{"id":"e28c5e75-8022-43b9-b188-0879244a3755","arxiv_id":"2509.02128","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"First JWST phase curves of TRAPPIST-1 b and c rule out thick, heat-redistributing atmospheres; b is likely airless, c may retain a tenuous O2 atmosphere.","lead":"JWST thermal phase curves of two temperate Earth-sized exoplanets, TRAPPIST-1 b and c, show almost no heat redistribution: b is likely bare rock, and c may have only a tenuous atmosphere. These are the first such measurements for temperate rocky planets and tightly constrain atmosphere retention around M dwarfs.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CO2+haze thick atmosphere remains viable; the 'no thick atmosphere' claim rests on a fine-tuning judgment, not a data-driven statistical test.","rationale":"The reader correctly identifies atmospheric model dependence as the weakest assumption, and explicitly mentions fine-tuned GCM haze parameters. However, the reader treats this as a handled caveat and accepts the paper's conclusion without requiring a more direct test. My concern is more pointed: the paper's headline-length claim ('no thick atmosphere') is contradicted by the paper's own admission that a 1 bar CO2 atmosphere with a high-altitude haze remains consistent with all data. The dismissal of this scenario relies on the judgment that the required haze properties are 'fine-tuned' and 'unlikely', but no quantitative prior or sensitivity analysis is presented. Furthermore, the comparison is performed on a compressed summary of the phase curve (three sinusoid-fit parameters), which may discard shape information that could distinguish the hazy atmosphere from an airless surface. A full light-curve fit with the GCM-generated phase curve would provide a decisive, data-driven test. Because this directly affects the central claim as stated in the title, I would condition acceptance on performing that test or softening the conclusion accordingly. The paper is otherwise excellent, with independent reductions and robust heat-redistribution limits, which is why I recommend conditional acceptance rather than rejection.","tokens_in":50517,"tokens_out":11258,"duration_ms":130417,"concrete_test":"Fit the GO 3077 light curve directly using the synthetic GCM phase curve for the 'Haze high' case (1 bar CO2 + high-altitude haze), including the same baseline, flare, transit/eclipse, and systematics model as the nominal Analysis #1. Compute the ΔBIC (or Bayesian evidence) between this model, the airless quasi-Lambertian model, and the fiducial sinusoidal model. If the Haze high model is disfavored by ΔBIC > 10, the fine-tuning dismissal is supported; if it is within ΔBIC < 10 of the airless model, the claim that no substantial atmosphere exists is not statistically justified and the wording should be softened to 'no atmosphere with efficient heat redistribution'.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central claim that TRAPPIST-1 b is unlikely to possess a substantial atmosphere requires ruling out the thick-atmosphere cases that survive the phase-curve parameter comparison. The paper explicitly identifies one such case—a 1 bar CO2 atmosphere with a high-altitude haze (the 'Haze high' 3D GCM case, Methods)—that matches both the 12.8/15 µm occultations and the observed dayside flux, nightside flux, and phase offset (Fig. 3). This case is dismissed not because the data exclude it, but because the haze parameters (fhaze, single-scattering albedo) are 'fine-tuned' and the required haze densities are 'unlikely'. No prior distribution or sensitivity analysis is provided to quantify this fine-tuning; only two grid points (Haze high/low) are shown. Moreover, the data/model comparison uses only three scalar parameters extracted from a first-order sinusoidal fit, discarding the full phase-curve shape, which could harbor additional discrimination between a hazy atmosphere and an airless surface. Thus, the title and abstract ('no thick atmosphere') are stronger than what the data alone support; the conclusion rests on a Bayesian prior that thick atmospheres require improbable microphysical parameters, not on a direct statistical rejection.