{"id":"5fc96649-9d23-4f0d-94f6-40d285f9844f","arxiv_id":"2412.12549","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"A two-flare model based on repeated partial disruptions of the same star reproduces the UV/optical light curves of AT 2022dbl and AT 2023adr and forecasts their third outbursts.","lead":"This paper builds a light curve model for tidal disruption events that brighten twice, under the idea that one star is partially torn apart by the same black hole on two separate close passes. It fits two real events and predicts when each star should flare a third time, giving observers a concrete date to test the model.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The model's monotonically dimming second flare rests on the unvalidated assumption that a stripped remnant immediately returns to a ZAMS radius; if tidal heating expands the remnant, the fitted b0, mass, and third-flare forecasts are unreliable.","rationale":"The paper develops a composite two-PTDE light-curve model and applies it to two re-brightening TDEs, with explicit third-flare predictions. The central claim is that a single star undergoing two partial disruptions can reproduce the observed UV/optical light curves, so the predicted third-flare times and magnitudes are reliable. The most load-bearing condition is the post-disruption stellar radius and the unchanged pericenter, because these determine the relative brightness of the two flares and the properties of the third flare. The reader's weakest-assumption analysis identifies exactly this: the ZAMS radius resets after each stripping, forcing later flares to be dimmer. I agree with that assessment. The manuscript itself states in Section 4.1 that tidal energy can cause expansion and brighter subsequent flares, and that the model is not appropriate for brighter repeaters. This is not an external or fringe concern; it is a known physical effect whose magnitude for the present targets is unquantified. A concrete MESA or hydrodynamic test of the remnant radius evolution would settle whether the assumption holds for these orbital periods and stripped masses. The fit-quality issues noted by the reader (reduced chi-square of 14 for AT 2022dbl, excluded data segments, and near-zero parameter uncertainties) are real and reinforce the need for caution, but they are secondary to the physics of the radius assumption: even a perfect statistical fit would not validate the forecast if the radial response is wrong. The forecast is genuinely falsifiable and is scheduled, which is credit to the paper. My recommended verdict remains CONDITIONAL: the model and the prediction are worth publishing, but the authors should either demonstrate that the ZAMS-radius assumption is appropriate for these targets or explicitly present the forecast as contingent on that assumption, alongside the requested code/data release.","tokens_in":22620,"tokens_out":7394,"duration_ms":67755,"concrete_test":"Compute the remnant radius of a 1 solar-mass ZAMS star after stripping ~0.1-0.2 solar masses at the relevant beta using a stellar evolution code (e.g., MESA) with tidal heating deposited according to Chen et al. (2024) or Liu et al. (2024a), then evolve for one orbital period (~680 days for AT 2022dbl). If the radius at the next pericenter exceeds the ZAMS radius at the same mass by more than ~10%, the Section 2.4 assumption fails and the fits and Table 3 forecasts must be redone with a self-consistent radius. A faster statistical check: refit both light curves allowing a free multiplicative factor f on the second-flare stellar radius (r*,1 = f * r_ZAMS) and examine whether f is constrained away from 1 and whether the third-flare predictions shift significantly.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central construction in Section 2.4 assumes r*,1 equals the ZAMS radius at the reduced mass (Tout et al. 1996) and rp,1 = rp,0 after the first partial disruption. Combined with the stripped-mass calculation, this forces rt and beta to decrease after each encounter, so every subsequent flare is dimmer. That monotonic dimming is not a derived result; it is imposed by the radius assumption. Section 4.1 explicitly acknowledges that tidal energy injection can make the remnant expand, producing brighter second flares, and that the model is 'clearly not appropriate' for objects with brighter second flares. For AT 2022dbl and AT 2023adr, the observed second flare is dimmer, so a smaller fitted b0 could compensate for an actual expansion; the best-fit parameters in Table 2 and the predicted third-flare