{"id":"6a06b02b-6916-4991-9e76-79b8ff1ac9a1","arxiv_id":"1908.05461","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Using Swift BAT hard X-rays, the authors show that NGC 4151's UV variability cannot come from a truncated disc within a few hundred gravitational radii, but is consistent with BLR-scale reprocessing at 1.5-20 light-days.","lead":"This paper shows that the standard model in which a hot X-ray corona illuminates an inner accretion disc cannot explain the observed UV variability of NGC 4151. Instead, the UV variations are consistent with X-ray reprocessing by gas on broad-line-region scales, suggesting the inner disc is replaced by a hot flow.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"BAT 15–50 keV driving light curve may not track the bolometric reprocessing photons, leaving the 1.5–20 day IRF and the 'rule out' statement dependent on an untested bandpass assumption.","rationale":"The reader identifies the same load-bearing assumption: the BAT 15–50 keV band tracks the bolometric X-ray driver. The paper's own framing makes this assumption central: it argues that previous campaigns failed because XRT below 10 keV did not track the bolometric flux, and that BAT is better because it reaches 15–50 keV. But the argument only relocates the bandpass problem upward; the bolometric luminosity of NGC 4151 during this campaign is stated to peak above 50 keV, so the 15–50 keV band is still a partial bolometric tracer. The paper does not demonstrate that the 15–50 keV band is linearly proportional to the full coronal luminosity on the relevant timescales, nor that spectral variability (which would break proportionality) is absent. I agree with the reader's conditional verdict, and with the emphasis on the missing uncertainties in the IRF. I would add that the absence of error bars on the IRF timescale and fractions (Section 4) combines badly with the bandpass proxy issue: even if the IRF fit is good, its physical interpretation as BLR-scale reprocessing assumes the input light curve is the true driver. A bolometric-proxy test is concrete and decisive because both the negative disc result and the positive BLR interpretation flow through the assumed driver. Independent support for the BLR interpretation exists (BLR line sizes, energetics check within the paper), but those support the plausibility of the interpretation, not the validity of the bandpass proxy. The paper is otherwise internally consistent: the SED decomposition robustness is addressed in Appendix A, the host-galaxy subtraction uncertainty is shown not to affect UVW1, and the physical model predictions are clearly described. Thus the single load-bearing concern is the BAT bandpass proxy, and the appropriate verdict remains CONDITIONAL until it is tested.","tokens_in":22259,"tokens_out":1948,"duration_ms":16715,"concrete_test":"Re-run the Section 3 and Section 4 calculations twice, once with the 15–50 keV BAT light curve and once with a bolometric-correction-modulated light curve derived by combining BAT with simultaneous spectral-index or harder X-ray information (e.g. INTEGRAL/SPI or NuSTAR 3–79 keV where available). If the disc-model rejection and the recovered 1.5–20 day IRF persist under the bolometrically-corrected driver, the bandpass concern is retired; if the IRF centroid or the fast-variability fraction changes by more than the quoted uncertainties, the headline claim is not supported.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that optically thick material in a disc geometry on scales < 1.5 light days is ruled out rests on the assumption that the Swift BAT 15–50 keV light curve tracks the bolometric X-ray variability that drives the UV (Sections 1 and 3). The paper argues this is superior to XRT (< 10 keV) because the bolometric luminosity peaks above 50 keV, but BAT itself only reaches 50 keV. If the flux above 50 keV (or the spectral shape of the coronal emission) varies independently of the 15–50 keV band, the driving light curve used in equations (3) and (5) is not the true driver. Then both the disc-model failure and the recovered 1.5–20 day IRF could be artifacts of an incorrect driver, rather than evidence for the absence of small-scale optically thick material. The authors do not validate the 15–50 keV band as a bolometric proxy within this campaign, e.g. by comparing BAT variability to harder X-ray coverage or to the spectral-index variability that would indicate band-dependent changes. The IRF recovery in Section 4 explicitly assumes fc + fs + fast fraction parameterization with the BAT curve as input; systematic bandpass error maps directly into the recovered IRF timescale and fractions. This is the weakest load-bearing condition for the 'rule out' claim, and it is not internally tested.