{"id":"2353808f-7150-437a-aa42-553b0be0dcb0","arxiv_id":"2501.14596","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"JWST transmission spectra of the super-Earth TOI-776b are flat, ruling out low-metallicity hydrogen atmospheres below about 100 times solar metallicity at a pressure of 1 millibar.","lead":"Two JWST transits of the super-Earth TOI-776b yield a flat 2.8-5.2 micron spectrum, and the authors rule out atmospheres below 100 times solar metallicity at a pressure of one millibar. The work adds a cautionary example of how different visits of the same planet can produce different atmospheric limits.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline claim hinges on Visit 2 ExoTiC-JEDI, whose 3σ metallicity limit sits exactly at 100× solar; the NRS1/NRS2 offset used to correct that visit is fitted to the same data and fixed without propagating its uncertainty or testing wavelength dependence, so the 3σ boundary is not robust.","rationale":"The reader's weakest assumption about the Visit 2 offset is the same concern I identify, and it is load-bearing because the central claim is a conservative lower limit that is exactly marginal for the least constraining reduction (Visit 2 ExoTiC-JEDI, 100× solar at 3σ). The offset is fit to the same data used for model comparison, and its uncertainty is not propagated; the paper even notes that adjacent metallicity models differ by only 5-7 ppm, similar to the offset uncertainty. The non-physical fits also show that the step offset is degenerate with a Gaussian feature in NRS1, so the correction could remove a real spectral feature. A correct treatment would marginalize over the offset or fit it simultaneously, and would test for wavelength dependence. The paper has genuine strengths: two independent reductions agree, the data products are public, and the authors explicitly quote the most conservative limit. These do not remove the fragility of the exact 3σ boundary. Because the concern is real but the verdict CONDITIONAL already captures the need for such robustness checks, no verdict change is needed.","tokens_in":19877,"tokens_out":15139,"duration_ms":129932,"concrete_test":"Recompute the Visit 2 ExoTiC-JEDI exclusion contours using a joint fit that treats the NRS1/NRS2 offset as a free parameter, allowing both a step amplitude and a possible linear slope across NRS1, simultaneously with each PICASO model. Evaluate the significance from Δχ² relative to the best offset model with the correct degrees of freedom. If the 3σ boundary at 10^-3 bar falls below 100× solar, the headline claim is not supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim (Section 6: 'most conservatively rule out atmospheres less than 100× solar metallicity' at 10^-3 bar) is exactly marginal for Visit 2 ExoTiC-JEDI, quoted as ruling out up to 100× solar at 3σ. The Visit 2 NRS1/NRS2 offset (~60-70 ppm, Table 3) is fitted to the same Visit 2 spectrum in Section 5.1, then fixed while computing χ²/σ for the PICASO models in Section 5.2, with no propagation of its uncertainty and no test of wavelength dependence. The offset uncertainty (7 ppm) is comparable to the transit-depth differences between adjacent metallicity models (5-7 ppm between 350× and 450× solar, per Section 6). Furthermore, the non-physical fits show that a Gaussian in NRS1 fits as well as the step offset (ΔlnZ ≤ 1, Table 3), so the step may be absorbing a real, narrow spectral feature; if so, subtracting it artificially flattens the data and biases the exclusion contours. Because the 3σ boundary for the weakest reduction sits exactly at 100× solar, a small shift from either effect would invalidate the headline claim.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents two JWST NIRSpec/G395H transit observations of the super-Earth TOI-776b, reduced independently with the ExoTiC-JEDI and Eureka! pipelines. The resulting 2.8–5.2 micron transmission spectra have a median precision of 34 ppm per 0.02 micron bin. The two reductions agree well for each visit, but the two visits show different overall structure: Visit 1 is consistent with a flat line, while Visit 2 requires a step offset between NRS1 and NRS2. After correcting for that offset, the authors compare the spectra to PICASO forward models over a grid of metallicity and opaque pressure. They conclude that atmospheres below 100x solar metallicity at an opaque pressure of 10^-3 bar are ruled out at >=3 sigma in all visits and reductions, with stronger visit- and reduction-dependent limits reaching 350–470x solar for Visit 1 and 100–130x