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents JWST/MIRI F1500W thermal phase curves of TRAPPIST-1 b and c (Program GO 3077), combined with earlier eclipse observations, from four independent reductions and global MCMC analyses. The nominal analysis finds for b a dayside flux of 840±56 ppm, nightside flux of 39+55/−27 ppm, and phase offset −6.5±6.4 deg; for c, dayside flux 392+75/−63 ppm, nightside flux 62+60/−43 ppm, and phase offset 10+25/−22 deg. These values rule out efficient heat redistribution for both planets and are consistent with an airless or very tenuous-atmosphere interpretation. Forward atmospheric modeling with a 1.5D climate-photochemical model and 3D GCMs is used to show that most thick, greenhouse-efficient atmospheres are disfavored, while thin O2/N2 atmospheres and, for b, a fresh ultramafic airless surface remain viable. The paper concludes that TRAPPIST-1 b is unlikely to possess a substantial atmosphere, while TRAPPIST-1 c may have a tenuous O2-dominated atmosphere or an airless reflective surface.","tokens_in":50797,"tokens_out":5052,"duration_ms":61482,"significance":"If the conclusions hold, this is a landmark dataset: the first thermal phase curves of temperate terrestrial exoplanets, providing much stronger constraints on heat redistribution than eclipse depths alone. The four independent reductions, Gaussian-process treatment of correlated noise, and forward self-consistent atmospheric models are notable strengths, as is the explicit comparison of multiple modeling frameworks. The measurements firmly exclude full heat redistribution and thereby strongly disfavor the thick, well-mixed atmospheres previously discussed for these planets. However, the paper's strongest claim—that TRAPPIST-1 b has 'no thick atmosphere'—is not strictly a data-driven exclusion: the 1-bar CO2 'Haze high' 3D GCM case is acknowledged to match the phase-curve and eclipse observables and is set aside largely on a fine-tuning judgment. The significance is high, but the headline conclusion needs to be carefully qualified.","major_comments":[{"comment":"The central claim that TRAPPIST-1 b has no substantial atmosphere is not fully supported by the statistical comparison presented. The paper explicitly states that the CO2 1 bar + 'Haze high' GCM case remains consistent with the dayside flux, nightside flux, and phase offset (Fig. 3b), and is rejected only because it 'appear[s] to be very fine-tuned' and requires 'unlikely densities' of haze. No prior distribution, sensitivity scan over fhaze or single-scattering albedo, or Bayes factor is provided to quantify this fine-tuning; only two haze grid points are shown. A reader cannot independently assess whether the parameters are improbable enough to overturn the model-data agreement. Please either provide a quantitative model-comparison (e.g., a prior-based marginal likelihood over the haze parameters) or soften the abstract/title to 'no thick atmosphere with efficient heat redistribution'","section":"Main text, 'Comparing the two sets...'; Methods '3-D Global Climate Modeling'; Fig. 3b"},{"comment":"The atmospheric-model comparison compresses the observed phase curve into three scalar parameters—dayside flux, nightside flux, and phase offset—derived from a first-order sinusoidal model, and the paper itself cautions that the full phase-curve shape is 'not a robust basis for comparison.' This discards potentially discriminating information: a hazy thick atmosphere and an airless surface could differ in higher-order phase-curve morphology (e.g., the sharpness of the substellar peak, ingress/egress shape, or second harmonic) even when their first-order Fourier amplitudes and offsets agree. Given that the surviving 'Haze high' scenario is distinguished from the airless case mainly by such shape details, the decision not to use the full phase curve weakens the empirical basis for the 'no thick atmosphere' conclusion. Please state explicitly what information is lost, or include a fit to th","section":"Methods, Eqs. (1)-(4) and Fig. 3"},{"comment":"The 'Haze high' GCM case, which is the only thick atmosphere that survives the phase-curve comparison, is tuned to the eclipse data by construction: the single-scattering albedo is set to 0.5 and the incoming stellar flux is 'artificially decreas[ed]' specifically to make the dayside emission match the MIRI occultations. Consequently, the subsequent agreement with the phase curve is not an independent prediction, and the model's free parameters (fhaze and single-scattering albedo) are adjusted in the region of parameter space that maximizes agreement. The paper should clarify this circularity and, ideally, show how the phase-curve predictions vary as the