{"id":"d7004ee4-1033-4eb0-b9c2-af77a7b65336","arxiv_id":"2607.14326","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":7,"one_line_summary":"LHS 1140b shows 1.24% metastable helium absorption during a 2024 transit, absent in 2025, pointing to variable escape from a helium-dominated upper atmosphere.","lead":"Astronomers report detecting helium escaping from the atmosphere of LHS 1140b, a rocky habitable-zone planet 15 parsecs away, in 2024 transit spectra—but not in a 2025 repeat. The signal suggests time-variable atmospheric escape and a hydrogen-poor, helium-rich upper atmosphere, with implications for whether small planets can hold onto atmospheres.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The hydrogen-depleted claim is not robust to unobserved EUV SED shape; the paper's 10x flux test scales X-ray too and does not span the plausible EUV/X-ray ratio.","rationale":"The reader's conditional verdict and the identified weakest assumption match my own reading: the detection is strong, but the helium-dominated, hydrogen-depleted composition claim depends on the p-winds retrieval and the assumed XUV SED. My analysis sharpens the reader's point: the paper's sensitivity test scales the entire SED, including X-ray, by 10, so it does not test the crucial unconstrained quantity—the EUV/X-ray ratio. The conclusion could change if LHS 1140's true EUV is higher relative to its X-ray flux than in the GJ 1132/GJ 699 analogues. The stated stability threshold for the GJ699 SED adds a second, more technical reason to treat the H:He lower bound cautiously. Because the central detection and the energy-limited escape-rate consistency are not affected by this concern, the verdict should remain CONDITIONAL rather than being upgraded or rejected. The concrete test proposed would settle whether the compositional claim is robust or should be downgraded to 'helium-dominated outflow' only under the assumed SED family.","tokens_in":27225,"tokens_out":8962,"duration_ms":101049,"concrete_test":"Using the archived WINERED 2024 data and p-winds, construct a set of LHS 1140 SEDs that keep the measured 0.25–2 keV X-ray flux fixed but vary the EUV flux at 100–911 Å independently by factors of 0.1, 0.3, 1, 3, and 10 relative to the GJ 1132 and GJ 699 templates (alternatively, use the two-temperature APEC model with different emission-measure ratios). Re-run the MCMC/grid retrieval for each SED and record the posterior on H:He and Mdot. If any SED consistent with the X-ray spectrum and stellar activity indicators yields H:He ≥ 0.01 with acceptable fits, the 'hydrogen-depleted' conclusion in the abstract is not settled; if all such SEDs yield H:He ≲ 0.01, the claim survives this test.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The 2024 detection (1.24+0.22/-0.23% at 10,833 Å, confirmed by an independent re-reduction, with checks against telluric, flare, and transit-light-source contamination) is convincing, and I do not dispute it. The load-bearing step is the extrapolation to a helium-dominated, hydrogen-depleted outflow (H:He ≈ 1e-3). That conclusion is produced by p-winds, which requires the stellar XUV SED as input. The SED for LHS 1140 is not measured; it is constructed by scaling the GJ 1132 and GJ 699 Mega-MUSCLES SEDs to the X-ray flux recovered from PSF-deblended XMM data (A=0.59 and 0.056; Materials and Methods, 'X-ray observations and analysis'). The only SED sensitivity test in the paper multiplies the entire scaled SED by a factor of 10 (Retrieving atmospheric properties; fig. 5). This changes the X-ray flux as well and does not explore independent variation of the unobserved EUV (≈100–911 Å) relative to X-ray. Because metastable helium production requires EUV photons to ionize He and H, a different EUV/X-ray ratio could relax the H:He constraint. Additionally, the posterior for the GJ699-based retrieval sits at the model's stated stability boundary (H:He ≥ 1e-3), so the lower envelope of that posterior may be a numerical artifact rather than data-driven. The escape detection itself is secure, but the hydrogen-depleted interpretation is not yet robust to SED-shape uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"Cherubim et al. report ground-based WINERED/Magellan transit spectroscopy of the M-dwarf system