magnitudes in Table 3 are therefore not uniquely tied to the physics. The zero-width uncertainties reported for b0 and Porb (e.g., 0.63+0.00-0.00) hide this degeneracy rather than resolve it. Consequently, the claim that the light curves are 'well fitted' does not by itself validate the radial-response assumption on which the forecast rests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a light-curve model for re-brightening tidal disruption events in which a single star undergoes two partial tidal disruptions on an eccentric orbit. The mass fallback rate for each encounter is built from the GRR2013 templates, shifted in energy for e<1 following Hayasaki et al. (2013), with the remnant's orbital energy updated after each encounter using the Chen et al. (2024) fitting formulae. The model is applied to AT 2022dbl and AT 2023adr, and the author reports that both light curves are well fitted and then forecasts the time and peak brightness of the third flare for each source (Table 3). The central claim is that the two-flare composite model reproduces the observations and that the third-flare predictions are reliable enough for observers to prepare follow-up observations.","tokens_in":22915,"tokens_out":5913,"duration_ms":55320,"significance":"The paper addresses a genuine gap: existing public TDE fitters assume a single parabolic encounter and cannot model two flares from the same star. The construction is largely transparent, using independently published simulation templates for fallback, eccentric-orbit energy shifts, and orbital-energy changes, and the third-flare forecast is a genuinely testable prediction whose time and brightness were not fed back into the fit. A working fitting code of this kind would be useful to the time-domain community, and the forecasts for AT 2022dbl and AT 2023adr are falsifiable within a few years. However, the current evidence for the central claim is mixed: one of the two fits has a reduced chi-square of 14, several data segments are removed post-hoc, and the forecast rests on a remnant-radius assumption that the paper itself identifies as potentially incorrect. If the caveats are addressed, the model would be a valuable contribution; in its present form the claims outrun the validation.","major_comments":[{"comment":"The reduced chi-square for AT 2022dbl is reported as chi²_red = 14.0, which is not consistent with the abstract's statement that the light curves are 'well fitted.' Please report the number of degrees of freedom, the chi-square contribution of each excluded data segment, and a fit-quality statistic that accounts for the rejected points; alternatively, soften the claim to reflect that the model captures the overall morphology but not the detailed photometry.","section":"Section 3, Table 2"},{"comment":"The exclusion of the MJD 59800-60300 segment for AT 2022dbl and of the first two g-band points for AT 2023adr is decided after inspecting residuals, and the paper states that including the AT 2022dbl points changes the black hole mass by about 0.2 dex. Because the fit and the forecast are conditioned on these post-hoc cuts, please show that the best-fit parameters in Table 2 and the third-flare predictions in Table 3 are robust to including those points with an alternative radiation treatment or an objective outlier criterion.","section":"Section 3, Figure 4"},{"comment":"The model imposes monotonic dimming by setting r*,1 to the ZAMS radius at the reduced mass (Tout et al. 1996) and by taking rp,1 = rp,0, so beta and the stripped mass necessarily decrease after each encounter. As Section 4.1 acknowledges, tidal energy injection can expand the remnant and make subsequent flares brighter, and the paper explicitly states that the model is 'clearly not appropriate' for brighter-second-flare sources. The fitted b0, stellar mass, and the Table 3 forecasts are therefore contingent on an unvalidated radial-response assumption; a concrete test would be to compare the assumed ZAMS radius evolution with the remnant radii from hydrodynamic or stellar-evolution calculations at the fitted parameters, or to introduce a free expansion parameter and examine how the forecast changes.","section":"Section 2.4, Section 4.1"},{"comment":"The same radiation parameters (eta, Rph0, l, t_nu) are used for both flares with no physical justification, as the author states directly in Section 4.3. For AT 2023adr, the discussion admits that stream-disk collision may dominate the emission and that a common eta may be inappropriate, which would change b0 and make the third flare fainter