{"id":"32e65df7-7bde-4ae8-ad1f-5aecf63febd9","arxiv_id":"2411.18917","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Mrk 1018 completed a full changing-look cycle in about a year, showing that broad emission lines track a brief accretion-powered outburst.","lead":"Astronomers monitored the changing-look active galaxy Mrk 1018 for seven years and caught its nucleus rebrightening briefly around 2020, with broad emission lines appearing and then vanishing within a year. The event supports the idea that such transitions are driven by changes in the rate at which gas falls onto the supermassive black hole.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Stress-test concern: the FWHM/σline dichotomy and the double-peaked profile interpretation depend on the forced two-Gaussian decomposition; without a nonparametric cross-check, the claimed FWHM/σline jump from 2.96 to 1.71 and the virial −0.27 exponent are not secure.","rationale":"The reader's weakest_assumption is that the two-Gaussian model captures the true line profile; my stress-test identifies the same loaded assumption as the critical gate. The paper's strongest and most broadly claimed result—the full-cycle changing-look transition around 2020—is well supported by the dense LJT spectroscopy and multi-band light curves, and the outburst timing, Gaussian light-curve fits, and rough time delays are internally consistent, though seasonal gaps and sparse rising-phase data limit the light-curve shape. However, the higher-level conclusions about BLR physics (virial exponent, turbulent broadening, state transition) depend on the derived line-width parameters. Without a nonparametric or single-component cross-check, the risk that the two-Gaussian decomposition manufactures the double-peaked profile and the FWHM/σline transition is too high to accept the full physical narrative. The also-noted tautology in computing Eddington ratio from the same Hα luminosity that defines the BLR properties is a secondary concern: it does not break the correlation, but it weakens the causal interpretation 'accretion regulates the BLR.' The paper contains useful independent support (host-galaxy-based M•, consistent M–σ, [O III] scatter of 6.9%), but none of that validates the line-profile decomposition. I therefore agree with the reader's CONDITIONAL verdict, with the concrete conditional being a nonparametric refit and profile-shape sanity check.","tokens_in":35318,"tokens_out":1997,"duration_ms":18888,"concrete_test":"Refit a representative subset of the 96 LJT spectra (at least all spectra with Hβ/[O III] > 0.1) using: (1) a single broad Gaussian plus a narrow-line system, (2) a single broad Gaussian plus an asymmetric wing component, and (3) a nonparametric profile where FWHM is measured at the actual half-maximum of the observed broad-line residual and σline is the second moment of that residual. If the single-component or nonparametric fits match the data equally well (Δχ²/dof < 1) and yield FWHM/σline values that stay near 2.35–3 across all epochs, then the claimed FWHM/σline jump from 2.96 to 1.71 and the peak-separation trend would not be robust; the paper should report both parametric and nonparametric measurements and show the profile-shape evolution for a few representative epochs.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central physical narrative—accretion-regulated type transition with a virialized BLR, a sharp FWHM/σline drop, and an ADAF-to-SSD transition—rests on the line-width measurements from the two-Gaussian decomposition and on how the time-dependent line profile is summarized into single FWHM/σline values. The key unvalidated step is that a two-Gaussian model genuinely captures the double-peaked BLR in every epoch, rather than being a convenient parameterization that absorbs an unresolved single broad component, blueshifted narrow-line residuals, or asymmetric profile shapes into a second Gaussian. If the true profile is a single broad component with an asymmetric wing or a narrower core, then the derived peak separation, FWHM, and σline are artifacts of the Gaussian decomposition, and the reported correlations in Figure 9 (peak-separation trend, FWHM∝(Lbol/LEdd)^−0.27, the FWHM/σline jump from 2.96 to 1.71 at the type 1.5/1.2 boundary) would not be physically meaningful. The claim that the −0.27 exponent follows from the Bentz et al. (2013) R–L relation is also partly tautological: the Eddington ratio is computed from the same Hα luminosity (via Greene & Ho 2005) used to define type and line strength, and the FWHM is measured from the same spectra, so part of the correlation may be driven by the shared L5100 in both axes. A decisive check is to refit the spectra without forcing two Gaussians on the broad Balmer lines, and to recompute FWHM/σline directly from the observed profile moments (e.g., line dispersion from the second moment and FWHM from the actual half-max points) with no parametric decomposition.