{"id":"1f291732-57ab-46af-84a9-0af44cb76703","arxiv_id":"2608.11300","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"The tidal disruption event ATLAS22kjn shows a resolved 9-day pre-peak bump, coronal lines that imply a stable ~1e6 K ionising source, and a dust echo whose covering fraction is much higher than typical TDEs.","lead":"Astronomers watched a supermassive black hole tear apart a star, catching a brief early flash, glowing high-energy gas lines, and a delayed infrared pulse. The event, ATLAS22kjn, reveals what a black hole's neighborhood looks like when we cannot see the X-rays directly.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed ≲10% cooling of the ionising source is not established by a successful model: Section 4's curves are not fitted, and the only quantitative attempt (Appendix C) did not converge and is consistent with no temperature change.","rationale":"The reader's weakest_assumption correctly identifies the photoionisation/same-source/blackbody assumptions as load-bearing. I agree with that, and I sharpen it: the central quantitative claim — a ≲10% temperature decrease — is not actually derived from a successful model. Section 4 makes only an eyeball comparison, and the one quantitative attempt (Appendix C) fails to converge and gives a temperature slope consistent with zero. This is a concrete gap between the data and the headline claim, independent of whether the broader assumptions hold. The observational core of the paper is strong: public multi-wavelength data, honest caveats, and a clear statement that the toy model failed. No red flag warrants rejection; the appropriate remedy is to soften the abstract and conclusions to frame the ≲10% decrease as an upper limit, or to support it with a converged photoionisation fit. My concern therefore does not change the reader's CONDITIONAL verdict; it reinforces the reason for it. Agreement is 'partial' because the reader's stated weakest assumption is the same-source/photoionisation premise, while my emphasis is on the additional unsupported mapping from line slopes to temperature, which the paper's own Appendix C demonstrates is not yet established.","tokens_in":58434,"tokens_out":8403,"duration_ms":81851,"concrete_test":"Run a grid of Cloudy photoionisation models with a blackbody SED at T0 ~ 1.2e6 K, gas densities n_H ~ 1e6–1e8 cm^-3, and a range of ionisation parameters, and compute the predicted luminosity evolution of [Ne V], [Fe VII], [Fe X], [Fe XI], [Fe XIV], [S XII], and [Ar XIV] for temperature changes of 0%, ±5%, and ±10% over 500 days, with the source radius either fixed or varying by ±10%. Compare the predicted slopes as a function of ionisation potential to the observed slopes in Figure 13, accounting for the full covariance of the line fits. If the best-fit temperature change is consistent with zero and the 95% upper limit exceeds 10%, the abstract and conclusions must be revised to report an upper limit rather than a detected decrease.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 4 infers a ≲10% temperature decrease of the ionising source from the IP-dependent slopes of CL luminosity evolution. This inference rests on the stated assumptions (photoionisation, same source as X-rays, blackbody SED), but even granting those, the quantitative claim is unsupported. The comparison in Figure 13 overlays observed slopes with curves for ±5% and ±10% temperature changes without any fit or goodness-of-fit statistic; the text says the data 'could be explained by' such a change, which is an upper limit, not a detection. The paper's own attempt to fit this model (Appendix C, Eq. C1–C3) failed: MCMC runs did not converge due to degeneracies between the line-strength parameters f_l and the temperature, and the linear-fit temperature slope s_T was 'of order 10 K/day (0.005 Tbar/500 days) and negative, and consistent with zero to within 1σ'. Thus the data are fully consistent with no temperature evolution, and the abstract's 'finding a decrease of ≲10% over 500 days' overstates what Section 4 establishes. Additionally, the toy model assumes L_l,s ∝ f_l R_s^2 ∫_{E_l}^∞ 2πB_E(E,T_s)/E dE, ignoring how line emissivity depends on ionisation parameter, density, column density, and the spatially stratified CL region (Figure 18). Mummery et al. (2025) required Cloudy to model these lines and found [Fe XIV] generally faint, whereas here it is the strongest CL, so the simplified mapping is not validated. The obscuration conclusion also inherits the same-source assumption, but the temperature claim is the most fragile because it additionally requires this unvalidated mapping.