tuning parameters are moved within a physically motivated range, rather than presenting the single surviving case as evidence against atmospheres.","section":"Methods, '3-D Global Climate Modeling of TRAPPIST-1b and c'"}],"minor_comments":[{"comment":"The entry '0.106+0.47/−0.50 R∗' appears to have a typo; presumably this should be 0.106+0.047/−0.050 R∗. Please correct.","section":"Table 1, planet b impact parameter"},{"comment":"The text defines γ(i,t) as the unocculted fraction, but the equations use Ω(i,t) for the occultation factor. Please make the notation consistent.","section":"Methods, Eqs. (1)-(3)"},{"comment":"Program numbers are inconsistent: the text refers to 'GO 2305' while the rest of the paper and Table S1 use GO 2304. Please verify.","section":"Methods, 'Data reduction and analysis (ED)'"},{"comment":"The caption says 'The data shown correspond to the nominal reduction (Analysis #1) by MG,' but the text says the analyses are consistent. Please specify which of the four reductions is plotted and why (binned at 60 minutes).","section":"Fig. 3 caption"}],"recommendation":"major_revision","confidential_remarks":"The paper is scientifically important and the photometric analysis appears careful, but the 'no thick atmosphere' headline is stronger than the evidence. The Haze high case is a genuine surviving model under the paper's own criteria, and the rejection relies on an unquantified fine-tuning argument. I would urge the editor to require a revised abstract and conclusion that either quantify the prior against hazy CO2 atmospheres or explicitly frame the result as ruling out efficient heat redistribution and disfavoring—but not excluding—thick atmospheres. The candidate planet 'i' is speculative but appropriately hedged and should not affect the scientific evaluation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—\n\nRead this one. It is the first JWST phase curve of temperate terrestrial planets, and the data are worth having: 59 hours of MIRI 15 µm photometry, four independent reductions (MG, ED, TJB, ZH) that agree within 1σ, and forward model comparisons. The core result is robust: TRAPPIST-1 b's nightside flux is consistent with zero, its phase offset is ~0°, and the airless, low-albedo surface model fits well. For c, the data are noisier and the interpretation is properly left open. I'd trust that neither planet has an atmosphere that efficiently moves heat to the nightside.\n\nThe soft spots, in order of weight. First, the title and abstract oversell. 'No thick atmosphere' is not what the data alone show. There remains a 1 bar CO2 atmosphere with a high-altitude haze ('Haze high') that matches both the 12.8/15 µm occultations and the measured dayside/nightside/offset. The paper dismisses it as 'fine-tuned' and 'unlikely,' but no prior, sensitivity analysis, or Bayesian evidence is given. That is a judgment, not a statistical rejection. I'd ask for a sensitivity scan over fhaze and single-scattering albedo, or at least a quantitative argument for why the required haze mass is physically implausible. Second, the model comparison uses only three scalars from a first-order sinusoid; the full phase curve shape is discarded, and the paper even warns against using shape—but that is where a hazy atmosphere and an airless surface might differ most. Not fatal, but worth a paragraph. Third, the 15% data cut (initial ramp, flares) and the handful of 'mini-flares' modeled to kill red noise—fine, but the error rescaling is doing a lot of work. The TJB no-GP fit shows how much the conclusions depend on the noise model.\n\nThe speculative planet i and the PPO detection are handled with appropriate caution: both are below decisive significance, and they do not affect the main conclusions. The surface modeling (ultramafic, space weathering) is a nice bonus and is properly hedged.