{"id":"04fdf98a-6d3d-4573-8910-93bab359c3f8","arxiv_id":"2607.16423","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"In 14 hard-X-ray-selected AGN monitored for two years, variability is strongest in X-rays and weakest in radio, and source variability explains only ~2-3% of the scatter in the Fundamental Plane.","lead":"This paper tracks 14 hard-X-ray-selected active galaxies across radio, optical and X-ray bands for two years, measuring how strongly each band flickers. It finds X-rays vary most, then optical, then radio, and that source flickering adds only a few percent to the scatter of the black-hole 'fundamental plane' relation.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Sample selection excludes previously radio-monitored sources and QSO/blazars, biasing the radio variability fraction and the amplitude hierarchy; the central claim of radio stability may be sample-dependent.","rationale":"The reader's weakest_assumption targets the FP variability-dispersion formula, but that assumption is not the most load-bearing: even if the plane slope or covariance differs, the qualitative conclusion that variability contributes only a few percent to the total scatter survives because the per-source radio and X-ray fractional variabilities are small (median 10% and 30%, respectively). The sample-selection bias, however, directly affects the headline numbers: the radio variability fraction and the median radio F_var are computed after deliberately excluding the objects most likely to be radio-variable (blazars, QSOs, and previously monitored sources). This makes the reported radio stability and the inferred core-dominated, non-variable 15 GHz emission potentially unrepresentative. The paper itself acknowledges the red-noise caveat on variability significances (Section 4) and the exclusion criteria (Section 2), but it does not test how those exclusions alter the central hierarchy. A concrete comparison with archival light curves of the excluded sources would settle this. Since this concern reinforces the reader's conditional verdict rather than overturning it, I recommend UNCHANGED.","tokens_in":35364,"tokens_out":13876,"duration_ms":134085,"concrete_test":"Recompute the radio variability statistics (detection fraction, median F_var, Mexican Hat variances) after adding published radio light curves for the excluded previously monitored sources (e.g., Mrk 110 from Panessa et al. 2022b, NGC 7213 from Bell et al. 2011, NGC 2992 from Fernandez et al. 2022), processed with the same F_var definition. If the combined radio median F_var exceeds ~15% or the detection fraction exceeds 50%, the reported hierarchy is an artifact of sample selection.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The sample is constructed by excluding BL Lacs, blazars, and QSOs (Section 2), and also 'the sources which have already been radio-monitored in the past' and those with only 15 GHz upper limits. The first two exclusions preferentially remove the most radio-variable AGN classes, and the third removes known radio variables from the literature (e.g., Mrk 110, NGC 7213). Consequently, the reported radio detection fraction (6/14 = 43% in the text, though Table 2 implies 5/14 = 36%) and the median radio F_var = 10% are likely lower limits. The central claim of a clear amplitude hierarchy with radio the least variable band, and the inference that 15 GHz emission is a stable, core-dominated component, rest on this possibly biased sample. The paper does not quantify how the excluded sources would affect the radio variability statistics. Section 2 states the exclusions but does not assess their impact on the variability conclusions.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a two-year multi-wavelength monitoring campaign of 14 hard X-ray selected AGN drawn from the INTEGRAL/IBIS catalog, using AMI-LA at 15 GHz, ZTF g/r photometry, and Swift/XRT. Variability is quantified with normalized excess variance, fractional rms amplitude, and Mexican Hat variances at 70 and 200 days. The authors report significant variability in 86% of the sample, an amplitude hierarchy (X-ray ~30%, optical g ~19%, optical r ~8.5%, radio ~10%), a red-noise power spectrum, and a Fundamental Plane (FP) analysis in which variability-induced dispersion accounts for only ~2–3% of the total FP scatter. They conclude that the 15 GHz emission is core-dominated and that non-simultaneity is not a dominant source of FP scatter.","tokens_in":35620,"tokens_out":6078,"duration_ms":60771,"significance":"If the results hold, this is a useful multi-band variability census of hard-X-ray-selected AGN, with the notable strength of quasi-simultaneous radio, optical, and X-ray coverage over a two-year baseline. The paper is transparent about the red-noise limitation in §4 and performs a careful RISS check in §6.2. However, the central claims — the radio-stability hierarchy and the small FP variability contribution — are currently weakened by sample-selection choices and by the FP propagation calculation. The dataset and methodology are valuable, and the stated caveats show good faith, but the headline numbers need revision or more careful qualification before the conclusions can be accepted.","major_comments":[{"comment":"The sample selection excludes BL Lacs, blazars, QSOs, and sources with previous radio monitoring or 15 GHz upper limits. These exclusions preferentially remove the most radio-variable AGN classes and known radio variables. The reported radio detection fraction (43% in the text, but 5/14 in Table 2) and the conclusion that 15 GHz emission is stable and core-dominated are therefore lower limits conditioned on censoring the most variable sources. Please quantify the impact of the excluded objects or explicitly restrict the conclusion to the selected sample.","section":"§2; Table 2; §7 item 1"},{"comment":"The variability-induced FP dispersion is computed as S_var = sqrt((ξ σ_X)^2 + σ_R^2) with ξ=0.6 taken from Merloni et al. (2003), the very relation being tested. For a mostly radio-quiet sample, the Bariuan et al. (2022) RQ slope (ξ=0.38) may be more appropriate; moreover, the radio and X-ray fluctuations are treated as independent, but no per-epoch covariance is reported. The conclusion that variability contributes only ~2–3% of the FP scatter depends directly on these choices. Please provide a sensitivity analysis and report the covariance or justify the independence assumption.","section":"§5.4; Eq. (1)"},{"comment":"The paper states that the additional scatter from red-noise stochasticity 'cannot be formally quantified' without simulations, yet §5.1 adopts a 3σ threshold on S = σ^2_NXS/σ^2_NXS,err and bases the 86% detection rate and the amplitude hierarchy on these thresholds. For red-noise light curves, the measurement-noise-only error understates the true uncertainty, so the detection significance is likely overestimated. The headline detection fraction should either be accompanied by simulation-based significance estimates or explicitly labeled as provisional modulo red-noise uncertainty.","section":"§4; §5.1"},{"comment":"There is an internal inconsistency in the radio variability count. Table 2 shows Y in the radio column for 5 sources (QSO B0241+62, LEDA 168563, MCG+08-11-11, NGC 4388, NGC 5252), while §7 states '6 (43%) in the radio.' The Table 2 Var. string format is also ambiguous (e.g., 'N - N - Y'). Please correct the count and clarify the column layout.","section":"Table 2; §7 item 1"},{"comment":"The ZTF variability threshold R≥3 is calibrated using NGC 4388 and IGR J23308+7120 as empirical noise baselines, and the same sources are then classified as non-variable in the optical. This circularity guarantees their optical non-variability and propagates into the statement that these two objects show no variability in any band. An independent noise sample (e.g., stars or non-varying sources in the field) or simulations should be used to set the threshold.","section":"Appendix B; Table B.1"}],"minor_comments":[{"comment":"The abstract and conclusions quote '~3%' for the variability-induced FP dispersion, while §5.4 states 2.16%; unify the numbers.","section":"Abstract; §5.4"},{"comment":"The header 'Var.' is described as 'ot' in the printed line; fix the typo.","section":"Table 2"},{"comment":"The symbol R is used both for the ZTF variability ratio (Appendix B) and for the radio-loudness parameters R_opt and R_X; rename one to avoid confusion.","section":"§4; §5.3"},{"comment":"The text says Mkn 3 'had no useful optical data' but Appendix B explains it was excluded because of error underestimation; please align these descriptions.","section":"§5.1; Appendix B"},{"comment":"The radio data points have very large uncertainties; consider using weighted means or a separate panel to show the trend more clearly.","section":"Fig. 3"}],"recommendation":"major_revision","confidential_remarks":"The paper presents a valuable simultaneous multi-wavelength dataset, and the core variability analysis is competently executed. However, the two headline conclusions — the radio-amplitude hierarchy and the small FP scatter contribution — are not yet robust to the sample-selection censoring and to the circularity in the FP propagation formula. These are fixable with additional analysis and/or softened wording. I recommend major revision before acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth a careful read. The paper delivers a genuinely new monitoring campaign: 14 hard-X-ray-selected AGN with simultaneous AMI 15 GHz, ZTF g/r, and Swift/XRT over two years. The headline results — variability in 86% of the sample, X-rays > optical > radio in fractional rms, red-noise PSD from the Mexican Hat analysis, and a ~2-3% contribution of source flickering to Fundamental Plane scatter — are new numbers for this INTEGRAL sample. The data handling is careful: phase-calibrator de-trending for the radio, a real attempt to validate ZTF photometric errors against DR19, and explicit statements about what the estimators can and cannot do. Credit where due: the paper does not oversell the significance of the variability detections; it states that red-noise scatter cannot be formally quantified.