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents a multi-wavelength analysis of the nuclear transient ATLAS22kjn (AT 2022fpx), arguing that it is a tidal disruption event in a gas-rich nucleus. The evidence includes a blue continuum, broad Balmer and He II lines, strong coronal lines with luminosities comparable to [O III], delayed soft X-ray emission, a pre-peak UV/optical bump resolved in TESS, and a mid-infrared dust echo. The authors use the evolution of coronal-line luminosities as a function of ionisation potential to infer that the ionising source cooled by less than about 10% over 500 days, and use line widths to place the coronal-line region between the broad-line region and the MIR-emitting dust. The bump is attributed most plausibly to stream-stream collision or wind-stream collision. The paper also derives a dust covering fraction of about 0.40 and constructs a spatial map of the nuclear emission regions.","tokens_in":58800,"tokens_out":5055,"duration_ms":56181,"significance":"If the central conclusions hold, this is a valuable addition to the small sample of coronal-line-emitting TDEs. The dataset is unusually rich, combining ATLAS, ASAS-SN, ZTF, TESS, Swift, XMM-Newton, and optical/NIR spectroscopy, and the paper carefully documents its analysis choices. The early bump measurement, the MIR dust echo, and the comparison with theoretical bump mechanisms are likely robust and will be useful benchmarks. The paper is also transparent: the assumptions behind the coronal-line temperature inference are stated explicitly, and the non-convergent MCMC alternative is reported honestly. The main weakness is that the headline temperature constraint is presented as a measurement even though the data are consistent with no temperature change; this needs to be reframed before publication.","major_comments":[{"comment":"The claimed 'decrease of less than 10% over 500 days' is not supported as a measurement. The curves in Figure 13 are not fitted to the data; no goodness-of-fit statistic or parameter uncertainty is reported, and the text says only that the data 'could be explained by' such a change. Appendix C's direct linear fit gives a temperature slope s_T that is consistent with zero at 1-sigma (Eq. C2-C3), so the data are fully consistent with no temperature evolution. The abstract and Section 8 ('suggestive of a 10% decrease') therefore overstate what the analysis establishes. Please reframe this result as an upper limit or consistency check, or provide a genuine fit with uncertainties.","section":"Section 4, Figure 13, Abstract"},{"comment":"The toy model L_{l,s} = f_l R_s^2 times the integral of the Planck function over photon energies above the line ionisation potential omits the dependence of each line's emissivity on ionisation parameter, density, column density, and the stratified geometry of the coronal-line region. The authors acknowledge the assumptions of photoionisation, a common source for the ionising and X-ray emission, and a blackbody SED, but these assumptions are load-bearing: Mummery et al. (2025) required Cloudy to model these lines and found [Fe XIV] generally faint, whereas here [Fe XIV] is the strongest CL. If the coronal lines are powered by shocks or by a separate EUV source, the inferred temperature evolution and the interpretation of the X-ray non-detections as obscuration would not follow. The paper should either validate the toy model against a photoionisation calculation or explicitly demote the temperature and obscuration claims to conditional interpretations.","section":"Section 4, Appendix C, Eq. C1"},{"comment":"The spatial ordering of emission regions uses virial estimates r ~ G M_BH / FWHM^2 for the broad lines and coronal lines. The assumption that the CL widths are virial is stated but not tested; if the CL gas is outflowing or turbulent, the inferred radii and the conclusion that the CL region lies between the broad-line region and the MIR-emitting dust are not robust. Please add a discussion of non-virial alternatives and how they would change Figure 17 and the associated text.","section":"Section 7, Figure 17"}],"minor_comments":[{"comment":"The sentence 'we mangled all our spectra to match multiple bands of follow-up photometry' appears to contain a typo; 'mangled' should likely be 'matched'.","section":"Section 2.5"},{"comment":"The sentence 'We are unable to make