\n\nBottom line: this is a real step forward, likely correct in its main claim, but the wording is stronger than the evidence. A serious referee should send it back with a request to soften the title/abstract and to address the fine-tuning question quantitatively. Still cite-worthy and worth putting in front of the group.","headline":"First JWST phase curves of temperate terrestrial planets; main result is solid, but 'no thick atmosphere' is an overstatement—the fine-tuned hazy CO2 case survives.","tokens_in":51467,"tokens_out":3039,"would_cite":true,"duration_ms":35021,"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 thermal phase curves of TRAPPIST-1 b and c rule out efficient heat redistribution, making thick atmospheres unlikely on either temperate rocky planet.","keywords":["exoplanets","TRAPPIST-1","thermal phase curves","JWST/MIRI","terrestrial planets","heat redistribution","secondary eclipses","ultracool dwarfs"],"falsifier":"A higher-precision or 12.8 µm phase curve of TRAPPIST-1 b that detects nightside emission above roughly 100 ppm, or a phase offset beyond about 15 degrees, would contradict the airless conclusion; conversely, detecting the O3 or H2O features predicted at 50–80 ppm in b's transmission spectrum would support the surviving thin-atmosphere scenarios. For c, the planned 12.8 µm eclipse measurement can decide between a bare reflective surface and a thin O2 atmosphere.","tokens_in":50415,"feed_emoji":"🪨","tokens_out":8966,"duration_ms":87310,"temperature":0.7,"pith_summary":"This paper presents the first JWST thermal phase curves of temperate terrestrial exoplanets: 59 hours of continuous 15 µm MIRI photometry of TRAPPIST-1, capturing full-orbit emission variations of planets b and c. Because an atmosphere transports heat from the perpetually sunlit dayside to the nightside, the day-night brightness contrast and the phase of the emission peak directly diagnose atmosphere presence. The measured curves show a hot dayside for b (490 K), nightside emission consistent with zero, and no phase offset; c shows a cooler dayside (369 K) and nightside emission statistically indistinguishable from b's. Efficient heat redistribution is firmly ruled out for b and appears unlikely for c, and atmospheric models with surface pressures of 1 bar or more with efficient greenhouse effects are strongly disfavored for both. The paper concludes that TRAPPIST-1 b almost certainly has no substantial atmosphere, while TRAPPIST-1 c may host a thin, greenhouse-poor O2-dominated atmosphere or be airless with a more reflective surface.","feed_headline":"JWST phase curves rule out thick air on TRAPPIST-1 b and c","feed_subtitle":"Full-orbit thermal maps show heat barely reaches the nightside — the bare-rock signature on both temperate planets.","key_machinery":"The thermal phase curve is the load-bearing observable: the planet-to-star flux measured continuously across the orbit, decomposed into a dayside term, a nightside term, and a longitudinal phase offset δ. An atmosphere redistributes heat, producing a nonzero nightside term and a peak that shifts away from the substellar point; an airless body emits from the dayside only, peaking at the substellar point with a quasi-Lambertian limb profile. The paper compares the measured parameters to (1) a two-column day-night climate-photochemical model generating synthetic phase curves for many atmospheric compositions and pressures, (2) 3D global climate model simulations of the surviving scenarios, incl","core_discovery":"The central discovery is that day-night heat redistribution — the observational signature of an atmosphere — is essentially zero for TRAPPIST-1 b and very low for TRAPPIST-1 c. In the nominal model, b's nightside flux is 39+55/−27 ppm against a dayside of 840±56 ppm, with phase offset −6.5±6.4 degrees; full heat redistribution has probability below 0.0006%. For c, the nightside is 62+60/−43 ppm against a 392+75/−63 ppm dayside. A model assuming both planets are bare rocks fits better (Bayes factor >1152). Previously viable scenarios for b — N2 ≥1 bar, thick O2, CO2-rich hazy atmospheres — are discarded because they would deliver measurable nightside flux. For c, steam atmospheres up to 10 ba","pith_inferences":["If TRAPPIST-1 b is truly airless, its modestly dark, fresh ultramafic-like surface offers a calibration point for space-weathering rates on M-dwarf planets, which the paper suggests may be faster than in the Solar System.","The phase-curve diagnostic used here should transfer to other temperate M-dwarf rocky planets: a single full-orbit 15 µm observation can break the eclipse-depth degeneracy between airless, hazy-thick, and thin-transparent atmospheres.","Because the two-column model may under-resolve the sharp substellar temperature peak for thin atmospheres, the 'no substantial atmosphere' claim for b is strongest for pressures ≳0.1 bar; extremely tenuous (~0.01 bar) atmospheres are excluded mainly by stability arguments (nightside CO2 collapse) rather than by the phase curve alone.","The suggestion that c may