than predicted. Please quantify the sensitivity of the fitted parameters and of the Table 3 forecasts to fitting the two flares with independent radiation parameters or to including the stream-disk efficiency estimate in the likelihood.","section":"Section 4.3"},{"comment":"The reported posterior for b0 in AT 2022dbl is 0.63+0.00-0.00, with Porb,0 = 680.44+0.16-0.17; the zero-width b0 uncertainty suggests that the MCMC chains are not resolving the posterior, likely because of a prior boundary or a degenerate direction, rather than a genuine 1-sigma constraint. Please provide corner plots for all fitted parameters and convergence diagnostics; as presented, the third-flare forecast inherits an artificially precise b0.","section":"Table 2"}],"minor_comments":[{"comment":"The interpolation of Delta_epsilon_orb between the gamma=4/3 and gamma=5/3 fitting formulae of Chen et al. (2024) is an ad hoc extension to hybrid stars; please state explicitly that this linear interpolation is an assumption rather than a result of the cited simulations.","section":"Equation (12)"},{"comment":"The composite magnitude in Equation (14) assumes that the two flares evolve independently as blackbodies and that their fluxes add linearly; this is discussed in Section 4.3, but the assumption should be flagged at the point of use.","section":"Equation (14)"},{"comment":"The statement that the eccentricity condition e > e5% is 'always satisfied' in the explored parameter space should be backed with the actual fitted eccentricities for the two targets, rather than only a typical example.","section":"Section 2.2"},{"comment":"Several quoted uncertainties, such as 18.1+0.0-0.0 for AT 2022dbl, are unrealistically small after rounding; please report the posterior widths with enough significant figures to be meaningful, or use a different summary statistic.","section":"Table 3"},{"comment":"The corner plots show only four parameters; because the model has ten free parameters, please show the full posterior or state explicitly that the remaining parameters are marginalized out.","section":"Figure 3"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Shiyan Zhong has put together a workable two-flare fitting model for re-brightening TDEs, built from published ingredients: the GRR13 fallback templates, the Hayasaki et al. eccentricity shift, and the Chen et al. orbital-energy update. The genuinely new pieces are the simultaneous two-flare composite and the concrete forecasts (third flare for AT 2022dbl near MJD 61061 and AT 2023adr near MJD 60788) that observers can check in 2025-2026. That is real and useful; existing tools like MOSFiT or TiDE cannot do this.\n\nThe application to two objects mostly works. For AT 2023adr the reduced chi-square is 1.3, which is fine. For AT 2022dbl it is 14, so the abstract's claim that both light curves are 'well fitted' oversells the first one. The author excluded a chunk of UV data and two early g points after looking at the residuals; the exclusions may be justifiable, but they should be reported as data-tailored rather than as a planned analysis. The zero-width uncertainties on b0 and Porb for AT 2022dbl (0.63+0.00-0.00) are a red flag that the MCMC did not actually sample the posterior; at minimum they need to be explained, or the parameters reported as unconstrained.\n\nThe bigger scientific soft spot is the radius assumption. In Section 2.4 the remnant is assumed to return to a ZAMS radius at the lower mass, with unchanged pericenter. That forces every later flare to be dimmer. The author is transparent in Section 4.1 that this is not appropriate when the second flare is brighter, so the limitation is acknowledged. For these two targets the second flares are in fact dimmer, so the assumption is consistent with the data, but it is still a choice, not a derived result. If tidal heating expands the remnant, the fitted b0 and the predicted third-flare brightness shift. The forecasts are therefore best read as predictions of this specific model, not as a robust astrophysical inevitability. That is fine as long as it is said, because the whole point is that they are testable.\n\nWhat is missing is the code and data release; the paper describes a fitting code but does not say where to get it. For a paper whose main product is a tool, that is a concrete omission.