sure the X-ray to NIR line ratio dignostics in Lamperti et al. (2017)' should read 'We are unable to use the X-ray to NIR line ratio diagnostics of Lamperti et al. (2017)', and 'dignostics' should be 'diagnostics'.","section":"Section 3.6"},{"comment":"In the concluding paragraph, 'as well as the to sample of CLEs' should be 'as well as to the sample of CLEs'.","section":"Section 8"},{"comment":"The bump duration and rise time depend on the chosen GP length scale ell = 4 days and the 20% threshold, but no sensitivity to the length scale is reported; a brief exploration of how the quoted duration changes with ell would strengthen the measurement.","section":"Section 3.3"},{"comment":"The sentence 'The slow rise of ATLAS22kjn to its main peak is also seen in the simulations of (Calderon et al. 2024)' should read 'in the simulations of Calderon et al. (2024)' without the leading 'of' or with a preceding 'by'.","section":"Section 5.6"}],"recommendation":"major_revision","confidential_remarks":"The paper is unusually transparent about its limitations, and the dataset is strong enough that the core observational results are likely publishable. The main issue is that the abstract and conclusions promote a temperature constraint that the paper's own analysis shows is not a detection; this framing should be corrected. I see no concerns about novelty or citation practices, and the use of Hinkle (2024) for the MIR method is appropriate because it is applied to new data."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nTwo things to know. First, this is a genuinely rich observational dataset: TESS resolves both rise and decline of the early bump, the coronal-line time series spans over 1000 days with ionisation-potential-dependent slopes, and the MIR dust echo gives a covering fraction of 0.40 ± 0.03, one to two orders of magnitude above typical TDEs. Second, the headline claim that the ionising source cooled by ≲10% over 500 days is not established. Figure 13 overlays no-fit curves, and the paper's own MCMC toy model in Appendix C did not converge and gave a temperature slope consistent with zero. The abstract overstates what Section 4 shows; the text itself says the data 'could be explained by' a temperature change, which is an upper limit, not a detection.\n\nWhat is actually new: the TESS-resolved bump, the CL luminosity evolution as a function of ionisation potential, and the radial ordering of emission regions. Koljonen et al. and Lin et al. already reported the object, but these additions are real. The photometric and spectroscopic reduction is careful, the paper is honest about the unphysical X-ray blackbody radii, and the dust echo analysis follows a published method cleanly.\n\nThe soft spots, in proportion. The temperature inference rests on three unverified assumptions: photoionisation, the same source for X-rays and CLs, and a blackbody SED. Even granting all three, the data are consistent with no temperature change, so the abstract and conclusion should be softened. The virial assumption for CL radii in Section 7 is load-bearing for the spatial map but not unreasonable. Minor: the GP length scale and the 20% bump-duration threshold are hand-chosen, and the bump-origin discussion is necessarily speculative.\n\nOverall, as an observational case study this is solid and useful. The classification as a CLE TDE is well supported, and the dust covering fraction is a clean result. I'd send it to peer review with a request to soften the temperature claim. The overclaim does not sink the paper; it just needs revision.","headline":"Solid observational case study of a CLE TDE with a real early bump and dust echo, but the headline '≲10% cooling' claim is not backed by a fit and should be softened.","tokens_in":59444,"tokens_out":3086,"would_cite":true,"duration_ms":26574,"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":"Coronal lines in the tidal disruption event ATLAS22kjn show its EUV/soft X-ray source was obscured early and late and cooled by less than about 10 percent over 500 days; its 9-day pre-peak bump favours stream-stream or wind-stream…","keywords":["tidal disruption events","coronal lines","AT 2022fpx","supermassive black holes","accretion","dust echo","pre-peak light curve bump","EUV/soft X-ray emission"],"falsifier":"Take a deep X-ray exposure of ATLAS22kjn at a late epoch while coronal lines are still detected: the paper's picture predicts the ionising source is still active but hidden, so a detection of soft X-rays at the