{"id":"890e0506-ca79-4f7f-8714-fa173d684503","arxiv_id":"2501.12444","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"AGN optical variability timescale and amplitude measured from ASAS-SN light curves define a plane that predicts supermassive black hole mass with about 0.4 dex scatter.","lead":"Astronomers combined a decade of all-sky brightness monitoring of 57 active galaxies with known black hole masses and found that the timescale and amplitude of the variability can predict black hole mass. This could let future surveys estimate masses for thousands of black holes using only photometry, without expensive spectroscopy.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"External validation shows a -0.53 dex offset and 0.70 dex dispersion, so the plane is not yet a reliable mass estimator for field AGN without recalibration.","rationale":"The reader's weakest assumption (Section 2.4 error correction) is a plausible root cause of the problem, but the most load-bearing concern is the direct contradiction between the claimed reliability and the external validation: a -0.53 dex offset and 0.70 dex dispersion against BASS. The error correction could cause this offset if it does not transfer from galaxies to AGN, but the offset is also consistent with selection differences or BASS systematics. Either way, the paper's own data show that Equation 2 does not provide accurate masses for field AGN in its present form. This is the weak point where the central claim is least secure. Because the correlation between variability and mass likely exists and the offset may be correctable, a conditional verdict remains appropriate; the concern does not move the reader's verdict but sharpens its basis. The concrete test would settle whether the offset is intrinsic to the plane or an artifact of the comparison sample.","tokens_in":18965,"tokens_out":5083,"duration_ms":51617,"concrete_test":"Re-fit the plane using the 42 BASS-overlap objects as the calibration truth (same DRW parameters and BASS masses) and compare the resulting coefficients and scatter to Equation 2. Also compute the offset and dispersion for the full 42 objects without 2-sigma clipping, and separately for H-beta and M-sigma BASS masses. If the re-fit coefficients differ from Equation 2 by more than the quoted uncertainties, or if the -0.53 dex offset persists in both mass-measurement subsets, the plane needs external recalibration before it can be considered reliable.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that Equation 2 provides reliable SMBH mass estimates—is contradicted by the paper's own external validation in Section 4. Against 42 BASS masses, the ASAS-SN masses have a systematic offset of -0.53 dex and a dispersion of 0.70 dex after 2-sigma clipping (Figure 7, Table 4). The offset means field masses are biased low by a factor of ~3.4, and the dispersion is nearly twice the in-sample 0.39 dex scatter. The paper reports these statistics but does not correct or explain the offset, and the abstract and discussion still claim reliability. The offset could arise from the Section 2.4 error-correction procedure—derived from constant galaxies but applied to variable AGN—from differences between the calibration sample (nearby Seyferts, z<0.2) and the field sample (quasars up to z~3.9), or from systematics in the BASS single-epoch virial and M-sigma masses. The 2-sigma clipping removes 10% of the validation points, and the additional cuts on tau_DRW and sigma_hat further restrict the field sample, so the validation is not robust. Without understanding the offset, Equation 2 cannot be used as a precision mass estimator for the broad range of masses and redshifts claimed.