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Mrk 1018 is a changing-look AGN observed with Lijiang 2.4 m spectroscopy over seven seasons (2017-2024). The paper reports a full-cycle changing-look transition around 2020 in which the AGN type varied from 1.8 to 1.5/1.2 and back within about one year, accompanied by a multi-wavelength outburst. Combining these spectra with nine archival epochs, the authors reconstruct a 45-year type sequence and derive Eddington ratios from broad H-alpha luminosities. They report that the broad Balmer lines are single-peaked in Type 1.0-1.2 and double-peaked in Type 1.5-1.8, that the double-peak separation and FWHM decrease with increasing Eddington ratio, with FWHM proportional to (Lbol/LEdd)^-0.27, and that FWHM/sigma_line drops from 2.96 to 1.71 across the Type 1.5/1.2 boundary. They interpret these trends as evidence for a virialized BLR with accretion-dependent turbulence and for a transition from an inner ADAF to an outer standard disk. A virial black hole mass of (1.64 +/- 0.98) x 10^8 solar masses, consistent with the M-sigma mass, is derived from the H-beta lag and FWHM.","tokens_in":35703,"tokens_out":7462,"duration_ms":61856,"significance":"The paper's main value is the dense, well-calibrated spectroscopic sampling of a full cycle of changing-look behavior in a single object, including the detection of an outburst with clear multi-band time delays and a full cycle of type transitions. The data reduction appears careful: the [O III] lambda5007 light curve has only 6.9% scatter, and the comparison with 1ES 1927+654 is instructive. If the line-profile interpretation is correct, the reported transition from double-peaked to single-peaked broad lines and the sharp change in FWHM/sigma_line would be an important constraint on BLR geometry and the accretion-state transition. However, the primary physical conclusions are currently conditional on the assumed two-Gaussian decomposition of the broad lines and on Eddington ratios derived from the same broad-line fluxes, and the significance is therefore limited until these are validated.","major_comments":[{"comment":"The two-Gaussian model for the broad Balmer lines is adopted a priori in every epoch, and the quantities that drive the main physical claims—FWHM, sigma_line, and double-peak separation—are measured from this model. The manuscript does not provide a comparison with a single-Gaussian fit or with nonparametric measurements of the line width, nor does it show that the second Gaussian is not absorbing narrow-line residuals or an asymmetric wing. Without such a cross-check, the reported FWHM/sigma_line drop from 2.96 to 1.71 at the Type 1.5/1.2 boundary and the decreasing peak-separation trend in Figure 9c could be artifacts of the forced two-component decomposition. I request a model-selection test (e.g., AIC/BIC for one vs. two Gaussians) and direct measurement of FWHM and the second moment on the continuum-subtracted profiles for representative spectra of each type.","section":"Section 2.2, Figure 9"},{"comment":"The Eddington ratio on the horizontal axes is obtained from the broad H-alpha luminosity via the Greene & Ho (2005) scaling, while the AGN type on the vertical axis in Figure 9a is defined by the broad H-beta/[O III] ratio and the FWHM in Figure 9d is measured from broad H-beta. Because both axes are constructed from the same broad-line spectra, part of the reported correlations is built in by construction: a larger broad-line luminosity raises the Eddington ratio and also makes the object appear more Type 1. An independent estimate of L_bol (for example from the fitted power-law continuum at 5100 Å, or from X-ray data) is needed to establish that the accretion rate genuinely regulates the type and line-width changes rather than merely re-expressing the same broad-line strength.","section":"Section 3.4, Figure 9d; Section 3.3, Figure 9a"}],"minor_comments":[{"comment":"There are numerous typographical and grammatical