level implied by the line luminosities would falsify the obscuration claim, while a continued non-detection at that depth would support it.","tokens_in":58264,"feed_emoji":"🔭","tokens_out":19884,"duration_ms":167805,"temperature":0.7,"pith_summary":"ATLAS22kjn is a tidal disruption event—a star shredded by a supermassive black hole—whose high-ionisation coronal lines let astronomers see the otherwise invisible extreme-ultraviolet and soft X-ray emission. The paper argues that these lines are present before the event's X-rays are detected and persist after the X-rays drop out, so the ionising engine is obscured by nuclear material at early and late times rather than simply switching off. Using the way each coronal line fades as a function of its ionisation potential, the authors conclude that the ionising source cooled by $\\lesssim 10\\%$ over 500 days. The same multi-wavelength dataset resolves a distinct 9-day bump in the ultraviolet/optical rise, which matches theoretical predictions for a collision between debris streams or between a wind and the debris stream. If correct, the result establishes coronal lines as a practical thermometer and calorimeter for the EUV/soft X-ray radiation that tidal disruption events otherwise hide.","feed_headline":"Coronal lines expose a hidden X-ray source in a shredded-star flare","feed_subtitle":"The flare's coronal lines show its X-ray engine was obscured and cooled less than 10% in 500 days.","key_machinery":"The central tool is the coronal-line region itself: forbidden emission lines from ions such as [Fe VII], [Fe X], [Fe XI], [Fe XIV], [Ne V], [Ar XIV], and [S XII], whose ionisation potentials lie between roughly 100 and 700 eV. Because photons at those energies are absorbed before reaching the observer, the lines act as a reprocessed thermometer for the extreme-ultraviolet and soft X-ray continuum. The argument compares the measured luminosity slope of each line (in dex per 500 days) against ionisation potential with predicted curves for a blackbody source whose temperature changes by $\\pm5\\%$ and $\\pm10\\%$ around the weighted X-ray temperature, separating temperature changes from radius changes. For the bump, the machinery is a Gaussian-process fit to high-cadence space-based photometry after subtracting a curved power-law model of the main rise; that fit yields the bump's duration, peak luminosity, integrated energy, blackbody radius, and implied outflow velocity.","core_discovery":"In its own terms, the paper establishes that ATLAS22kjn is a coronal-line-emitting tidal disruption event in a gas-rich, dusty nucleus. The coronal lines appear before the ultraviolet/optical peak, precede the first X-ray detection by roughly 300 days, and survive after X-rays are no longer detected; the paper reads this as evidence that the same ionising source is hidden by obscuring material at both early and late times. From the luminosity evolution of lines spanning ionisation potentials from about 97 eV ([Ne V]) to 686 eV ([Ar XIV]), it finds that higher-ionisation lines fade faster than lower-ionisation lines, matching a $\\lesssim 10\\%$ decrease in the temperature of a blackbody ionising source whose weighted X-ray temperature is $\\bar{T}_X \\simeq 1.2\\times10^6$ K. The ultraviolet/optical bump peaks $125^{+5}_{-3}$ rest-frame days before the main flare, lasts $9^{+4}_{-2}$ days, and reaches a peak luminosity of $(3.0^{+0.2}_{-0.4})\\times10^{42}$ erg s$^{-1}$; its properties are most consistent with a stream-stream or wind-stream collision. The mid-infrared dust echo has a rest-frame lag of about 200 days and a covering fraction of $f_c = 0.40\\pm0.03$, placing the coronal-line region between the gas producing broad Balmer lines and the dusty torus.","pith_inferences":["If the photoionisation interpretation holds, the same ionisation-potential-slope method could be applied to other coronal-line events to build a census of EUV/soft X-ray temperatures for tidal disruption events that are invisible to X-ray telescopes.","The bump was fully resolved only after high-cadence space-based photometry was added, suggesting that similar precursor features in earlier transients may have gone unnoticed; all-sky high-cadence ultraviolet/optical surveys should reveal whether such bumps are common.","The high dust covering fraction implies coronal-line-emitting tidal disruption events preferentially occur in gas-rich, dusty nuclei; accounting for this selection effect could revise optical TDE rate estimates.","ATLAS22kjn