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper calibrates a scaling relation between optical variability properties of AGN and SMBH mass. The authors fit damped random walk models to ~11-year ASAS-SN light curves for 57 AGN with reverberation-mapping or dynamical mass measurements, and derive Equation (2): a plane relating log M_BH to log τ_DRW and log σ̂, with an in-sample intrinsic scatter of 0.39 dex. They apply this relation to 203 bright Milliquas AGN and compare the resulting masses with BASS masses for 42 overlapping objects, finding a systematic offset of -0.53 dex and a dispersion of 0.70 dex. The paper also forecasts the reach of LSST and ASAS-SN for future variability-based mass measurements.","tokens_in":19193,"tokens_out":5448,"duration_ms":59407,"significance":"If the calibration were validated, this would be a valuable method for estimating SMBH masses from photometry alone, using a homogeneous, long-baseline survey dataset without the need for costly spectroscopy. The paper's strengths include the use of a homogeneous ASAS-SN dataset, careful attention to under-reported photometric uncertainties, and an explicit external-validation comparison that most similar studies omit. However, the external validation shows a large unexplained systematic offset and a dispersion nearly twice the claimed in-sample scatter, so the central claim that Equation (2) provides reliable mass estimates is not yet established. The paper is a useful calibration study, but its conclusions and abstract need to be reconciled with the validation statistics.","major_comments":[{"comment":"The external validation against BASS is the decisive test of the paper's central claim, and it currently fails: the ASAS-SN masses are offset by -0.53 dex and have a 1σ dispersion of 0.70 dex after 2σ clipping that removes 10% of the sample. These numbers are reported but not explained, and the abstract and Section 5 nevertheless state that the relation provides 'reliable M_BH estimates' with a 'typical scatter of 0.39 dex'. The in-sample scatter is not the predictive accuracy of the relation when applied to field AGN. At minimum, the paper must discuss the offset, test whether it can be traced to the BASS single-epoch virial/M-σ mass scale rather than to the variability-based masses, and either recalibrate the zero point (e.g., by including a BASS-based cross-calibration term) or carefully restrict the claims to the calibration locus.","section":"Section 4, Figure 7, Table 4"},{"comment":"The error-correction procedure is load-bearing because τ_DRW and σ̂ are derived from the flux uncertainties, and the authors themselves note that underestimated uncertainties bias τ_DRW low. The correction is calibrated on 50,000+ random galaxies, most of which are quiescent, and is then applied to AGN light curves, including strongly variable point sources. The manuscript does not demonstrate that galaxy-based corrections transfer to AGN, nor does it test the sensitivity of Equation (2) to the details of this correction (e.g., by refitting without the correction or by using the independent variable-star correction of Jayasinghe et al. 2018). A quantitative robustness test is needed to show that the plane parameters are not driven by this assumption.","section":"Section 2.4"},{"comment":"The path from the initial 17,000 Milliquas sources above the variability threshold to the final 203 objects involves several cuts: the stellar-contamination cut to 1,200 sources, the removal of ~20% of sources with σ̂→0, and the restriction to 10 days < τ_DRW < 10^3.5 days. These selection effects are not modeled in the BASS comparison, and the 42 overlapping AGN may not be representative of the full field sample. In addition, the 2σ clipping removes 10% of the validation points before the offset and dispersion are computed, so the reported statistics depend on the clipping procedure. The authors should report the BASS comparison without clipping, with alternative clipping thresholds, and ideally with selection weights, to show that the validation result is robust.","section":"Section 4 selection and clipping"},{"comment":"The paper reports the in-sample intrinsic scatter of 0.39 dex as 'the typical scatter' of the relation and compares it to other scaling relations, but this is an in-sample fit statistic, not a prediction error. The external validation yields a dispersion of 0.70 dex, which is the quantity that matters for the claimed application to field AGN. The manuscript should distinguish clearly between the two, report the external predictive scatter alongside the in-sample value in the abstract and discussion, and avoid implying that 0.39 dex is the expected accuracy of masses estimated with Equation (2) for arbitrary field objects.","section":"Section 3 and Section 4"}],"minor_comments":[{"comment":"The text states that NGC 4151 is the only source with