{"id":"ae138fe7-0194-410b-bab9-080870d202d1","arxiv_id":"2607.22426","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"TWA 10 hosts a ~191 au debris disk—the largest known around any M-dwarf—discovered in JWST NIRCam scattered light.","lead":"JWST NIRCam images reveal a huge, previously unknown debris disk around the M-dwarf TWA 10, with a radius of about 191 au, and new sharp views of the known TWA 25 disk. This makes TWA 10 the largest M-dwarf debris disk found, hinting that faint disks around cool stars have been missed by older surveys.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"TWA 10's 'largest M-dwarf disk' claim rests on a single ring-model r0 that is degenerate with the unconstrained scattering phase function; Powell/emcee differ by 34 au and model uncertainty is not propagated.","rationale":"The reader's conditional verdict is appropriate, and I identify the same weakest link. The discovery of an extended scattered-light structure around TWA 10 is probably real: it appears in F200W, is plausibly traced at F444W, and the companion candidates pass jackknife tests. Those parts of the paper are not the problem. The problem is the quantitative 'largest known' statement. It uses the emcee r0 of 191 au as a headline, while the Powell fit gives 157 au and the SPF—one of the main factors controlling where scattered light peaks—is explicitly unconstrained. The statistical error bars from emcee therefore understate the true uncertainty by ignoring model degeneracy. Since the comparison sample of M-dwarf disks has radii ≲70 au, the claim 'largest known' is not sensitive to the difference between 157 and 191, but it is sensitive to whether an equally plausible model could put r0 below ~100 au or lower. The proposed sensitivity test — refit with alternate SPF and free h0 — would directly decide this. If the r0 posterior is stable above ~100 au, the abstract's claim is fine; if not, the paper should present the radius as model-dependent and move the 'largest' claim to a caveat. This is a correctable, not fatal, issue, so the reader's conditional verdict stands unchanged.","tokens_in":24852,"tokens_out":7572,"duration_ms":93387,"concrete_test":"Re-run the TWA 10 F200W forward-model fit (same MCRDI images) with: (i) a single-component HG SPF, (ii) the two-component HG with g2 sampled from a wide prior instead of the current best fit, and (iii) h0 free over 0.01–0.2, starting emcee from both the Powell and emcee best-fit values. If the 95% credible interval for r0 remains above ~100 au in all variants, the largest-disk claim is robust; if any variant yields r0 consistent with ≤100 au (or below the ~70 au scale of other M-dwarfs), the headline should be downgraded to a model-dependent candidate.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline claim — that TWA 10 hosts the largest known M-dwarf debris disk, at ~191 au — depends directly on r0 from the emcee forward model in Table 3. The model is a single axisymmetric ring (Augereau et al. 1999) with a two-parameter Henyey-Greenstein SPF and h0 fixed at 0.05. The paper states in Section 3.1 that the SPF is difficult to constrain, and in Section 5.1 that no constraint is obtained for g2 in F200W or for any SPF parameter in F444W. Since scattered-light surface brightness is the product of the radial density and an anisotropic SPF, r0 is strongly degenerate with SPF and vertical structure. The Powell fit gives r0=157 au, not 191 au; the 34 au gap is larger than the quoted 1σ uncertainty (±6.5 au) and is not included in the claim. A plausible alternative SPF or a free h0 could move the inferred peak radius well below 191 au. Even if the disk is real, the 'largest known' claim is not yet robust to model choice. The discrepancy is acknowledged, but the Abstract and Summary state 191 au without a systematic uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents JWST NIRCam coronagraphic F200W/F444W observations of two M-dwarfs in TW Hydra, TWA 10 and TWA 25. It reports the discovery and first resolved imaging of a debris disk around TWA 10, re-images the known TWA 25 disk, and fits both disks with an axisymmetric Augereau et al. (1999) ring model. From these fits the authors derive a radius of ~191 au for TWA 10, claim this is the largest known M-dwarf debris disk, and argue that the TWA 25 disk has a steep inner edge and a brightness asymmetry consistent with planetary sculpting. The paper also uses MCFOST to estimate SEDs and dust masses, presents 5σ contrast curves and planet-detection probability maps, and identifies three point-source candidates that are shown to be likely background objects. The central discovery claim—the exceptionally large TWA 10 radius—is interesting but, as discussed below, is not yet robust to the modeling assumptions used to derive it.","tokens_in":25181,"tokens_out":4733,"duration_ms":62290,"significance":"If the ~191 au radius is robust, the TWA 10 disk would be a landmark object: it would be the largest resolved debris disk around an M-dwarf, challenge the