is the only coronal-line emitter in the paper's comparison sample showing [Ar XIV], which may encode information about the hardness of the ionising SED and could be tested with dedicated photoionisation modelling of non-iron species."],"forward_implications":["Coronal-line monitoring can recover the temperature history of a tidal disruption event's EUV/soft X-ray source even when the source itself is not directly detectable in X-rays.","X-ray non-detections that bracket a coronal-line-bright phase point to persistent obscuration by nuclear gas or dust rather than delayed disk formation or a single expanding reprocessing outflow.","The bump's 9-day duration and $(3.0^{+0.2}_{-0.4})\\times10^{42}$ erg s$^{-1}$ luminosity favour stream-stream or wind-stream collisions, and disfavour nozzle shocks and cooling unbound debris as the dominant mechanism.","The approximately 200-day infrared lag and dust covering fraction of $0.40\\pm0.03$ place the coronal-line region between the broad-line region and the dusty torus, giving a radial map of the sub-parsec black-hole environment.","The faster fading of higher-ionisation lines is a quantitative sign that the ionising source remained nearly stable in temperature, cooling by $\\lesssim 10\\%$ over 500 days."],"supporting_citations":[{"why":"Identifies coronal-line emitters as objects whose CL luminosities rival [O III], establishing the photoionisation context for tidal disruption events.","marker":"Komossa et al. (2008)"},{"why":"Supplies the comparison sample of coronal-line-emitting TDEs and their mid-infrared dust covering fractions.","marker":"Hinkle et al. (2024)"},{"why":"Provides the photoionisation simulations of coronal-line emission in TDEs used to interpret line radii and luminosity trends.","marker":"Mummery et al. (2025)"},{"why":"Earlier study of ATLAS22kjn that first identified the precursor feature and reported X-ray and polarization behaviour.","marker":"Koljonen et al. (2024)"},{"why":"Earlier study of ATLAS22kjn's X-ray, mid-infrared, dust, and spectral properties, which this analysis extends and compares with.","marker":"Lin et al. (2025)"},{"why":"Compiles TDEs with pre-peak bumps and links them to stream collisions, giving the comparative framework for the bump.","marker":"Wang et al. (2024)"},{"why":"Simulates wind-stream collisions and predicts a short precursor, one of the two mechanisms the bump is most consistent with.","marker":"Calderón et al. (2024)"},{"why":"Computes stream-stream collision luminosities and efficiencies that support the bump's energy budget.","marker":"Huang et al. (2023b)"},{"why":"Provides the host-subtraction and covering-fraction method used to measure the mid-infrared dust echo.","marker":"Hinkle (2024)"}],"fun_headline_variants":["Coronal lines unveil a hidden X-ray source in a TDE","TDE's coronal lines outlast its X-rays, hinting at obscuration","Early bump and dust echo mark a coronal-line TDE","Coronal lines expose a cool, obscured X-ray engine in a shredded star","Obscured X-rays, early bump, dust echo: a coronal TDE"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the coronal lines are powered by photoionisation from the same blackbody source that produces the X-rays, while the spatial map additionally assumes the line-emitting gas is virialised.","fun_headline_variants_meta":{"raw":{"variants":["Coronal lines unveil a hidden X-ray source in a TDE","TDE's coronal lines outlast its X-rays, hinting at obscuration","Early bump and dust echo mark a coronal-line TDE","Coronal lines expose a cool, obscured X-ray engine in a shredded star","Obscured X-rays, early bump, dust echo: a coronal TDE"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000235,"raw_usage":{"total_tokens":1654,"prompt_tokens":1253,"completion_tokens":401,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":869,"completion_tokens_details":{"reasoning_tokens":300}},"tokens_in":869,"tokens_out":401,"duration_ms":3645,"temperature":1.0,"reasoning_tokens":300,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:12:59.886756+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a deep X-ray exposure of ATLAS22kjn at a late epoch while coronal lines are still detected: the paper's picture predicts the ionising source is still active but hidden, so a detection of soft X-rays at the level implied by the line luminosities would falsify the obscuration claim, while a continued non-detection at that depth would support it.","supporting_citations":[],"review_version":1}