errors, including 'a accretion disk' (Section 1), 'qui pped' (Section 2.1), 'the the telluric' (Section 2.1), 'Out findings' (Section 3.4), and 'a a reference' (Section 2.2); a thorough language edit is needed.","section":"Throughout"},{"comment":"In panel (a), the y-axis labels for H-alpha and H-beta list units as 'erg s cm^-2' rather than 'erg s^-1 cm^-2'.","section":"Figure 6"},{"comment":"The conclusion that the BLR size exceeds the R-L relation during the outburst relies on time delays derived from Gaussian peak offsets, but the rising phase is poorly sampled and the Gaussian shape is assumed; the authors should present cross-correlation lags and explicitly discuss the impact of the sparse rising-phase sampling on the peak-time measurements.","section":"Section 4.2, Figure 10"},{"comment":"The reference 'Wu et al. 2024, submitted' is to an unpublished manuscript; it should be replaced with a published reference or removed, or its status clearly marked as 'in preparation'.","section":"Section 3.4"},{"comment":"The phrase 'full-cycle changing-look transition' is used to describe a transition from Type 1.8 to 1.2 and back to 1.8; the authors should define what constitutes a 'full cycle' in this context, since the broad lines do not completely disappear.","section":"Abstract and Section 5"}],"recommendation":"major_revision","confidential_remarks":"The paper is well suited to the journal if the model-dependence of the line-width measurements is addressed. The circularity concern regarding the Eddington-ratio axes could be resolved with an independent luminosity estimate. I also note that the use of an 'in preparation' reference in a submitted manuscript is not appropriate for a refereed journal. The central observational dataset is unique and valuable, and the authors have done a careful job with calibration and decomposition, but the physical narrative currently outpaces the validation of the key assumption."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a valuable single-object study that catches the full cycle of a changing-look transition in Mrk 1018, something the field has not had with this density. The 2020 outburst and the rapid Type 1.8→1.2→1.8 swing are well documented with seven seasons of LJT spectra, and the [O III] stability at 6.9% scatter speaks to careful calibration. The Balmer decrement turnover and the decreasing double-peak separation with Eddington ratio are genuinely new, and the comparison with 1ES 1927+654 that rules out a TDE is thoughtful.\n\nThe soft spots are real but not fatal. The biggest one is the two-Gaussian decomposition of every broad Balmer line. All the profile-width trends—FWHM, σline, FWHM/σline, peak separation—depend on that choice. The authors say they tried Lorentzians and Gaussians won, but they never show a nonparametric measurement (e.g., line dispersion from the second moment, FWHM from half-max points) to confirm the FWHM/σline jump from 2.96 to 1.71 is not an artifact of forcing two components. A referee should ask for that cross-check.\n\nThe accretion-rate correlation is also partly built in, since the Eddington ratio is computed from the same Hα luminosity that defines the broad-line strength. That does not invalidate the type-accretion link—the narrow [O III] is constant—but it weakens the claim that the correlation is independent evidence. Similarly, the FWHM∝(L/LEdd)^−0.27 relation is a prediction from the Bentz R–L relation and the virial assumption, not a new empirical fit; the paper frames it as consistency, which is honest, but it should not be sold as a discovery.\n\nThe archival points are sparse and heterogeneous, so the 45-year trends rest on a handful of epochs. Again, that is acknowledged and the new data carry most of the weight.\n\nBottom line: the core observational result—a full-cycle changing-look transition within a year with correlated multi-band variability—is solid and deserves to be published. The BLR physics interpretation is suggestive but needs the nonparametric line-shape check and a clearer statement about circularity. I would send it to a serious referee, with a request for those analyses.","headline":"Dense monitoring catches a full changing-look cycle in Mrk 1018 with novel profile trends, but the line-decomposition assumptions need a nonparametric cross-check before the BLR physics claims fully land.","tokens_in":36296,"tokens_out":2941,"would_cite":true,"duration_ms":25934,"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":"Seven