\n\nThe soft spots are real but not fatal. The sample is cut to exclude blazars, QSOs, and previously radio-monitored sources. That means the radio variability fraction (5/14 by my count of Table 2, despite '6/14 = 43%' in the text) and the median radio F_var are lower limits. The inference that 15 GHz emission is a stable, core-dominated component is the one that depends most on this selection. I would want that discussed. Second, the FP dispersion estimate uses xi=0.6 from Merloni et al. (2003) — the very relation being tested — and assumes independent radio and X-ray fluctuations. With xi=0.38 (Bariuan et al. RQ), the contribution changes; the conclusion that non-simultaneity is not the dominant scatter is probably safe, but the 2-3% number is softer than it looks. Third, the ZTF variability thresholds were calibrated on the sample's own Seyfert 2s, then applied to those same objects. Minor circularity; acceptable heuristic, but worth acknowledging. Fourth, no data or code are released, so the light curves and the Mexican Hat values cannot be independently checked.\n\nNone of this sinks the central claims. The chi-square values are huge; the amplitude hierarchy is robust. The paper is a solid, if not revolutionary, contribution. I would send it to a referee. I'd probably ask the referee to check the radio selection effects and the FP propagation formula, and to enforce a fix of the 5/6 radio-detection inconsistency.\n\nFor a reading group: if you work on AGN variability or the Fundamental Plane, worth an hour. I would cite it for the FP scatter estimate, with caveats.","headline":"Solid observational campaign with a useful new variability census; the radio-stability conclusion is sample-dependent and the FP scatter estimate is softer than the abstract suggests.","tokens_in":36098,"tokens_out":3789,"would_cite":true,"duration_ms":55799,"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":"This paper claims that variability-induced shifts account for only ~2% of the Fundamental Plane's scatter in a sample of 14 AGN, even though 86% of them vary significantly across radio, optical and X-ray.","keywords":["active galactic nuclei","AGN variability","Fundamental Plane of black hole activity","radio loudness","accretion-ejection coupling","red noise","multi-wavelength monitoring","X-ray corona"],"falsifier":"Measure the joint X-ray-radio covariance on the same epochs; if the covariance term is substantial or using a flatter plane slope (0.38) raises the variability contribution above ~10%, the claim that variability contributes only ~2% fails.","tokens_in":35302,"feed_emoji":"🔭","tokens_out":10180,"duration_ms":85818,"temperature":0.7,"pith_summary":"This paper tries to establish how and where AGN variability arises by tracking 14 hard X-ray selected AGN for two years at 15 GHz, in the optical g and r bands, and in X-rays, all at roughly the same times. It finds that 86% of the sources vary significantly, with a clean amplitude hierarchy — X-rays wiggle most (median ~30%), then optical g (~19%), then radio and optical r (~10% and ~8.5%) — and that the power grows from 70-day to 200-day timescales, a classic red-noise signature. The load-bearing result concerns the Fundamental Plane of black hole activity: although individual sources wander along the relation, the variability-induced dispersion (0.118 dex) amounts to only ~2% of the plane's total scatter. If true, this means that using archival, non-simultaneous radio and X-ray data to build the plane does not materially inflate its scatter, and that the compact 15 GHz core is a stable feature — likely a jet base or magnetized corona — rather than an incidental extended structure.","feed_headline":"86% of AGN flicker, yet black-hole scaling barely moves","feed_subtitle":"A two-year, three-band campaign shows the black-hole scaling law survives source flicker, so archival data can be trusted.","key_machinery":"The central machinery is the multi-band light-curve analysis, specifically the normalized excess variance and fractional rms amplitude (F_var) as intrinsic variability estimators, and the Mexican Hat filter — a time-domain wavelet filter that isolates the variance at selected timescales (here 70 and 200 days) without red-noise leakage. The argument that variability barely affects the Fundamental Plane rests on the variance-propagation formula S_var = sqrt((ξ σ_X)^2 + σ_R^2), where σ_X and σ_R are the X-ray and radio fractional variability amplitudes and ξ is the X-ray slope of the plane, here fixed at 0.6, the classical value. This formula converts the observed per-band flicker into an expec","core_discovery":"Across a sample of 14 hard X-ray selected AGN monitored quasi-simultaneously in the radio (15 GHz), optical (g and r), and X-ray (2-10 keV) bands over 2018-2020, significant intrinsic variability is detected in 86% of the sample. The fractional rms amplitude is stratified by band: X-ray ~30% (range 11-67%), optical g ~19% (2-33%), optical r ~8.5% (0.2-24%), and radio ~10% (4-23%). The Mexican Hat wavelet filter shows variance increasing from the 70-day to the 200-day scale in all bands, indicating red-noise, long-term-dominated power spectra. Placing the sources on the Fundamental Plane of black hole activity (radio luminosity vs. X-ray luminosity and black hole mass) shows the sample follow","pith_inferences":["If the 2% estimate holds up, it implies the Fundamental Plane's ~0.8-dex scatter is dominated by intrinsic physical differences (e.g., black hole spin, magnetic field topology, accretion state), not by measurement timing — a conclusion that would refocus theoretical work on explaining that intrinsic spread.","A sharper test would be to compute the joint radio-X-ray covariance at each epoch; the current analysis treats the bands as independent, so any correlated behavior would change the variance budget. If correlated fluctuations are present, the variability contribution could be higher than reported, and the plane might not be as robust as claimed.","Applied to larger samples with denser cadence, this variance-decomposition method could identify which sources are genuinely off-plane because of state changes, versus those that are simply flickering, turning the plane into a tool for classifying accretion-ejection states.","Since the Mexican Hat shows red noise with power still rising at 200 days, longer monitoring campaigns should reveal even larger long-timescale excursions; if those excursions are correlated across bands, the 2% figure may be a lower bound for the variability contribution."],"forward_implications":["If the ~2-3% variability contribution is general, non-simultaneous archival radio and X-ray measurements can be combined to build the Fundamental Plane without a major scatter penalty, extending its use to larger samples.","The amplitude hierarchy (X-ray > optical g > radio and optical r) and its red-noise character imply physically distinct emission regions: a compact X-ray corona, an extended accretion disk smoothing optical fluctuations, and a stable compact radio core.","The stability of the 15 GHz core over a two-year baseline supports models where radio emission in radio-quiet AGN comes from a jet base or corona rather than large-scale outflows or star-forming regions.","The diversity in variability patterns — correlated multi-band flicker in some sources, decoupled radio or X-ray in others — suggests that AGN occupy a range of corona-jet coupling strengths, and that radio-loudness alone is not sufficient to predict that coupling."],"fun_headline_variants":["AGN flicker in every band, but black-hole scaling holds","86% of AGN flicker, but black-hole scaling stands","Three-band survey: red noise everywhere, scaling intact","Flickering AGN add just 3% scatter to the fundamental plane","X-rays flicker most, but the black-hole plane stays put"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The estimate that variability adds only about 2% of the plane's scatter assumes the X-ray and radio signals flicker independently and obey a particular scaling slope; if they flicker together, or the true slope is flatter, the contribution could be much larger.","fun_headline_variants_meta":{"raw":{"variants":["AGN flicker in every band, but black-hole scaling holds","86% of AGN flicker, but black-hole scaling stands","Three-band survey: red noise everywhere, scaling intact","Flickering AGN add just 3% scatter to the fundamental plane","X-rays flicker most, but the black-hole plane stays put"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000616,"raw_usage":{"total_tokens":2799,"prompt_tokens":947,"completion_tokens":1852,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":691,"completion_tokens_details":{"reasoning_tokens":1772}},"tokens_in":691,"tokens_out":1852,"duration_ms":15466,"temperature":1.0,"reasoning_tokens":1772,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T20:58:16.179558+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the joint X-ray-radio covariance on the same epochs; if the covariance term is substantial or using a flatter plane slope (0.38) raises the variability contribution above ~10%, the claim that variability contributes only ~2% fails.","supporting_citations":[],"review_version":1}