empirical radius–luminosity relation, and support the idea that many M-dwarf disks are too faint for previous SED surveys. The strengths of the paper include the careful reduction with public pipelines (spaceKLIP/Winnie), the use of forward-model-constrained RDI, jack-knife tests of candidate companions, and a sensible color-based rejection of background objects. The data are also made available through a MAST DOI. However, the headline radius currently rests on a single model that fixes h0 and leaves the scattering phase function largely unconstrained, while the two fitting methods used by the authors disagree by 34 au—five times the quoted 1σ uncertainty. Because the abstract and summary state the emcee radius without this systematic uncertainty, the main quantitative claim needs additional robustness testing or a substantially more cautious presentation.","major_comments":[{"comment":"The headline 'largest known M-dwarf debris disk' is not yet robust. Table 3 gives Powell r0 = 157.13 au and emcee r0 = 191.15 ± 6.53 au for TWA 10 in F200W. The 34 au difference is much larger than the quoted 1σ uncertainty and is not propagated into the abstract or summary, which quote 191 au as the measured radius. This matters because the scattered-light surface brightness is the product of the radial density profile and an anisotropic, two-parameter Henyey-Greenstein SPF; Section 5.1 explicitly states that the SPF is difficult to constrain, with no constraint on g2 in F200W and no constraint on any SPF parameter in F444W. The model also fixes h0 = 0.05. A robust claim requires either a systematic uncertainty that covers the Powell–emcee range, a model-selection argument for preferring the emcee solution, or a test of sensitivity to h0 and SPF parameterization. At minimum, the paper s","section":"§3.1, Table 3; §5.1; Abstract and §7"},{"comment":"The statement that TWA 10 and TWA 25 were 'likely missed by previous disk detection surveys due to low luminosity fractions' is presented in the Summary as a conclusion, but the MCFOST SEDs in §4.3 are normalized to the F200W best-fit model and the dust mass is scaled to match that model's surface brightness. The resulting luminosity fractions (~3.6e-5 and ~6.9e-5) are therefore illustrative estimates, not independent predictions from the SED. The paper acknowledges degeneracies, but the language in the Summary should be softened, and the dependence of the disk SED on r0 should be noted: if r0 shifts from 191 to 157 au, the dust mass and peak flux of the model SED would also change.","section":"§4.3, §5.1"},{"comment":"For TWA 25, the derived radius of 63.8 ± 3.3 au is formally inconsistent with the SPHERE estimate of 76 ± 2 au at about 3.5σ. The paper attributes this to ADI self-subtraction and to SPHERE's ellipse-fitting approach, which is plausible but not demonstrated. Since the same axisymmetric ring model is used for both disks and a significant part of the paper's credibility on TWA 10 depends on this modeling approach, it would strengthen the analysis to test TWA 25 with at least one alternative model (e.g., free h0 or a single-component HG SPF) and quantify how the radius changes. This would also give a better sense of the model uncertainty that likely affects TWA 10.","section":"§5.2, Table 3"}],"minor_comments":[{"comment":"Typo: 'stis appears axisymmetric' should be 'still appears axisymmetric'.","section":"§7"},{"comment":"Typo: 'circled in whie' should be 'circled in white'.","section":"Figure 12 caption"},{"comment":"The TWA 10 2MASS Ks flux of 3.5e-2 ± 9.9e-3 Jy appears anomalously low compared to the listed J and H fluxes and to WISE3/WISE4. This may be a typographical or unit error; please check.","section":"Table 5"},{"comment":"The text lists Ren et al. (2023) as a previous detection but only compares the new results with NICMOS and SPHERE. A brief comparison with the STIS-based analysis would be useful.","section":"§5.2"},{"comment":"The paper uses both 'TWA' and 'TW A' (e.g., 'TW A 10' in text and title, 'TWA 10' in tables/abstract). Please unify the notation.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The central result is likely correct in spirit—the TWA 10 disk appears large and no obvious planet is detected—but the headline radius and the 'largest known' claim need to be made robust against the acknowledged SPF/h0 degeneracies. The authors should be asked to report a systematic error or a credible range, and to re-word the abstract and Summary accordingly. I do not see grounds to reject, as the data and reduction appear sound and the manuscript is generally careful; the issue is one of quantitative robustness and presentation rather than a fundamental flaw."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The headline result is a discovery, but the record claim is not yet robust. The F200W image of TWA 10 shows a real debris disk, and the paper does a careful job with RDI/MCRDI, forward modeling, and an honest treatment of how faint the disk is — the F444W detection needs heavy binning, and the MIPS/PACS non-detections are fully consistent with a faint disk. The new result matters: TWA 10 is only the second M-dwarf debris disk discovered with JWST, and it pushes the sample to larger radii.