solar for Visit 2.","tokens_in":20177,"tokens_out":5201,"duration_ms":47044,"significance":"If the central claim holds, this is a useful addition to the growing JWST sample of super-Earth atmospheric constraints, and it strengthens the COMPASS program's statistical approach. The paper's strengths include the use of two independent reduction pipelines, the public release of data products on Zenodo, the explicit discussion of visit-to-visit and reduction-to-reduction differences, and the framing of a conservative lower metallicity limit rather than a claimed detection. The conclusion that TOI-776b likely has a very thin, very cloudy, or highly metal-rich atmosphere is credible and informative for future population-level interpretations.","major_comments":[{"comment":"The abstract reverses the visit-specific metallicity limits. The abstract states \"Visit 1 ruling out ≲100× solar while the lower limits for Visit 2 extend beyond ∼350× solar,\" but Section 6 and Figure 6 report the opposite: at 10^-3 bar, Visit 1 excludes 350× solar (ExoTiC-JEDI) and 470× solar (Eureka!) while Visit 2 excludes only 100× solar (ExoTiC-JEDI) and 130× solar (Eureka!). Since the visit comparison is a key result and the conservative 100× solar floor is the headline, the abstract must be corrected to match the body of the paper.","section":"Abstract and Section 6"},{"comment":"The Visit 2 NRS1/NRS2 offset is fitted to the binned transmission spectrum in Section 5.1 and then treated as a fixed correction when computing chi-square/N and sigma for the PICASO models. The offset uncertainties (7.2 ppm for ExoTiC-JEDI, 6.2 ppm for Eureka!) are comparable to the 5–7 ppm differences between adjacent high-metallicity models quoted in Section 6, yet the significance contours in Figure 6 do not propagate this uncertainty or test whether the offset is wavelength-dependent or time-varying. Because the weakest case (ExoTiC-JEDI Visit 2) places the 3-sigma boundary at exactly 100× solar, the headline claim rests on the stability of this single-step correction. Please propagate the offset uncertainty into the model comparison, or demonstrate that a wavelength-dependent or time-varying offset moves the 10^-3 bar exclusion boundary by less than the metallicity grid spacing.","section":"Section 5.2, Table 3, Figure 6"},{"comment":"For Visit 2, the step-offset model and the five-parameter Gaussian-in-NRS1 model have Delta lnZ <= 1 for both reductions, so the step function is not statistically preferred over a narrow spectral feature. The paper argues that the Gaussian is not consistent between visits or reductions, but that does not rule out the possibility that the step offset is partially astrophysical in origin, especially given that the Gaussian central wavelength (3.43–3.45 um) lies in the methane band. If the step subtraction removes a real spectral feature, the flatness of the corrected data would be artificially enhanced and the metallicity exclusion contours biased. Please quantify the amplitude and width of the best-fit Gaussian and re-run the physical model comparison treating the offset as a free parameter (or allowing a wavelength-dependent offset) to verify that the 100× solar exclusion at 10^-3 bar is not an artifact of the chosen step correction.","section":"Section 5.1, Table 3"}],"minor_comments":[{"comment":"The term \"reduced-χ2\" is an unusual construction; consider using \"reduced χ²\" or simply \"χ²/N\" throughout for consistency with standard terminology.","section":"Section 5.2"},{"comment":"Table 4 reports chi-square/N and sigma values at 1 bar opaque pressure, while the headline claim concerns 10^-3 bar. Since the 100× solar row for Visit 2 ExoTiC-JEDI shows sigma = 2.6 at 1 bar, a footnote clarifying that the table is for 1 bar only would prevent readers from interpreting these entries as contradicting the abstract.","section":"Table 4"},{"comment":"In the bottom panel of Figure 6, the four 3-sigma contours are hard to distinguish because of small labels and overlapping line styles; direct labels on each contour would improve readability.","section":"Figure 6"},{"comment":"The sentence describing removal of 15 integrations around the HGA move would benefit from clarifying whether this removal was performed before or after the iterative 4-sigma outlier rejection, and whether the alternative of retaining those integrations was tested.","section":"Section 3.1"},{"comment":"The citation \"May & MacDonald et al. 2023\" is inconsistent with the author-list style used elsewhere; it should be \"May, MacDonald, et