] 5007 Å line in 12 disk-dominated AGN, the authors detect outflows in 10 of them and measure a mean outflow rate of $0.95\\pm0.14\\,M_\\odot\\,\\mathrm{yr}^{-1}$, about 18 times the mean black hole accretion rate of $0.054\\pm0.039\\,M_\\odot\\,\\mathrm{yr}^{-1}$. From energy conservation this implies an average inflow rate of roughly $1.01\\pm0.14\\,M_\\odot\\,\\mathrm{yr}^{-1}$, a level that simulations show bars, spiral arms, and cold accretion can supply. The paper also reports that these secularly fuelled AGN accrete about five times faster, yet drive outflows about five times weaker, than a comparison sample of AGN with merger-dominated histories, attributing the difference to smoother planar inflow spinning up the black hole and a biconical outflow geometry.","feed_headline":"Merger-free disk galaxies feed black holes at 1 solar mass per year","feed_subtitle":"Outflows outpace black hole accretion 18-to-1, so secular inflows alone can fuel SMBH growth.","key_machinery":"The machinery is narrowband [O III] $\\lambda5007$ imaging: continuum subtraction removes starlight, PSF subtraction removes the AGN's own narrow-line emission, and the remaining flux inside the Petrosian radius is summed as the outflow. The [O III] luminosity is converted to an outflowing gas mass via the Carniani et al. (2015) scaling, assuming solar metallicity and $n_e=500\\,\\mathrm{cm}^{-3}$, and dividing by the outflow timescale, set by the blueshifted velocity and the maximum spatial extent of the emission, gives the mass outflow rate. Adding the black hole accretion rate, taken from SSL17 as $\\dot m=L_{\\rm bol}/\\eta c^2$ with $\\eta=0.15$, yields the inflow rate, with an energy-coupling factor $f=1$ so the quoted inflow rates are upper limits. The comparison claim rests on two-sample Kolmogorov–Smirnov tests between this sample and the Bae et al. (2017) merger-dominated sample.","core_discovery":"The central claim is that merger-free, disk-dominated hosts can supply all the gas needed to power both an AGN and its outflow: the combined inflow rate needed is about $1\\,M_\\odot\\,\\mathrm{yr}^{-1}$, within the range produced by bars, spiral arms, and cold accretion in simulations. The mean outflow rate in the 12-source disk-dominated sample, $0.95\\pm0.14\\,M_\\odot\\,\\mathrm{yr}^{-1}$, exceeds the mean SMBH accretion rate, $0.054\\pm0.039\\,M_\\odot\\,\\mathrm{yr}^{-1}$, by a factor of roughly 18, so the galaxy must feed its centre at least this fast. In a direct comparison, the disk-dominated sample shows about 5 times higher black hole accretion rates ($4.2\\sigma$) and about 5 times lower outflow rates ($2.6\\sigma$) than a sample of 20 AGN with merger-dominated histories. The authors interpret this as evidence that smooth, coplanar secular inflow both spins the black hole up, raising accretion efficiency, and suffers less feedback interception from a biconical outflow than does the chaotic quasi-spherical inflow in merger-fuelled systems.","pith_inferences":["Inference: because the paper assumes $f=1$ in the energy-balance inflow equation, the quoted inflow rates are upper limits; directly measuring the unseen driving wind velocity with X-ray or UV absorption lines would tighten the true rate that secular mechanisms must supply.","Inference: the comparison with Bae et al. (2017) mixes narrowband imaging with integral-field spectroscopy and uses different black-hole mass and bolometric calibrations, so part of the 5x/5x accretion/outflow contrast could be methodological; a same-method comparison would be needed to confirm the spin and geometry story.","Inference: if planar inflow indeed spins black holes up, then within disk-dominated samples the highest-accretion objects should show the highest spin indicators, while merger-fuelled AGN should be preferentially low-spin; X-ray reflection spectroscopy could test this directly.","Inference: the marginal 2.1$\\sigma$ bar excess suggests that a larger sample of disk-dominated AGN, controlling for stellar mass, colour, and environment, could turn the bar-outflow connection into a definitive result."],"forward_implications":["Secular processes alone can fuel SMBH growth: the required inflow of roughly $1\\,M_\\odot\\,\\mathrm{yr}^{-1}$ is within the range of bars, spiral arms, and cold accretion and can last longer than the longest outflow timescale of about 920 Myr.","The outflow rate exceeds the accretion rate by a factor of about 18, so most inflowing gas is blown back out rather than consumed; galaxy gas budgets must therefore account for ejected