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper applies the Gardner & Done (2017) spectral-timing method to the 2016 Swift monitoring campaign of NGC 4151, using the BAT 15-50 keV light curve as the driving X-ray signal. The authors fit a truncated disc plus warm Comptonization model (agnsed) to the mean SED, inferring L/L_Edd ~ 1.4%. They then predict the UVW1 light curve from the observed BAT light curve and demonstrate that the predicted curve has far more fast variability and a shorter cross-correlation lag than the observed UVW1 curve. They next fit phenomenological linear transfer functions to the UVW1/BAT data, finding that a slowly responding component on 1.5-20 light-day timescales plus a small fast fraction describes the light curve well, and they interpret this as reprocessing by dense gas in the broad-line region. The paper concludes that optically thick material in a disc geometry is ruled out within 1.5 light days, supporting a changing-state picture in which the inner disc is replaced by a radiatively inefficient flow.","tokens_in":22648,"tokens_out":13646,"duration_ms":129281,"significance":"If the conclusions hold, the paper provides a strong observational constraint on accretion geometry in low-luminosity AGN: it would be the first to use a hard X-ray (BAT) driver to test disc reprocessing predictions, and it would support both the truncation of the thin disc at low Eddington ratios and the domination of the UV by diffuse BLR continuum. The manuscript is commendable for making an explicit, falsifiable prediction, for using a band less affected by absorption than the XRT, and for demonstrating robustness of the variability prediction to alternative SED decompositions in Appendix A. It also offers an energetic plausibility check for BLR reprocessing. However, the central claim rests on the unvalidated assumption that the 15-50 keV BAT band tracks the bolometric X-ray driver, and the recovered IRF parameters are presented without statistical uncertainties; both issues need to be addressed before the conclusions can be accepted at face value.","major_comments":[{"comment":"The central claim that the disc/warm-Compton reprocessing model is ruled out for radii < 1.5 light days assumes that the Swift BAT 15-50 keV light curve faithfully tracks the bolometric X-ray luminosity that drives the UV reprocessing. The paper argues that BAT is preferable to XRT because the latter is affected by variable absorption, but it does not provide any direct validation that the 15-50 keV band traces the harder X-rays where the bolometric luminosity peaks. If the flux above 50 keV has different variability amplitude or power spectrum, the predicted UVW1 light curve would change and the disc model could be reconciled with the data. The authors should either support this proxy assumption with auxiliary data (e.g., simultaneous INTEGRAL/NuSTAR coverage or a coherence analysis between BAT and XRT after correcting for absorption) or explicitly restate the rule-out as conditional on this assumption. This is load-bearing for the abstract's headline conclusion.","section":"Section 3, Eq. (3) and Fig. 8"},{"comment":"The positive result - that the UVW1 is predominantly reprocessed on BLR scales of 1.5-20 light days - is obtained by fitting the UVW1 light curve with the BAT light curve as the input driver. A systematic error in the driver (for instance, if the true bolometric flux is less variable on short timescales than the 15-50 keV band) would propagate directly into the recovered IRF timescales and the inferred fast fraction. The paper does not test how the recovered IRF changes for plausible alternative drivers, so the BLR-scale conclusion is not uniquely determined. The authors should include a sensitivity analysis (e.g., a driver with a different power spectrum or with spectral-index-dependent variability) or qualify the interpretation accordingly.","section":"Section 4, Eq. (5) and Fig. 10"},{"comment":"The IRF fits are presented without confidence intervals on the parameters fc, fs, centroids, variances, and derived size scales. The only uncertainty discussed is the systematic difference between the double-Gaussian and exponentially-modified-Gaussian forms. Since the 1.5 light-day lower limit and the 90% slow fraction are quantitative claims, a bootstrap or Monte Carlo error estimate on the recovered IRF parameters and the implied length scales is needed to support them.","section":"Section 4, parameter uncertainties"}],"minor_comments":[{"comment":"In the second term of the exponentially modified Gaussian IRF, the argument of the complementary error function uses lambda_1 sigma_2^2 where it should use lambda_2 sigma_2^2; as written the formula is internally inconsistent.","section":"Eq. (6)"},{"comment":"The reference entries for Edelson et al. (2015) and Edelson et al. (2017) are