al. 2023\" (or similar) to match the other multi-author citations.","section":"Section 1, references"}],"recommendation":"major_revision","confidential_remarks":"The main risk to the central claim is the treatment of the Visit 2 NRS1/NRS2 offset: it is fitted to the same data used for the atmospheric model comparison, and the 100× solar boundary sits exactly on the weakest reduction. If the authors can propagate the offset uncertainty and show that a wavelength-dependent or spectral-feature interpretation does not move the boundary, the conservative conclusion will be solid. The abstract's reversal of the visit limits is a serious presentation error that should be fixed before acceptance. The paper is otherwise well within the scope and methodological standards of the journal."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here’s my read of the COMPASS TOI-776b paper. It’s a solid, incremental addition to the super-Earth transmission-spectrum sample. The genuinely new piece is the first published NIRSpec/G395H spectrum of TOI-776b, and the analysis is careful in the ways that matter: two independent reductions (ExoTiC-JEDI and Eureka!) give consistent per-visit spectra, the data products are on Zenodo, and the authors explicitly test whether the choice of solar abundance table changes the constraints (it doesn’t). The demonstration that visit-to-visit and reduction-to-reduction differences can shift quoted metallicity limits by factors of 3–4 is a useful caution for the field.\n\nThe main result—flat, high-mean-molecular-weight spectrum consistent with >100× solar metallicity at 1 mbar—is credible. Visit 1 rules out below ~350–470× solar, Visit 2 only ~100–130×, and the two-pipeline agreement is genuine. But there are two soft spots that should be fixed before publication.\n\nFirst, the abstract gets the visit direction backwards: it says Visit 1 rules out ≲100× solar and Visit 2 extends beyond ~350×, while the body and Figure 6 show the opposite. That’s a simple swap, but it’s confusing for anyone skimming.\n\nSecond, and more substantive, the Visit 2 NRS1/NRS2 offset is fit to the same spectrum that is then compared to the PICASO models, and the offset’s uncertainty is not propagated into the reported σ contours. The offset is ~65 ppm with a ~7 ppm error, and the non-physical fits show a Gaussian in NRS1 fits nearly as well as the step (ΔlnZ ≈ 1). So the step could be absorbing real spectral structure. This matters most for the Visit 2 ExoTiC-JEDI bound, which sits right at 100× solar. If the offset were wavelength-dependent, that single reduction’s 3σ boundary could move. I don’t think this overturns the paper’s conservative claim—Visit 1 alone gives a much stronger limit, and Eureka’s Visit 2 is a bit above 100×—but the “across all visits and reductions” phrasing is stronger than the analysis supports.\n\nThe paper deserves a serious referee. The issues are fixable: correct the abstract, propagate the offset uncertainty, and perhaps test an offset with a linear slope across NRS1. I’d send it out, with encouragement to the authors to soften the headline claim or strengthen the offset analysis. For the reading group, it’s a useful case study in reduction-induced systematics, but not a must-read.\n\nRecommendation: accept after minor revision.","headline":"New TOI-776b spectrum with careful dual-reduction analysis, but the abstract reverses the visit limits and the Visit 2 detector offset is fit to the same data, leaving the headline 3σ bound a bit softer than the prose claims.","tokens_in":20804,"tokens_out":5471,"would_cite":true,"duration_ms":45597,"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":"The paper reports that two JWST transits of the super-Earth TOI-776b rule out any clear, low-metallicity hydrogen atmosphere at the 1-millibar level.","keywords":["Exoplanet atmospheric composition","Exoplanet atmospheres","Exoplanets","Infrared spectroscopy","super-Earth","transmission spectroscopy","JWST NIRSpec G395H","TOI-776b"],"falsifier":"A decisive check would be a third transit of TOI-776b observed at a different telescope roll angle, or with NIRSpec PRISM, so that no NRS1/NRS2 step correction is needed; if a methane band at ~3.3 µm or a CO2 band at ~4.3 µm then appears with an amplitude consistent with a <100× solar atmosphere, the paper's exclusion is wrong. A cheaper test is available now: re-fit the Visit 2 spectrum allowing the NRS1/NRS2 offset to be wavelength-dependent and see whether the 3σ excluded-metallicity boundary shifts by more than the stated visit-to-visit