seasons of spectroscopy catch a full changing-look cycle in Mrk 1018 in one year, tied to an accretion-driven outburst.","keywords":["changing-look AGN","Mrk 1018","broad-line region","Eddington ratio","accretion disk instability","double-peaked emission lines","Balmer decrement","reverberation mapping"],"falsifier":"A decisive test would be to obtain dense, high-signal-to-noise spectroscopy covering the full rising phase of a similar outburst in a low-Eddington AGN; if the broad lines remained single-peaked while the type moved from 1.8 to 1.2, or if the double-peak separation did not shrink as the Eddington ratio rose, the accretion-regulated BLR restructuring described here would be falsified. Alternatively, resolving the line profile with a non-parametric model rather than two fixed Gaussians in the Type 1.5-1.8 state of Mrk 1018 would show whether the double peaks are true components or a fitting artifact.","tokens_in":35122,"feed_emoji":"🌌","tokens_out":5385,"duration_ms":41430,"temperature":0.7,"pith_summary":"The paper reports that the changing-look active galactic nucleus Mrk 1018 completed a full cycle of type transitions, from Type 1.8 to Type 1.5/1.2 and back, in about one year around 2020, riding on a nucleus outburst. Using seven seasons of spectroscopy plus archival data spanning 45 years, the authors argue that the transition is driven by the accretion rate, which varied by a factor of 1000 and peaked at 2 percent of Eddington. They tie the outburst to a disk instability in the transition region between an outer thin disk and an inner advection-dominated accretion flow. The broad Balmer lines change from single-peaked to double-peaked as the type drops, with the double-peak separation shrinking as accretion rises, and the FWHM follows the virialized scaling FWHM proportional to (Lbol/LEdd)^-0.27. The result matters because it shows changing-look behavior can be a fast, accretion-regulated restructuring of the broad-line region rather than only obscuration or a rare tidal event.","feed_headline":"Mrk 1018 flips its AGN type twice in one year","feed_subtitle":"Seven-season spectra tie the rapid flip to an accretion outburst and a disk instability.","key_machinery":"The central observable machinery is the decomposition of every broad H-$\\alpha$ and H-$\\beta$ profile into two Gaussians, which yields the FWHM, the velocity dispersion sigma_line, and the peak separation of the double-peaked profiles across 45 years. This decomposition, combined with the Eddington ratio as the driving variable, produces the paper's core scaling relations: FWHM proportional to (Lbol/LEdd)^-0.27, a decreasing double-peak separation with accretion, and a FWHM/sigma_line ratio that jumps between 2.96 (rectangular profiles, Type 1.8-1.5) and 1.71 (Lorentzian-like, Type 1.2-1.0). The proposed mechanism is a radiation-pressure instability in the transition zone between the outer standard thin disk and the inner advection-dominated accretion flow, which produces the observed one-year outburst and the synchronized changing-look behavior.","core_discovery":"We find that Mrk 1018 underwent a full-cycle changing-look transition within one year, around 2020, from Type 1.8 to Type 1.5/1.2 and back to Type 1.8, synchronized with a multi-band nucleus outburst. Over 45 years its Eddington ratio swung by a factor of 1000, reaching 0.02 at the peak, and the type follows accretion rate. For the first time we observe a turnover in the Balmer decrement with accretion, a decrease of double-peak separation as accretion increases, and a sharp rise in sigma_line at high Eddington ratio that drops FWHM/sigma_line from 2.96 to 1.71, signaling enhanced turbulence. The FWHM obeys the virial expectation FWHM proportional to $L5100^{-0}$.27, so the BLR remains virialized while turbulent motions grow with accretion, likely due to a transition from an advection-dominated to a standard disk. These observations establish accretion-regulated BLR evolution as the driver of changing-look phenomena in this source.","pith_inferences":["If the same accretion-regulated behavior holds in other CL-AGNs, the diversity of broad-line profiles across the AGN population may be a single sequence in Eddington ratio rather than a set of distinct physical classes.","The one-year cycle implies that past claims about decades-long changing-look timescales may be biased by sparse sampling; denser monitoring could reveal many hidden fast