missing the publication years, and the entry for Welsh and Horne is listed as 2016 instead of 1991.","section":"References"},{"comment":"The phrase 'to produce give an even sampling' contains a typo; it should read 'to produce an even sampling.'","section":"Section 3, first paragraph"},{"comment":"The caption reads 'which is more consisted with the Bentz-corrected data'; this should be 'more consistent with.'","section":"Fig. 5 caption"},{"comment":"The significance of the BAT-UVW1 correlation is not quoted for the UVW1 band itself; the text cites E17's value for UVW2. The authors should report the significance for the specific band used in the analysis.","section":"Section 3.1"}],"recommendation":"major_revision","confidential_remarks":"The manuscript addresses an important question and the analysis is carefully conducted, but the headline claim is broader than the evidence supports because the BAT bandpass assumption is not validated. If the authors can provide additional support or appropriately qualify the claims, the paper would likely be suitable for MNRAS. No issues with citation or novelty were found."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper's negative result is strong and worth taking seriously; its positive BLR-reprocessing interpretation is plausible but much less certain. The analysis is careful, the writing clear, and the authors are honest about tensions in their picture.\n\nWhat's new: this is, to my knowledge, the first spectral-timing test of the lamppost reprocessing model driven by Swift BAT 15–50 keV light curves. NGC 4151 is the only AGN where BAT can be used this way during the E17 campaign. The authors build a full spectral-timing model with a hot corona, warm Comptonisation and outer disc, fit the time-averaged SED, and then predict the UVW1 variability from the observed BAT curve without re-fitting the UV data. The predicted light curve has fast peaks that are absent in the data, and the cross-correlation shows a much shorter lag and higher coherence than observed. That failure is robust to the SED decomposition (Appendix A) and to host-galaxy subtraction. It is a clean kill of the standard inner-disc picture for this source.\n\nThe soft spots are two. First, the 15–50 keV band is assumed to track the bolometric X-ray luminosity. The paper doesn't test whether the spectrum above 50 keV varies independently, and it doesn't compare with harder X-ray coverage. If the true bolometric driver were smoother than the BAT curve, the disc model might not fail as badly, and the recovered IRF could be a distortion of the wrong input. I think this is a real limitation, not a fatal one — the predicted fast variability is so strong that you'd need an implausibly smooth driver to rescue the disc — but the paper should have at least checked for spectral index variations or used NuSTAR data if available. Second, the IRF recovery in Section 4 is phenomenological, with no error bars on the recovered timescales or component fractions. The double-Gaussian is motivated by the CCF shape, but the improvement over a single Gaussian is not quantified in the final fits, and the exponentially modified Gaussian adds parameters for modest gain. The claimed 1.5–20 light-day range is consistent with BLR sizes, but without uncertainties I don't know how tightly to take it.\n\nThe paper includes a useful control in Appendix B: the same IRF recovery fails on NGC 5548 with the XRT band, which supports the idea that the BAT band is doing something real. It also discusses the apparent contradiction with line reverberation lags and gives a plausible resolution.\n\nNet: this deserves a serious referee. I would recommend acceptance after the authors either validate or more carefully caveat the bandpass assumption, and add some error assessment to the IRF. I'd cite it if I were working on AGN variability.","headline":"A strong negative result on inner-disc reprocessing in NGC 4151, with a plausible but undertested BLR-scale interpretation.","tokens_in":23146,"tokens_out":5362,"would_cite":true,"duration_ms":50006,"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 paper claims that the ultraviolet light of NGC 4151 is produced by X-ray reprocessing in broad-line-region gas, not by an inner accretion disc, which had vanished into a hot flow.","keywords":["accretion discs","active galactic nuclei","reverberation mapping","X-ray reprocessing","broad line region","changing-state AGN","NGC 4151","spectral timing"],"falsifier":"Higher-cadence, simultaneous monitoring with a hard-X-ray instrument covering above 50 keV would settle the picture: if the UVW1 versus greater-than-50 keV cross-correlation shows a peak at lags below 1.5 days, the claimed absence of inner optically thick material is wrong. A high signal-to-noise X-ray spectrum showing a