be airless-but-more-reflective implies that 'airless' is not a single observable state; surface albedo variation could complicate population-level interpretations of rocky exoplanet emission."],"forward_implications":["TRAPPIST-1 b, very likely bare rock, can serve as a clean stellar-contamination reference for transmission spectroscopy of the outer TRAPPIST-1 planets, provided the small residual features predicted by surviving thin-atmosphere models are checked.","TRAPPIST-1 c's ambiguity between an airless reflective surface and a thin O2-dominated atmosphere is directly testable: planned 12.8 µm eclipse observations should separate the two.","The divergent states of b and c, despite similar sizes and compositions, constrain how quickly M-dwarf planets lose secondary atmospheres and suggest atmospheric loss is not uniform across the system.","Phase curves, not just eclipse depths, are validated as the decisive measurement for distinguishing bare rocks from hazy or thin atmospheres on temperate rocky exoplanets.","The tentative 0.2 R⊕ candidate planet i, seen at ~4σ in one analysis but below 3σ in others, is not robustly detected and requires confirmation."],"supporting_citations":[{"why":"Supplies the first 15 µm dayside eclipse measurement of TRAPPIST-1 b that the combined phase-curve analysis extends.","marker":"[4]"},{"why":"Provides the 12.8 µm eclipse data whose airless-vs-hazy-CO2 degeneracy the phase curve now breaks.","marker":"[6]"},{"why":"Supplies the 15 µm eclipse measurement of TRAPPIST-1 c and the atmospheric scenarios it allowed.","marker":"[7]"},{"why":"Introduces the two-column day-night climate-photochemical model used to generate synthetic atmospheric phase curves.","marker":"[8]"},{"why":"Provides the steam-atmosphere GCM simulations for TRAPPIST-1 c that the phase curve disfavors.","marker":"[9]"},{"why":"Supplies the sinusoidal day/night phase-curve model with offset used in the nominal analysis.","marker":"[16]"},{"why":"Supplies the quasi-Lambertian airless phase-curve model used in the alternative fits.","marker":"[17]"},{"why":"Provides the system ephemerides, masses, radii, and priors for the MCMC fits.","marker":"[19]"},{"why":"Provides the 3D global climate model framework used for the climate simulations of TRAPPIST-1 b and c.","marker":"[26]"},{"why":"Supplies the bare-rock thermal model with surface materials and space weathering used for b's surface interpretation.","marker":"[31]"}],"fun_headline_variants":["JWST: TRAPPIST-1 b and c are bare rocks, no thick air","No thick atmosphere for TRAPPIST-1 b and c, JWST shows","TRAPPIST-1 b and c lose atmosphere fight: JWST finds bare rock","JWST phase curves: TRAPPIST-1 b and c have no thick air","TRAPPIST-1 b and c: JWST sees bare rock, no heat redistribution"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The conclusion rests on the atmospheric models predicting how much heat even a thin or hazy atmosphere would move to the nightside; if such an atmosphere transports less heat than modeled, or the models miss the sharp dayside hot spot, a thin atmosphere could remain hidden in the data.","fun_headline_variants_meta":{"raw":{"variants":["JWST: TRAPPIST-1 b and c are bare rocks, no thick air","No thick atmosphere for TRAPPIST-1 b and c, JWST shows","TRAPPIST-1 b and c lose atmosphere fight: JWST finds bare rock","JWST phase curves: TRAPPIST-1 b and c have no thick air","TRAPPIST-1 b and c: JWST sees bare rock, no heat redistribution"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000995,"raw_usage":{"total_tokens":4094,"prompt_tokens":831,"completion_tokens":3263,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":575,"completion_tokens_details":{"reasoning_tokens":3151}},"tokens_in":575,"tokens_out":3263,"duration_ms":23254,"temperature":1.0,"reasoning_tokens":3151,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-05T11:50:36.639753+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A higher-precision or 12.8 µm phase curve of TRAPPIST-1 b that detects nightside emission above roughly 100 ppm, or a phase offset beyond about 15 degrees, would contradict the airless conclusion; conversely, detecting the O3 or H2O features predicted at 50–80 ppm in b's transmission spectrum would support the surviving thin-atmosphere scenarios. For c, the planned 12.8 µm eclipse measurement can decide between a bare reflective surface and a thin O2 atmosphere.","supporting_citations":[],"review_version":1}