\n\nWhat this paper does well: the reduction uses public tools (spaceKLIP/Winnie) and the data are public on MAST; the forward-modeling residuals look reasonable; the TWA 25 re-analysis improves on earlier NICMOS/SPHERE constraints, revises the radius downward, and finds a steep inner edge and an asymmetry; the candidate companions are sensibly rejected as likely background objects using color-color comparison; and the SED modeling is explicitly labeled as illustrative, not a prediction.\n\nThe soft spot is the 191 au radius. Table 3 shows Powell giving 157 au and emcee giving 191 ± 6.5 au. That 34 au gap is far larger than the quoted 1σ uncertainty, and neither the abstract nor the summary reports a systematic error. The model is a single ring with h0 fixed at 0.05 and a two-parameter HG SPF that the authors themselves say is poorly constrained (no constraint on g2 in F200W, none at all in F444W). In scattered light, r0 is degenerate with the SPF and vertical structure; a different phase function or a free h0 could plausibly move the peak well below 191 au. Even at 157 au the disk would be among the larger M-dwarf disks, but 'largest known' needs to be defended against this model dependence, and the comparison sample of resolved M-dwarf disks is very small.\n\nThe planet non-detection discussion is fine but inherits the same inner-radius uncertainty. The 'we should have detected it' claim is only as strong as the assumption that the inner edge is planet-sculpted and that the radius is correct. The SED-based dust masses and luminosity fractions are scaled to the F200W model, so they are not independent; the authors mostly acknowledge this.\n\nBottom line: this is a solid observational paper that deserves peer review, not a desk rejection. Send it out, but ask the authors to present the radius as a range (or at least to propagate the Powell/emcee disagreement as a systematic uncertainty) and to temper the record claim unless they can show the radius is robust to SPF/scale-height choices. I'd cite it with a caveat, and it's a good reading-group paper for the debris-disk and JWST imaging community.","headline":"Real and useful disk discovery, but the '191 au record' radius needs a model-systematic error before it can be quoted as the largest known M-dwarf disk.","tokens_in":25870,"tokens_out":4194,"would_cite":true,"duration_ms":48939,"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":"JWST reveals the largest known M-dwarf debris disk, a ~191 au ring around TWA 10.","keywords":["TWA 10","TWA 25","M-dwarf debris disk","JWST NIRCam","scattered-light imaging","disk radius","planet sculpting","infrared surveys"],"falsifier":"ALMA submillimeter imaging of TWA 10 that resolves the thermal emission from large grains: if the millimeter continuum peaks at a radius substantially below ~191 au (e.g., <100 au), then the scattered-light ring is tracing small, radiation- or wind-blown grains rather than the planetesimal belt, and the claim of an exceptionally large disk is wrong. Alternatively, a re-fit of the NIRCam data with a more flexible model (free scale height, broken power law, or eccentric ring) that yields a significantly smaller r0 would falsify the headline claim.","tokens_in":24713,"feed_emoji":"🪐","tokens_out":5448,"duration_ms":55351,"temperature":0.7,"pith_summary":"This paper reports the first resolved images of a debris disk around the M-dwarf TWA 10, taken with JWST's NIRCam coronagraph, and new images of a previously known disk around TWA 25. The central claim is that the TWA 10 disk is exceptionally large, with a modeled radius of about 191 astronomical units, making it the largest debris disk ever found around an M-dwarf. If correct, this single system breaks the empirical trend that M-dwarf disks are compact, and it suggests that many similar disks are too faint in the infrared for earlier surveys to have detected. For TWA 25, the data sharpen the disk's inner edge to a steep density slope and reveal a moderate 2-micron brightness asymmetry, hinting at planet sculpting and recent collisions. The paper also argues that both disks were missed by Spitzer and Herschel because their infrared luminosity fractions are extremely low, not because such disks are rare.","feed_headline":"JWST spots largest known M-dwarf debris disk","feed_subtitle":"A 191-au ring around TWA 10 suggests many M-dwarf disks are too faint for earlier surveys.","key_machinery":"The central tool is forward modeling of the coronagraphic images: a simple axisymmetric ring model (fiducial radius r0, scale height h0=0.05, radial power-law exponents α_in and α_out, and a two-parameter scattering phase function) is fitted to the NIRCam F200W and F444W data after reference-star differential imaging. The same model is then fed to a radiative-transfer code to predict the