cycles.","The apparent coupling between the low-ionization BLR and the hot dust torus (H-alpha lag close to W1 lag) suggests that simultaneous optical and mid-IR reverberation mapping of CL-AGNs might directly test whether the BLR is launched from the torus.","The sharp rise of sigma_line at high Eddington ratio could be observable as a profile-shape diagnostic in large spectroscopic surveys, potentially identifying the ADAF-to-SSD transition without X-ray data."],"forward_implications":["If the accretion-regulated cycle is real, then type transitions in CL-AGNs need not be slow or rare: a full cycle can occur on roughly a year timescale when the accretion rate surges.","The observed FWHM-Lbol/LEdd scaling means single-epoch virial black hole masses in changing-look sources can remain reliable even during type flips, as long as the broadening is corrected.","The Balmer decrement turnover at an Eddington ratio near 0.004 offers a spectroscopic indicator of the accretion state transition between ADAF and standard disk regimes.","The double-peak separation decreasing with rising accretion provides a direct test for disk-wind or disk-like BLR models such as FRADO or Keplerian disk models."],"supporting_citations":[{"why":"Documents the original type change of Mrk 1018 from 1.9 to 1.2, establishing the historical changing-look baseline.","marker":"Cohen et al. 1986"},{"why":"Records the return of Mrk 1018 to type 1.9 in 2015, providing the other end of the decades-long cycle.","marker":"McElroy et al. 2016"},{"why":"Supplies the earlier spectra of Mrk 1018 that the authors recalibrated to extend the 45-year spectral baseline.","marker":"Kim et al. 2018"},{"why":"Presents the disk-instability model in the transition region between SSD and ADAF that the authors use to explain the outburst and changing-look timescale.","marker":"Sniegowska et al. 2020"},{"why":"Provides the radius-luminosity relation used to derive the expected FWHM scaling and to compare measured BLR sizes.","marker":"Bentz et al. 2013"},{"why":"Defines the FWHM/sigma_line diagnostics for line-profile shape (Gaussian at 2.35, Lorentzian below 2.35) used to interpret the profile transitions.","marker":"Kollatschny & Zetzl 2011"},{"why":"Supplies the H-alpha luminosity to 5100 angstrom luminosity relation used to estimate Eddington ratios.","marker":"Greene & Ho 2005"}],"fun_headline_variants":["Mrk 1018 flips AGN type twice in one year, accretion-driven","Accretion rate swings 1000x as Mrk 1018 changes look repeatedly","First observed Balmer turnover in changing-look AGN Mrk 1018","Disk instability triggers rapid full-cycle changing-look in Mrk 1018","Virialized BLR with turbulent motions: new AGN changing-look model"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that every broad Balmer line across all 45 years is truly composed of two Gaussian components, so the measured FWHM, sigma_line, and double-peak separation reflect real kinematics rather than fitting artifacts.","fun_headline_variants_meta":{"raw":{"variants":["Mrk 1018 flips AGN type twice in one year, accretion-driven","Accretion rate swings 1000x as Mrk 1018 changes look repeatedly","First observed Balmer turnover in changing-look AGN Mrk 1018","Disk instability triggers rapid full-cycle changing-look in Mrk 1018","Virialized BLR with turbulent motions: new AGN changing-look model"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000894,"raw_usage":{"total_tokens":3957,"prompt_tokens":1153,"completion_tokens":2804,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":769,"completion_tokens_details":{"reasoning_tokens":2696}},"tokens_in":769,"tokens_out":2804,"duration_ms":19269,"temperature":1.0,"reasoning_tokens":2696,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T10:44:53.303543+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive test would be to obtain dense, high-signal-to-noise spectroscopy covering the full rising phase of a similar outburst in a low-Eddington AGN; if the broad lines remained single-peaked while the type moved from 1.8 to 1.2, or if the double-peak separation did not shrink as the Eddington ratio rose, the accretion-regulated BLR restructuring described here would be falsified. Alternatively, resolving the line profile with a non-parametric model rather than two fixed Gaussians in the Type 1.5-1.8 state of Mrk 1018 would show whether the double peaks are true components or a fitting artifact.","supporting_citations":[],"review_version":1}