\n\nNet: this deserves a serious referee. The model is an assembly rather than a new mechanism, but the combination plus the forecasts is a contribution the subfield will want. I would suggest major revision with emphasis on honest fit diagnostics, explanation of zero uncertainties, and public release of code and data. When the third flares do or do not appear, the paper will be judged by them.","headline":"A useful two-flare fitting tool with testable forecasts, but the 'well fitted' claim outruns the chi-square and the third-flare brightness rests on a ZAMS-radius assumption that needs to be presented more honestly.","tokens_in":23458,"tokens_out":3142,"would_cite":true,"duration_ms":27074,"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":"A single star stripped twice by a black hole can explain re-brightening tidal disruption events, and the model predicts when the next flares will appear.","keywords":["tidal disruption events","partial tidal disruption","re-brightening transients","light curve modeling","supermassive black holes","AT 2022dbl","AT 2023adr","mass fallback rate"],"falsifier":"Watch the two predicted third flares: if AT 2022dbl does not re-brighten near MJD 61061 at about 18--19 mag, or AT 2023adr does not re-brighten near MJD 60787 at about 22 mag, the model's predicted remnant state is wrong. A more targeted test is to catch the third flare's rise: if it is much brighter than the second flare, the remnant must have expanded after absorbing tidal energy, contradicting the assumed main-sequence radius recovery.","tokens_in":22356,"feed_emoji":"🔭","tokens_out":5671,"duration_ms":46886,"temperature":0.7,"pith_summary":"Re-brightening tidal disruption events, where a flare fades and then brightens again, are hard to explain with existing single-flare models. This paper tries to establish that both flares come from one star that is partially torn apart twice: the first encounter strips some mass, the surviving remnant keeps orbiting the black hole on an eccentric orbit, and a second partial disruption produces the second flare. The author builds a composite mass fallback rate from two such partial disruptions, converts it into UV/optical light curves, and fits the observed data of AT 2022dbl and AT 2023adr. The fit works, and on that basis the paper predicts when and how bright the third flare of each source should be.","feed_headline":"One star, stripped twice, explains re-brightening TDEs","feed_subtitle":"A two-partial-disruption model fits AT 2022dbl and AT 2023adr and predicts their next flares.","key_machinery":"The load-bearing object is the composite mass fallback rate for two consecutive partial tidal disruptions of the same star. It is constructed by taking the debris energy distribution $dm/d\\epsilon$ from a parabolic disruption, shifting it by the orbital energy of an eccentric orbit, computing the stripped mass, updating the remnant's mass and radius through a zero-age main-sequence mass-radius relation, updating its orbital energy with a fitted formula from simulations, and then summing the fallback of the first and second disruptions. The amount of stripping is set by a scaled penetration factor $b$, related to the ratio of tidal radius to pericenter distance. This composite rate is converted to a bolometric luminosity through a constant radiation efficiency and a viscous delay, then to multi-band magnitudes through a luminosity-dependent photosphere with a black-body spectrum; the two flares are combined in flux space. The whole chain lets one set of physical parameters describe both flares at once.","core_discovery":"The central claim is that the observed double-flare light curves of AT 2022dbl and AT 2023adr can be reproduced by a single surviving star that experiences two partial tidal disruptions. The key physical updates relative to standard parabolic-disruption models are that the star's orbit is eccentric, so the debris energy distribution is shifted toward negative binding energy, and that the remnant's orbital energy, period, mass, and radius are updated after each encounter using prescriptions from hydrodynamical simulations. With these ingredients, the model fits both light curves, yields black hole masses around $10^{6.9}$--$10^{7.2}\\,M_\\odot$, and yields the forecast that AT 2022dbl will re-brighten near MJD 61061 at roughly 18--19 mag and AT 2023adr near MJD 60787 at roughly 22 mag.","pith_inferences":["If the predicted third flares arrive at the stated times and magnitudes, that would strongly favor repeated partial disruption over alternative double-star or two-phase accretion explanations; a non-detection at the predicted brightness would instead point to the remnant star expanding rather than returning to its main-sequence radius.","The same two-flare construction should be