material, not just accreted mass.","The higher accretion rates and lower outflow rates of the disk-dominated sample imply that black hole spin and accretion geometry, not just gas supply, determine how much inflowing gas is consumed versus ejected.","Barred galaxies in the sample host the brightest outflows at a marginally significant excess over the parent sample, suggesting bars are a plausible but not yet confirmed fuel-delivery mechanism.","The observed outflow velocities lie below the galaxy escape velocity, so these outflows redistribute gas within the galaxy rather than ejecting it into the intergalactic medium."],"supporting_citations":[{"why":"Supplies the parent sample of 101 disk-dominated AGN, along with the black hole masses, Eddington ratios, and black hole accretion rates used throughout.","marker":"SSL17"},{"why":"Provides the 20 merger-dominated AGN used as the comparison sample in the 5x/5x accretion and outflow contrast.","marker":"B17"},{"why":"Provides the scaling relation that converts [O III] luminosity into outflowing gas mass.","marker":"Carniani et al. (2015)"},{"why":"Gives the broad-H$\\alpha$ virial recipe used by SSL17 to compute black hole masses.","marker":"Greene & Ho (2005)"},{"why":"Provides GANDALF, the spectral fitting code used to detect the blueshifted [O III] outflow components in the SDSS spectra.","marker":"Sarzi et al. (2006)"},{"why":"Supplies the planar-inflow/biconical-outflow versus quasi-spherical-inflow/outflow model used to explain differences from merger-fuelled AGN.","marker":"Nayakshin et al. (2012)"},{"why":"Simulation result that 65% of SMBH growth since z~3 is secular, the context this work supports.","marker":"Martin et al. (2018)"},{"why":"Provides the bolometric correction applied to WISE 12 $\\mu$m luminosities from which accretion rates are derived.","marker":"Richards et al. (2006)"}],"fun_headline_variants":["Disk galaxies feed black holes without mergers","Outflows outpace black hole accretion 18-to-1","Secular inflows suffice to fuel AGN outflows","Merger-free disk AGN grow black holes fivefold faster"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The paper's numbers rely on the assumption that the [O III] flux left after subtracting the continuum and the AGN's point-spread function comes only from outflowing gas; if star-forming regions also contribute ionized flux, the outflow gas masses, outflow rates, and inferred inflow rates would be overestimated.","fun_headline_variants_meta":{"raw":{"variants":["Disk galaxies feed black holes without mergers","Outflows outpace black hole accretion 18-to-1","Secular inflows suffice to fuel AGN outflows","Merger-free disk AGN grow black holes fivefold faster"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000574,"raw_usage":{"total_tokens":2814,"prompt_tokens":1153,"completion_tokens":1661,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":769,"completion_tokens_details":{"reasoning_tokens":1597}},"tokens_in":769,"tokens_out":1661,"duration_ms":12518,"temperature":1.0,"reasoning_tokens":1597,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:20:22.918022+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the same 12 galaxies with an integral-field spectrograph and map the kinematics of the residual [O III] emission: if most of the flux outside the AGN PSF has narrow line widths and [O III]/H$\\beta$ ratios typical of star formation rather than a broad, blueshifted AGN-driven component, then the measured outflow masses and derived inflow rates would not stand.","supporting_citations":[{"cited_title":"E., Ho L","cited_arxiv_id":null,"evidence_quote":"Gives the broad-H$\\alpha$ virial recipe used by SSL17 to compute black hole masses."},{"cited_title":"L., Bacon R., Bureau M., Cappellari M., de Zeeuw P","cited_arxiv_id":null,"evidence_quote":"Provides GANDALF, the spectral fitting code used to detect the blueshifted [O III] outflow components in the SDSS spectra."},{"cited_title":"R., 2012, , 753, 15","cited_arxiv_id":null,"evidence_quote":"Supplies the planar-inflow/biconical-outflow versus quasi-spherical-inflow/outflow model used to explain differences from merger-fuelled AGN."},{"cited_title":"D., Devriendt J","cited_arxiv_id":null,"evidence_quote":"Simulation result that 65% of SMBH growth since z~3 is secular, the context this work supports."},{"cited_title":"T., Lacy M., Storrie-Lombardi L","cited_arxiv_id":null,"evidence_quote":"Provides the bolometric correction applied to WISE 12 $\\mu$m luminosities from which accretion rates are derived."}],"review_version":1}