spread.","tokens_in":19649,"feed_emoji":"🪐","tokens_out":10639,"duration_ms":89146,"temperature":0.7,"pith_summary":"Using two independent reductions of two NIRSpec/G395H transit visits, the authors measure TOI-776b's 2.8–5.2 µm transmission spectrum at a median precision of 34 ppm per 0.02 µm bin. They find no reliable molecular features: Visit 1 is best described by a flat line, and Visit 2 by a flat line plus a ~65 ppm offset between the NRS1 and NRS2 detectors, which they treat as instrumental. Comparing the spectra to PICASO forward models, they conclude that at an opaque pressure of $10^{-3}$ bar any atmosphere below 100× solar metallicity is excluded at ≥3σ in every visit and every reduction; at 1 bar, 10× solar and below are excluded by more than 8σ. The exact lower limit moves between visits and reductions, from ~100× solar in Visit 2 to ~350–470× solar in Visit 1, so the paper cautions against over-interpreting any single metallicity number. The result bears on what kind of atmosphere, if any, this 1.85 Earth-radius planet can retain, favouring a very thin, heavily clouded, or highly metal-enriched atmosphere over a clear hydrogen-dominated one.","feed_headline":"JWST rules out low-metallicity air on super-Earth TOI-776b","feed_subtitle":"Two transits at 34 ppm precision rule out clear hydrogen air below 100 times solar metallicity.","key_machinery":"The central machinery is a grid of PICASO forward transmission models spanning 1–1000× solar metallicity in 20 logarithmically spaced steps and opaque pressure levels from 1 to $10^{-4}$ bar, where the opaque pressure plays the role of an agnostic, wavelength-independent cloud deck or the planetary surface. PICASO is a radiative-transfer code that computes transmission spectra from chemical-equilibrium abundances supplied by photochem, on temperature-pressure profiles parameterized following Guillot (2010), using the Resampled Opacities database (Batalha et al. 2022) with CH4, CO, CO2, H2O, NH3, Na, K, and dozens of minor species. Each model is rebinned to the data resolution and scored by reduced chi-squared, and the 3σ exclusion contours are obtained by 2D cubic interpolation in the metallicity–opaque-pressure plane. A second piece of machinery is the set of non-physical fits (zero-slope, sloped, step offset between NRS1 and NRS2, and Gaussian features) that establishes the spectral shape preferred by each visit and supplies the offset applied before the physical-model comparison.","core_discovery":"TOI-776b's transmission spectrum is featureless at the achieved precision, and the paper's central claim is a conservative exclusion: using chemical-equilibrium PICASO forward models in which an opaque pressure level represents either a cloud deck or the surface, the authors rule out atmospheres below 100× solar metallicity at $10^{-3}$ bar to ≥3σ across both visits and both reductions. At 1 bar, the exclusion is much stronger, with 10× solar and below rejected by more than 8σ and 100× solar rejected at 7σ for Visit 1. After subtracting the NRS1/NRS2 offset in Visit 2, the 3σ boundary at $10^{-3}$ bar spans ~350× solar (ExoTiC-JEDI) and ~470× solar (Eureka!) for Visit 1, and ~100× and ~130× solar for Visit 2. The atmospheres that survive are a very thin layer (opaque pressure at or below $10^{-4}$ bar), a high-metallicity atmosphere near ~1000× solar at 1 bar, or any of these with a wavelength-independent cloud deck; a bare rock is disfavoured because the planet's density and radius-valley position require some low-density material.","pith_inferences":["An implication the authors leave implicit is that the unexcluded 1× solar, 10^-4 bar corner is unlikely to be a physically persistent atmosphere, so future observing time is better spent distinguishing a bare rock from a thin but stable high-metallicity or cloudy atmosphere than refining the high-metallicity boundary.","If the Visit 2 NRS1/NRS2 offset is actually wavelength-dependent or time-varying, a scalar step correction could be hiding real spectral structure; a third transit at a different roll angle or with NIRSpec PRISM would test whether the 100× solar exclusion survives without the offset correction.","A joint analysis of both visits with a shared systematic model for the detector offset might either sharpen the combined 3σ contour or reveal that the visit-to-visit spread is larger than the photon noise, which would argue for treating multi-visit super-Earth spectra as correlated measurements rather than independent confirmations.","Applying the same metallicity-pressure