LHS 1140. They detect a 1.24 ± 0.22% metastable helium absorption feature at 10,833 Å during the 2024 transit of the habitable-zone rocky planet LHS 1140b, confirm the detection with an independent re-reduction, and do not detect the feature in a 2025 transit of the same planet or in the transit of the inner planet LHS 1140c. Using the p-winds code, they retrieve mass-loss rates of 2–4×10^8 g/s and H:He ratios of ~1×10^-3, leading them to interpret the upper atmosphere as helium-dominated and hydrogen-depleted. They also discuss variable escape, a leading/trailing tail, and consistency with the cosmic shoreline.","tokens_in":27662,"tokens_out":8611,"duration_ms":89234,"significance":"If the compositional interpretation holds, this is a landmark observation: direct evidence of escaping helium from a rocky, habitable-zone exoplanet, with implications for atmospheric retention, fractionation, and the cosmic shoreline. The paper's concrete strengths are the robust 2024 detection (confirmed by an independent pipeline), the careful exclusion of telluric, flare, and transit-light-source contamination, and the public data/code archive. The central weakness is that the helium-dominated/hydrogen-depleted conclusion relies on a p-winds retrieval whose input XUV SED is not directly measured; the only sensitivity test scales the entire SED, leaving the EUV/X-ray ratio unvaried. This makes the compositional claim less robust than the escape detection itself.","major_comments":[{"comment":"The helium-dominated conclusion rests on the p-winds retrieval, whose input XUV SED is constructed by scaling the GJ 1132 and GJ 699 Mega-MUSCLES SEDs to the X-ray flux from a blended XMM detection (A=0.59 and 0.056). The EUV band (≈100–911 Å) that drives He ionization is not observed for LHS 1140 and is assumed to follow the adopted template ratio. The only sensitivity test multiplies the entire SED by 10, preserving the EUV/X-ray ratio. Because the H:He constraint is set by hydrogen attenuation of EUV, an independent EUV variation could relax the constraint. I request retrievals that vary the EUV segment independently (e.g., 0.1–10×) while holding the X-ray normalization fixed, or a formal justification that the adopted templates bracket the plausible SED space. In addition, the GJ 699 run's stability boundary is H:He ≥ 1×10^-3, and the posterior in Fig. S6 is truncated near that bound","section":"Materials and Methods, 'X-ray observations and analysis' and 'Retrieving atmospheric properties'; Figs. S4, 5, S5, S6"},{"comment":"The abstract states that the 2024 detection and 2025 non-detection indicate time-variable atmospheric escape. However, the paper does not quantify the probability of the 2025 non-detection under the 2024 best-fit model: Fig. 4 shows example models with low XUV flux or high T_wind falling below the 0.6% limit, but no posterior predictive distribution is computed. No X-ray observation of LHS 1140 between the two epochs is presented. To support 'time-variable atmospheric escape' as a measured result rather than a plausible speculation, the authors should (a) compute the fraction of posterior predictive realizations with XUV and T_wind varied over the adopted ranges that yield absorption below the 2025 limit, or (b) soften the abstract and conclusion to state that the non-detection could be due to variable escape, stellar activity, or other effects.","section":"Variable helium escape; Fig. 4"}],"minor_comments":[{"comment":"The quoted Galactic neutral hydrogen column density reads NH = 7.4^{+5.1}_{-4.9}×10^-8 cm^-2; the exponent should likely be 10^18, based on the text and the allowed range 0–4.6×10^18 cm^-2.","section":"Table S1"},{"comment":"The text says 'thermal broadening dominates over instrumental and natural broadening' but the observed FWHM is 23.9 km/s, far larger than thermal broadening for helium at T≈5000–5800 K. The line width is plausibly dominated by the radial velocity gradient in the outflow; please rephrase to avoid a misleading statement.","section":"Spectroscopic observations of LHS 1140"},{"comment":"Typo: '1σ and 3σ derivations' should be 'deviations'. Also, the caption's description of the 10× XUV flux points ('Data points with dotted error bars') could state explicitly that this is the only SED-sensitivity test.","section":"Fig. 5 caption"},{"comment":"The abstract says 'other volatile species trapped at lower altitudes'; the text discusses cold trapping (e.g., water) and crossover-mass dragging. 