σ̂ > 40 mJy/days^1/2, but the preceding sentence says most sources have σ̂ between ~0.20 and 3.5 mJy/days^1/2; the value 40 appears to be a typo (possibly 0.40 or 4.0) and should be corrected.","section":"Section 3"},{"comment":"The abstract should mention the external validation offset and dispersion, or at least qualify the word 'reliable', so that readers are not misled by the in-sample scatter alone.","section":"Abstract"},{"comment":"The reported uncertainties on the two slopes and the intercept do not include the covariance between the fitted parameters; because τ_DRW and σ̂ are likely correlated, the authors should provide the covariance matrix or bootstrap uncertainties so that mass errors can be propagated correctly.","section":"Equation (2)"},{"comment":"The authors note in Section 5.1 that Equation (2) is assumed not to evolve with redshift, but this assumption is used to extrapolate to z~3.9 in the field sample and to forecast LSST/ASAS-SN capabilities. This important caveat should be stated earlier, in Section 4, where masses at high redshift are first presented.","section":"Section 5.1"},{"comment":"The choice of the 85th percentile of galaxy scatter as the variability threshold is arbitrary; at least a brief justification or a sensitivity check would help the reader assess how this selection affects the calibration sample.","section":"Section 2.3"}],"recommendation":"major_revision","confidential_remarks":"The core calibration is a useful contribution and the authors are transparent in reporting the BASS comparison. However, the offset and dispersion in Table 4 contradict the abstract's claim of reliability. In my view this is fixable in a revision: the authors can either recalibrate the zero point, restrict the claims, or demonstrate that the offset is entirely due to the BASS mass scale. If the offset is left unexplained, the paper should not be accepted as is. I would also encourage the editor to ask for the covariance of Equation (2) and the unclipped validation statistics."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague, here is my take on arXiv:2501.12444. The paper does something useful: it fits the tau_DRW–MBH relation on a homogeneous sample of 57 AGN with RM/dynamical masses using 11-year ASAS-SN light curves, and it shows that adding the variability amplitude sigma_hat reduces the scatter from 0.44 to 0.39 dex. The data handling is careful, the methods are reproducible from public data, and the ancillary tables are there. That is a legitimate extension of the known tau–mass correlation, not just a rehash.\n\nThe soft spot is exactly where the stress-test points: the external validation against BASS. For 42 overlapping field AGN, the ASAS-SN masses come out 0.53 dex low, with a dispersion of 0.70 dex, about twice the in-sample 0.39 dex. The 2-sigma clipping removes 10% of the validation points, and the field sample is further trimmed by cuts on tau and sigma_hat. The authors report these numbers honestly, but they do not explain the offset, and the abstract and discussion still say the method gives reliable masses. That is an overclaim. It is possible that part of the offset comes from BASS single-epoch virial and M-sigma systematics, but the paper does not test that. It is also possible the Section 2.4 error correction, derived from constant galaxies and applied to AGN, shifts the plane.\n\nThe core correlation is almost certainly real, given prior work, and the in-sample fit is clean. So this is not a reject. But the precision claim needs external recalibration or a convincing explanation of the offset before the relation is used as a mass estimator.\n\nAudience: AGN variability specialists and anyone wanting cheap photometric mass estimates from LSST or ASAS-SN. It is a reasonable first step, and the survey projections are useful. I would bring it to reading group and would cite it if working on variability–mass relations, but I would not yet use the plane for science. It deserves a serious referee; the right outcome is major revision, not desk rejection.","headline":"Good homogeneous-data calibration and a real improvement from adding sigma_hat, but the external validation offset means the paper overclaims precision as a mass