grid approach to other COMPASS targets could separate cloudy high-metallicity atmospheres from genuinely bare-rock planets, a distinction that density and radius-valley arguments alone cannot make."],"forward_implications":["If the central claim is correct, TOI-776b cannot host a clear, hydrogen-dominated, low-metallicity atmosphere down to the 1-millibar level, so any atmosphere it retains must be metal-rich, very thin, or hidden by an opaque cloud or haze deck.","The conservative 100× solar exclusion at 10^-3 bar extends the COMPASS sample's growing pattern that JWST super-Earth transmission spectra are featureless and point to high mean molecular weights or clouds rather than detectable molecular bands.","Visit- and reduction-dependent limits (roughly 100× to 470× solar at 10^-3 bar) imply that comparing single quoted metallicities across planets or programs without accounting for these systematics could produce misleading population-level conclusions.","The paper's identification of allowed low-metallicity, low-pressure parameter space shows that a non-detection of molecular features does not by itself distinguish a bare rock from an extremely thin atmosphere, since a 10^-10 mass-fraction hydrogen layer is photoevaporation-equivalent to a bare rock."],"supporting_citations":[{"why":"Supplies the PICASO radiative-transfer code used to generate all forward transmission models.","marker":"Batalha et al. (2019)"},{"why":"Provides the updated PICASO implementation that computes the model grid in this paper.","marker":"Mukherjee et al. (2023)"},{"why":"The photochem chemistry module that fixes chemical-equilibrium abundances and the Asplund et al. solar abundance scale.","marker":"Wogan et al. (2023)"},{"why":"The Resampled Opacities database containing the CH4, CO, CO2, H2O and other opacities used by PICASO.","marker":"Batalha et al. (2022)"},{"why":"Parameterizes the temperature-pressure profiles on which the model atmospheres are built.","marker":"Guillot (2010)"},{"why":"Establishes the precedent of fitting the NRS1/NRS2 offset before deriving atmospheric constraints from a scaled transit spectrum.","marker":"Moran & Stevenson et al. (2023)"},{"why":"Preceding COMPASS analysis whose ExoTiC-JEDI reduction and modeling approach this paper follows.","marker":"Alderson et al. (2024)"},{"why":"The Eureka! pipeline that produces the second independent reduction of the same data.","marker":"Bell et al. (2022)"},{"why":"Provides the stellar parameters, ephemeris, and fixed orbital elements used in the light-curve fits.","marker":"Luque et al. (2021)"}],"fun_headline_variants":["JWST: Super-Earth TOI-776b has no clear low-metallicity air","TOI-776b's flat spectrum rules out low metallicity","JWST NIRSpec: Super-Earth TOI-776b's air not metal-rich","JWST's flat spectrum on TOI-776b rejects solar-metal air"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument assumes that the ~65 ppm offset between the NRS1 and NRS2 detectors in Visit 2 is a fixed instrumental artifact that a single step function can remove before comparing the spectrum to physical models: if this offset actually changes with wavelength or time, the derived metallicity exclusion contours would be biased.","fun_headline_variants_meta":{"raw":{"variants":["JWST: Super-Earth TOI-776b has no clear low-metallicity air","TOI-776b's flat spectrum rules out low metallicity","JWST NIRSpec: Super-Earth TOI-776b's air not metal-rich","JWST's flat spectrum on TOI-776b rejects solar-metal air"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000834,"raw_usage":{"total_tokens":3723,"prompt_tokens":1109,"completion_tokens":2614,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":725,"completion_tokens_details":{"reasoning_tokens":2537}},"tokens_in":725,"tokens_out":2614,"duration_ms":18263,"temperature":1.0,"reasoning_tokens":2537,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T14:59:37.854978+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive check would be a third transit of TOI-776b observed at a different telescope roll angle, or with NIRSpec PRISM, so that no NRS1/NRS2 step correction is needed; if a methane band at ~3.3 µm or a CO2 band at ~4.3 µm then appears with an amplitude consistent with a <100× solar atmosphere, the paper's exclusion is wrong. A cheaper test is available now: re-fit the Visit 2 spectrum allowing the NRS1/NRS2 offset to be wavelength-dependent and see whether the 3σ excluded-metallicity boundary shifts by more than the stated visit-to-visit spread.","supporting_citations":[],"review_version":1}