relativistically broadened iron line from within roughly 1.5 light-days would also directly contradict the truncated-flow geometry.","tokens_in":22041,"feed_emoji":"🛰️","tokens_out":9087,"duration_ms":79774,"temperature":0.7,"pith_summary":"This paper asks where the ultraviolet emission of the active galaxy NGC 4151 is produced during a faint, changing-state phase. Building a spectral-timing model with an outer disc, a warm Comptonising layer, and a hot X-ray corona, the authors fit the time-averaged spectrum and then use the observed 15-50 keV light curve to predict how the near-ultraviolet should vary if the UV is simply reprocessed X-rays from optically thick material within a few hundred gravitational radii. That prediction fails badly: the model UV is far too fast and too strongly correlated with the X-rays. The data are instead matched by a linear reprocessing response spread over 1.5-20 light-days, the size scale of the broad line region, and the authors conclude that NGC 4151's inner thin disc had effectively been replaced by a hot, radiatively inefficient flow during the campaign, at a luminosity near 1.4 percent Eddington. A sympathetic reader cares because this is the first time the full bolometric X-ray driver, rather than a soft X-ray proxy, has been used to test the standard reprocessing picture of AGN optical and UV continuum.","feed_headline":"NGC 4151's UV comes from gas 1.5-20 light-days out, not the disc","feed_subtitle":"If right, the inner disc had evaporated into a hot flow at 1.4 percent Eddington: a changing-state AGN.","key_machinery":"The central machinery is the agnsed spectral-timing model: it assumes Novikov-Thorne gravitational dissipation at each radius, but thermalises it to a blackbody only outside the warm-Compton radius, releases it in a warm Comptonising layer at intermediate radii, and puts the innermost radii into a hot corona, so the UV in the canonical fit originates at 90-390 $R_g$. The companion tool is an impulse-response decomposition in which the UVW1 light curve is a constant, plus a slow component formed by convolving the BAT light curve with a free impulse-response function, plus a small fast component. A double-Gaussian and then an exponentially modified double-Gaussian impulse response localise the slow component to 1.5-20 light-days, and the light-travel-time argument against any alternative at less than 1.5 light-days is what rules out the disc.","core_discovery":"The central discovery is that the near-ultraviolet (UVW1) light curve of NGC 4151 cannot be powered by optically thick accretion-disc material inside roughly 1.5 light-days, and can instead be powered by X-ray reprocessing in broad-line-region gas. The canonical spectral fit places the UV source in a warm Comptonising region spanning 90-390 gravitational radii from the black hole, but when the Swift BAT hard X-ray light curve is convolved through the light-travel-time impulse response of that geometry, the predicted UV variability has far too much power on sub-day timescales and too high a correlation with the X-rays. Replacing the physical model with a free impulse-response function gives a good fit with 90 percent of the UV flux responding on lags of 1.5-20 days, matching known BLR size scales, and only 5-10 percent responding fast. The paper's conclusion is that the inner disc was absent during this epoch, replaced by an optically thin hot flow, and that the UV continuum is dominated by diffuse emission from dense BLR gas, with the inferred luminosity of 1.4 percent Eddington placing the source in the changing-state regime.","pith_inferences":["A direct test of the paper's logic: monitor a sample of AGN across the Eddington-fraction range 0.01-0.03 with simultaneous UV and greater-than-10 keV X-ray coverage and check whether fast (less than 1 day) UV response appears as the Eddington fraction rises, as expected if the thin disc reforms.","If BLR diffuse continuum dominates the UV, the recovered impulse response should correlate with the known He II, H beta and H alpha lags of roughly 2-3, 6 and 11 days; a longer campaign could measure the UV lag structure and test this.","The same analysis applied to other low-state AGN, including a re-analysis of NGC 5548 with a 15-50 keV driver, could show whether a truncated flow plus BLR reprocessing is generic rather than unique to NGC 4151."],"forward_implications":["The standard irradiated-disc plus warm-Compton geometry cannot produce the observed UVW1 light curve of NGC 4151, even when the soft-excess region is removed or the corona is made vertically extended.","A linear impulse response with delays of 1.5-20 days reproduces the UVW1 light curve, meaning the slow UV variability is consistent with reprocessing by dense gas in the broad line region.","Only about 5-10 percent of the UVW1 