estimator.","tokens_in":19854,"tokens_out":2814,"would_cite":true,"duration_ms":29949,"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":"The paper claims that a two-parameter plane of AGN optical variability predicts supermassive black hole masses with 0.39 dex scatter.","keywords":["AGN","supermassive black hole masses","damped random walk","optical variability","reverberation mapping","ASAS-SN","photometric mass estimation"],"falsifier":"Measure $\\tau_{\\rm DRW}$ and $\\hat{\\sigma}$ for AGN with independent dynamical masses, for example megamaser disks that were not in the calibration sample, and check whether Equation 2 recovers their masses within the quoted 0.39 dex scatter; alternatively, repeat the DRW fits on light curves from a survey with independent error calibration, such as ZTF or LSST, and test whether the inferred masses agree with the ASAS-SN estimates.","tokens_in":18667,"feed_emoji":"🔭","tokens_out":7427,"duration_ms":64189,"temperature":0.7,"pith_summary":"This paper tries to establish that the stochastic optical flickering of an active galactic nucleus (AGN) encodes the mass of its central supermassive black hole. Using 11 years of homogeneous, near-daily ASAS-SN light curves for 57 AGN with masses measured by reverberation mapping or stellar/gas dynamics, the authors fit a damped random walk model and show that the characteristic timescale $\\tau_{\\rm DRW}$ alone correlates with mass, and that adding the variability amplitude $\\hat{\\sigma}$ tightens the relation to an intrinsic scatter of 0.39 dex. The resulting plane, Equation 2, turns photometry alone into a black hole mass estimate, which matters because reverberation mapping and dynamical measurements are expensive and limited to nearby, bright objects. The paper applies the plane to 203 bright field AGN and compares 42 overlapping sources with spectroscopic masses from the BAT AGN Spectroscopic Survey.","feed_headline":"AGN flicker rates reveal black hole masses to 0.39 dex","feed_subtitle":"Photometric variability alone rivals reverberation mapping as a mass estimator for active galaxies.","key_machinery":"The load-bearing object is the damped random walk (DRW) model of AGN variability, a CARMA(1,0) stochastic process characterized by a damping timescale $\\tau_{\\rm DRW}$ and an amplitude $\\sigma$. The authors fit this model to each light curve by maximum likelihood on a grid in $\\log_{10}\\tau_{\\rm DRW}$ and $\\log_{10}\\hat{\\sigma}$, using $\\hat{\\sigma}^2 = 2\\sigma^2/\\tau_{\\rm DRW}$ because that combination decorrelates the two fitted parameters. Because underestimated photometric errors bias $\\tau_{\\rm DRW}$ low, they compute per-camera, per-filter error corrections from tens of thousands of non-variable galaxy light curves and add those corrections in quadrature before fitting. The homogeneous 11-year baseline and daily cadence of ASAS-SN are what make the $\\tau_{\\rm DRW}$ estimates reliable across the mass range.","core_discovery":"The central discovery is the calibrated variability\\,--\\,mass plane: $\\log_{10}(M_{\\rm BH}/M_\\odot) = (2.27\\pm0.20)\\log_{10}(\\tau_{\\rm DRW}/200\\, {\\rm days}) + (1.20\\pm0.20)\\log_{10}(\\hat{\\sigma}/1\\, {\\rm mJy\\, days^{-1/2}}) + 7.68\\pm0.08$, with an intrinsic scatter of 0.39 dex. The $\\tau_{\\rm DRW}$-only relation already shows a significant correlation with mass (Kendall's tau of 0.50, p-value $5.0\\times10^{-8}$) at 0.44 dex scatter, but the residuals correlate with $\\hat{\\sigma}$, so incorporating the amplitude improves the fit. The final plane shows no statistically significant residual trends with redshift or Eddington\\-ratio proxy. Compared with earlier work, the homogeneous ASAS-SN light curves and the error correction yield typically longer $\\tau_{\\rm DRW}$ values, and the plane gives the smallest dispersion (0.70 dex) when tested against the BASS validation masses.","pith_inferences":["If the plane is as stable as claimed, short-cadence surveys such as TESS or future LSST data could pre-select low-mass AGN candidates for follow-up reverberation mapping, effectively turning the mass estimator into a discovery tool.","The absence of a redshift trend in the calibration sample does not guarantee the plane is non-evolving; a natural extension, not pursued in the paper, would be to fit Equation 2 in redshift bins once