'Cold-trapped' or 'concentrated at lower altitudes' would be more precise, since the mechanism differs for different species.","section":"Atmospheric composition"}],"recommendation":"major_revision","confidential_remarks":"The escape detection appears secure and the contamination checks are convincing. The load-bearing issue is the SED-dependent H:He retrieval; the requested EUV-variation runs are straightforward and should be feasible. The authors' previous theoretical work predicted helium dominance, but this observation is an independent test rather than a restatement, so I see no circularity problem. The paper fits the journal's scope well."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe punchline: the 2024 helium detection is solid and, if it stands, it is the first direct evidence of ongoing atmospheric escape from a rocky planet in the habitable zone. The more dramatic interpretation attached to it — a helium-dominated, hydrogen-depleted upper atmosphere — is much more conditional than the abstract implies, and the paper's own retrieval makes that clear.\n\nWhat's genuinely new: a ~5.6σ metastable helium absorption at 1.24% during the transit of LHS 1140b, confirmed by an independent re-reduction, with careful checks against telluric lines, stellar flares, and the transit light source effect. The non-detection in 2025 and the non-detection for LHS 1140c add internal consistency and strengthen the case that the signal is really planetary escape. The line profile (FWHM, amplitude ratio) matches the helium triplet. That is a real first.\n\nThe soft spots are where you'd expect them. The hydrogen-depleted conclusion comes from p-winds, which needs the stellar SED as input. LHS 1140's EUV is unobserved; the SED is built by scaling GJ 1132 and GJ 699 templates to XMM X-ray fluxes, with ~10% residual contamination in the aperture. The only sensitivity test multiplies the whole SED by 10, which does not explore changes in the EUV/X-ray ratio. That is a genuine gap: metastable helium production depends on EUV photons in a way that X-ray alone doesn't capture. On top of that, the GJ 699 posterior sits right at the model's H:He ≥ 1e-3 stability boundary, so the lower envelope of that posterior may be numerical flooring rather than data. The leading/trailing tail regions were picked by eye, and the 2025 variability explanation has no contemporaneous stellar baseline, but those are secondary.\n\nI don't think these issues kill the paper. The escape detection itself is convincing, and the retrieved escape rates (2–4e8 g/s) are consistent with energy-limited values. The helium-dominated composition is a reasonable hypothesis, supported by prior models from the same group that predicted it — that's a test, not a restatement. But the abstract should be careful. I'd rather see the strong claim labeled as model-dependent until someone does a proper SED-shape sensitivity run.\n\nFor the reading group: yes. For citation: yes, at least for the detection. For review: definitely send to a serious referee. The referee should push on the SED sensitivity and the H:He boundary behavior, but the core result deserves to be out in the open.","headline":"A convincing first detection of helium escape from a rocky habitable-zone planet, but the 'hydrogen-depleted' interpretation is more conditional than the abstract suggests.","tokens_in":28176,"tokens_out":2155,"would_cite":true,"duration_ms":22974,"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":"This paper reports the detection of metastable helium escaping from the rocky habitable-zone exoplanet LHS 1140b, with absorption present in 2024 and absent in 2025, and interprets the outflow as helium-dominated and hydrogen-depleted.","keywords":["metastable helium","atmospheric escape","rocky exoplanet","LHS 1140b","habitable zone","transmission spectroscopy","XUV-driven escape","helium-dominated atmosphere"],"falsifier":"A direct EUV observation of LHS 1140 (e.g., 100–911 Å) showing a flux more than about ten times the scaled analogue-star values would falsify the hydrogen-depleted