flux can be fast-correlated with the hard X-rays, leaving no room for a significant disc or warm-Compton contribution inside roughly 1.5 light-days.","The inner accretion flow in NGC 4151 at this epoch is inferred to be radiatively inefficient, consistent with its low Eddington fraction and with the absence of a relativistically broadened iron line."],"supporting_citations":[{"why":"Supplies the 69-day multiwavelength campaign: Swift UVOT, XRT and BAT light curves and the UV-X-ray cross-correlation that the model must reproduce.","marker":"E17"},{"why":"Provides the spectral-timing prediction method and the impulse-response geometry used to model disc reprocessing; its NGC 5548 failure is what this paper extends with a bolometric driver.","marker":"GD17"},{"why":"First proposed that diffuse continuum emission from broad-line-region gas contributes to the optical/UV, the physical mechanism invoked to explain the slow response.","marker":"Korista & Goad 2001"},{"why":"Quantifies the diffuse recombination and free-free continuum from dense BLR clouds, showing that such reprocessing can produce UV at the recovered size scale.","marker":"Lawther et al. 2018"},{"why":"Provides the warm Comptonisation sandwich geometry with energy-balance-fixed spectral index Gamma = 2.7 that underlies the time-averaged SED fit.","marker":"Petrucci et al. 2018"},{"why":"Establishes the changing-state transition at low Eddington fraction used to interpret NGC 4151's 1.4 percent Eddington luminosity.","marker":"Noda & Done 2018"},{"why":"Characterises the complex neutral absorption and reflection that must be included before the agnsed continuum can be fit to the X-ray spectrum.","marker":"Beuchert et al. 2017"},{"why":"Independent X-ray spectral evidence for no relativistic iron line, supporting the inference that the inner disc is absent.","marker":"Zoghbi, Miller & Cackett 2019"}],"fun_headline_variants":["NGC 4151's UV arises from BLR gas, not the disc","Inner disc gone: UV from gas 1.5-20 light-days out","Changing-state AGN: UV reprocessed in BLR, not disc","Discarded disc: NGC 4151's UV from broad-line region"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes the Swift BAT 15-50 keV light curve tracks the true bolometric X-ray luminosity that drives reprocessing; if the flux above 50 keV varied independently, both the failure of the disc model and the recovered 1.5-20 day response would be artifacts.","fun_headline_variants_meta":{"raw":{"variants":["NGC 4151's UV arises from BLR gas, not the disc","Inner disc gone: UV from gas 1.5-20 light-days out","Changing-state AGN: UV reprocessed in BLR, not disc","Discarded disc: NGC 4151's UV from broad-line region"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000245,"raw_usage":{"total_tokens":1632,"prompt_tokens":1137,"completion_tokens":495,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":753,"completion_tokens_details":{"reasoning_tokens":420}},"tokens_in":753,"tokens_out":495,"duration_ms":4066,"temperature":1.0,"reasoning_tokens":420,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:13:35.623293+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Higher-cadence, simultaneous monitoring with a hard-X-ray instrument covering above 50 keV would settle the picture: if the UVW1 versus greater-than-50 keV cross-correlation shows a peak at lags below 1.5 days, the claimed absence of inner optically thick material is wrong. A high signal-to-noise X-ray spectrum showing a relativistically broadened iron line from within roughly 1.5 light-days would also directly contradict the truncated-flow geometry.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"First proposed that diffuse continuum emission from broad-line-region gas contributes to the optical/UV, the physical mechanism invoked to explain the slow response."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Quantifies the diffuse recombination and free-free continuum from dense BLR clouds, showing that such reprocessing can produce UV at the recovered size scale."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the warm Comptonisation sandwich geometry with energy-balance-fixed spectral index Gamma = 2.7 that underlies the time-averaged SED fit."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the changing-state transition at low Eddington fraction used to interpret NGC 4151's 1.4 percent Eddington luminosity."},{"cited_title":"et al., 2017, A&A, 603, A50","cited_arxiv_id":null,"evidence_quote":"Characterises the complex neutral absorption and reflection that must be included before the agnsed continuum can be fit to the X-ray spectrum."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Independent X-ray spectral evidence for no relativistic iron line, supporting the inference that the inner disc is absent."}],"review_version":1}