LSST provides high-redshift AGN with independent mass anchors.","Because the error correction is the load-bearing step, a clean test of the paper's systematics would be to fit DRW parameters to the same AGN using a completely independent light-curve pipeline, such as forced photometry rather than image subtraction, and compare the resulting masses."],"forward_implications":["Equation 2 turns existing all-sky photometry into black hole masses for thousands of AGN; the paper reports 203 such estimates, including 60 low-mass AGN below $2\\times10^6\\,M_\\odot$.","With a 25-year ASAS-SN baseline, damping timescales up to roughly 2.5 years become recoverable, corresponding to masses near $10^{9.9}\\,M_\\odot$, and a higher-cadence sub-survey could reach down to $10^5\\,M_\\odot$.","The 10-year LSST survey should be able to measure the plane for $7 \\lesssim \\log_{10}(M_{\\rm BH}/M_\\odot) \\lesssim 9$ out to $z\\sim1$ and for $\\log_{10}(M_{\\rm BH}/M_\\odot)\\sim8$ out to $z\\sim4$.","The 0.39 dex scatter is comparable to or better than other black hole scaling relations, so photometric variability offers a competitive mass estimator where spectroscopic or dynamical data are unavailable."],"supporting_citations":[{"why":"Supplies the AGN Black Hole Mass Database with the reverberation-mapping and dynamical masses used to calibrate the plane.","marker":"Bentz & Katz (2015)"},{"why":"Defines the damped random walk parametrization and the $\\hat{\\sigma}^2 = 2\\sigma^2/\\tau_{\\rm DRW}$ variable used in the fit.","marker":"Kozłowski et al. (2010)"},{"why":"Established the $\\tau_{\\rm DRW}$\\,--\\,mass correlation that the paper confirms and extends.","marker":"Kelly et al. (2009)"},{"why":"Provides the DRW fitting framework and the wavelength dependence of $\\tau_{\\rm DRW}$ used to justify rest-frame corrections.","marker":"MacLeod et al. (2010)"},{"why":"Documents the baseline bias that underestimates $\\tau_{\\rm DRW}$, motivating the 11-year homogeneous ASAS-SN light curves.","marker":"Suberlak et al. (2021)"},{"why":"Describes ASAS-SN Sky Patrol V2.0, the source of every light curve used for calibration and field-sample masses.","marker":"Hart et al. (2023)"},{"why":"Provides the LtsFit maximum-likelihood fitter used to derive Equations 1 and 2 with intrinsic scatter.","marker":"Cappellari et al. (2013)"},{"why":"Supplies the BASS spectroscopic comparison masses used to validate the field-sample estimates.","marker":"Koss et al. (2022a)"}],"fun_headline_variants":["AGN flicker plane nails black hole masses to 0.39 dex","Flickering AGN yield SMBH masses to 0.39 dex scatter","AGN variability plane gives BH mass with 0.39 dex scatter","AGN flicker reveals SMBH mass: 0.39 dex intrinsic scatter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The correction for underestimated ASAS-SN uncertainties, derived from quiescent galaxy light curves, is assumed to transfer to point-source AGN light curves; if that transfer is wrong, $\\tau_{\\rm DRW}$ and $\\hat{\\sigma}$ are systematically biased and the whole plane shifts.","fun_headline_variants_meta":{"raw":{"variants":["AGN flicker plane nails black hole masses to 0.39 dex","Flickering AGN yield SMBH masses to 0.39 dex scatter","AGN variability plane gives BH mass with 0.39 dex scatter","AGN flicker reveals SMBH mass: 0.39 dex intrinsic scatter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000758,"raw_usage":{"total_tokens":3561,"prompt_tokens":1334,"completion_tokens":2227,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":950,"completion_tokens_details":{"reasoning_tokens":2141}},"tokens_in":950,"tokens_out":2227,"duration_ms":17112,"temperature":1.0,"reasoning_tokens":2141,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T17:11:10.622889+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure $\\tau_{\\rm DRW}$ and $\\hat{\\sigma}$ for AGN with independent dynamical masses, for example megamaser disks that were not in the calibration sample, and check whether Equation 2 recovers their masses within the quoted 0.39 dex scatter; alternatively, repeat the DRW fits on light curves from a survey with independent error calibration, such as ZTF or LSST, and test whether the inferred masses agree with the ASAS-SN estimates.","supporting_citations":[],"review_version":1}