testable on other gradually dimming re-brightening TDEs, while systems with a brighter second flare would require allowing the remnant radius or pericenter to change between encounters.","Because the model predicts both time and brightness, it turns each re-brightening TDE into a scheduling tool for catching the early rise of the next flare, which is exactly the phase where competing shock-powered emission models differ most."],"forward_implications":["If the model is right, re-brightening TDEs with a dimmer second peak are the expected outcome of repeated partial disruptions, with the flare gap set by the remnant's updated orbital period.","AT 2022dbl should re-brighten near MJD 61061, around January 2026, at roughly 18--19 mag in the UV/optical bands, bright enough for ground and space telescopes.","AT 2023adr should re-brighten near MJD 60787, around April 2025, at roughly 22 mag, requiring deeper monitoring.","Existing single-flare fitting tools are inadequate for these systems, because the first flare still contributes during the second peak and the eccentric-orbit fallback differs from parabolic fallback.","The same machinery can be applied to single-flare TDEs with slightly non-parabolic orbits, turning them into candidates for a future second flare."],"supporting_citations":[{"why":"Supplies the original debris energy distributions and scaling relations for partial disruption fallback that the model starts from.","marker":"GRR2013"},{"why":"Provides the luminosity-dependent photosphere radiation model and viscous delay conversion used to turn fallback into light curves.","marker":"Mockler et al. (2019)"},{"why":"Justifies shifting the energy distribution by the orbital energy for eccentric, e < 1 disruptions.","marker":"Hayasaki et al. (2013)"},{"why":"Provides the fitting formulae for the orbital energy change of the remnant star after each partial disruption.","marker":"Chen et al. (2024)"},{"why":"Supplies the zero-age main-sequence mass-radius relation used to update the remnant radius after stripping.","marker":"Tout et al. (1996)"},{"why":"Supports the assumption that the pericenter of the remnant orbit stays nearly unchanged after a partial disruption.","marker":"Ryu et al. (2020)"},{"why":"Supplies the AT 2022dbl multi-band photometry and the earlier identification of it as a repeating partial TDE.","marker":"Lin et al. (2024)"},{"why":"Validates the eccentric-orbit energy shift and quantifies the small tidal-distortion deviations from parabolic cases.","marker":"Liu et al. (2023a)"}],"fun_headline_variants":["Two partial disruptions from one star explain re-brightening TDEs","Eccentric orbit model fits double-flare TDEs and predicts next flares","One star, two partial disruptions: new model for re-brightening TDEs","Partial disruption model predicts upcoming flares of two TDEs","Re-brightening TDEs: same star disrupted twice, model fits and predicts"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that after each partial stripping the surviving star immediately shrinks back to the radius of a normal zero-age main-sequence star of its new mass and keeps the same closest approach distance, which forces every later flare to be dimmer than the one before it.","fun_headline_variants_meta":{"raw":{"variants":["Two partial disruptions from one star explain re-brightening TDEs","Eccentric orbit model fits double-flare TDEs and predicts next flares","One star, two partial disruptions: new model for re-brightening TDEs","Partial disruption model predicts upcoming flares of two TDEs","Re-brightening TDEs: same star disrupted twice, model fits and predicts"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000239,"raw_usage":{"total_tokens":1500,"prompt_tokens":920,"completion_tokens":580,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":536,"completion_tokens_details":{"reasoning_tokens":479}},"tokens_in":536,"tokens_out":580,"duration_ms":5465,"temperature":1.0,"reasoning_tokens":479,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:58:09.706562+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Watch the two predicted third flares: if AT 2022dbl does not re-brighten near MJD 61061 at about 18--19 mag, or AT 2023adr does not re-brighten near MJD 60787 at about 22 mag, the model's predicted remnant state is wrong. A more targeted test is to catch the third flare's rise: if it is much brighter than the second flare, the remnant must have expanded after absorbing tidal energy, contradicting the assumed main-sequence radius recovery.","supporting_citations":[],"review_version":1}