disk's spectral energy distribution, demonstrating that the disks could have escaped detection in previous infrared surveys. The fitted r0 is the quantity that carries the paper's headline claim.","core_discovery":"The discovery is that TWA 10 hosts a debris disk with radius ~191 au, more than twice the typical size of previously imaged M-dwarf debris disks and far outside the radius–luminosity scaling seen for brighter stars. The authors derive this from forward models of the NIRCam scattered-light images, using a ring-like disk with a two-parameter scattering phase function, and they note the radius is formally uncertain (the two fitting runs give 191 and 157 au). In addition, the paper characterizes TWA 25's disk as having a very steep inner density power-law (~12–17) and a 10–25% brightness asymmetry in the F200W filter, which they interpret as likely evidence of an inner planetary system or a rece","pith_inferences":["If M-dwarf disks are commonly this faint, then the known population of M-dwarf debris disks is heavily biased toward the brightest and most compact examples; a systematic JWST survey of nearby M-dwarfs could reveal a much wider range of disk sizes and frequencies.","The TWA 10 disk's scattered-light radius may not coincide with the planetesimal belt if stellar winds push small grains outward; the paper assumes the peak surface brightness marks the belt, but longer-wavelength imaging of large grains would test this.","The steep inner edge of TWA 25 adds to a pattern seen in other debris disks where inner edges are too steep for pure collisional grinding and too shallow for a single planet, supporting a multi-planet or migration scenario.","A direct follow-up prediction: deep ALMA observations should detect thermal emission from large grains at ~191 au in TWA 10 and ~64 au in TWA 25; if the millimeter radius of TWA 10 is much smaller, the scattered-light ring is dominated by blowout grains and the 'giant disk' claim would be weakened."],"forward_implications":["If the 191 au radius holds, TWA 10 becomes the largest M-dwarf debris disk known, requiring a re-examination of the disk radius–stellar luminosity scaling.","The low luminosity fractions derived for these disks (about 3.6e-5 and 6.9e-5) imply that previous Spitzer and Herschel surveys would have missed them, meaning the census of M-dwarf debris disks is incomplete.","TWA 25's sharp inner edge (α_in ~ 12–17) and the F200W brightness asymmetry are consistent with a sculpting planet or a recent collision; the paper's planet-search shows that if a single massive planet were responsible, it should have been detected, so the architecture must be different (multiple smaller planets, inward migration, or non-planetary sculpting).","The three point sources found in the images are all likely background objects based on their F200W–F444W colors, so no new companions are confirmed in these systems."],"fun_headline_variants":["JWST finds largest M-dwarf debris disk yet, ~191 au","Record-size debris disk discovered around dim star TWA 10","TWA 10's 191-au disk breaks M-dwarf debris disk record","JWST reveals largest debris disk around any M-dwarf","Giant debris ring around TWA 10 hints at hidden M-dwarf disks"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The 'largest M-dwarf disk' claim rests on a forward model that assumes a single axisymmetric ring with a fixed scale height and a two-parameter scattering phase function; the two fitting methods disagree on the radius (191 vs 157 au) and the phase function is unconstrained, so the true disk radius could be significantly smaller if the model is too simple.","fun_headline_variants_meta":{"raw":{"variants":["JWST finds largest M-dwarf debris disk yet, ~191 au","Record-size debris disk discovered around dim star TWA 10","TWA 10's 191-au disk breaks M-dwarf debris disk record","JWST reveals largest debris disk around any M-dwarf","Giant debris ring around TWA 10 hints at hidden M-dwarf disks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00088,"raw_usage":{"total_tokens":3679,"prompt_tokens":818,"completion_tokens":2861,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":562,"completion_tokens_details":{"reasoning_tokens":2766}},"tokens_in":562,"tokens_out":2861,"duration_ms":19640,"temperature":1.0,"reasoning_tokens":2766,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T04:44:45.458354+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"ALMA submillimeter imaging of TWA 10 that resolves the thermal emission from large grains: if the millimeter continuum peaks at a radius substantially below ~191 au (e.g., <100 au), then the scattered-light ring is tracing small, radiation- or wind-blown grains rather than the planetesimal belt, and the claim of an exceptionally large disk is wrong. Alternatively, a re-fit of the NIRCam data with a more flexible model (free scale height, broken power law, or eccentric ring) that yields a significantly smaller r0 would falsify the headline claim.","supporting_citations":[],"review_version":1}