outflow, because hydrogen would then absorb the ionizing flux and suppress metastable helium; conversely, a 2025-style non-detection coinciding with a measured drop in stellar XUV would support the variability interpretation.","tokens_in":27146,"feed_emoji":"🪐","tokens_out":6878,"duration_ms":67612,"temperature":0.7,"pith_summary":"LHS 1140b, a rocky planet in the habitable zone of a nearby M dwarf, shows absorption from metastable helium during its 2024 transit, making it one of the first rocky exoplanets with a detected atmospheric escape signal. The same feature is absent in 2025, so the escape is time-variable. Modeling the line profile with a one-dimensional isothermal outflow yields a mass-loss rate of roughly 2–4×10^8 g/s and a hydrogen-to-helium ratio near 10^-3, implying the escaping upper atmosphere is dominated by helium and depleted in hydrogen. The authors argue this matches predictions of escape-driven fractionation, in which lighter hydrogen is preferentially lost and heavier volatiles are trapped below. If correct, the result shows that temperate rocky planets can retain detectable outflows and that single-epoch non-detections do not prove an atmosphere is absent.","feed_headline":"Rocky habitable-zone exoplanet is losing helium","feed_subtitle":"Helium detected escaping a rocky habitable-zone planet in 2024, gone in 2025 — sign of a hydrogen-poor atmosphere.","key_machinery":"The central probe is the metastable helium triplet at 10,833 Å, a near-infrared absorption feature that traces the extended, escaping upper atmosphere of a planet; the two red components are blended at the instrument resolution, and their amplitude ratio (measured 6.7, expected 8) confirms the line identification. The physical model is a one-dimensional, isothermal Parker wind, which converts an assumed stellar XUV spectrum, an escape rate, a temperature, and an H:He ratio into a predicted helium line profile. The key constraint is ionization balance: at H:He ≳0.01, neutral hydrogen absorbs the stellar flux below 911 Å, starving the helium ionization–recombination cycle that populates metast","core_discovery":"During a 6.5-hour observation on 2024 September 23, the authors detected the metastable helium triplet at 10,833 Å in the transmission spectrum of LHS 1140b, with a blended-line absorption depth of 1.24% and a Doppler shift consistent with the planet's motion, plus a leading tail of absorption before ingress. The inferred opaque radius is 1.52 planetary radii. A re-observation in 2025 showed no helium absorption, with a 0.6% detection limit, indicating the escape signal varies between epochs. Using a spherically symmetric, isothermal Parker-wind model, the authors retrieve an atmospheric mass-loss rate of 2.0–4.2×10^8 g/s, an outflow temperature near 5000–5900 K, and a hydrogen-to-helium rat","pith_inferences":["If the hydrogen depletion holds, the same technique could be used to search for helium outflows around other habitable-zone M-dwarf planets, effectively using helium as a proxy for an atmosphere that might otherwise be invisible.","The year-to-year variability could reflect an activity cycle of the host star rather than intrinsic atmospheric changes; monitoring over several years would distinguish a stellar-cycle modulation from a stochastic escape event.","A direct measurement of the star's EUV spectrum would be the sharpest test of the hydrogen-depletion claim; for now, the conclusion rests on scaled spectra of analogue stars.","The crossover-mass argument implies that any water in the planet's interior would be cold-trapped and retained, so a helium-dominated outflow is compatible with a water-rich bulk composition—connecting this detection to the water-world interpretation of LHS 1140b's density."],"forward_implications":["Ground-based monitoring of metastable helium can reveal atmospheric escape from temperate rocky planets, not just hot gas giants, and can do so from the ground.","A single non-detection of helium is not evidence that a rocky planet is airless; the 2024–2025 variation shows escape can switch on and off.","If helium-dominated upper atmospheres are common among temperate rocky planets, transmission spectra will show small scale heights and muted hydrogen features, making such atmospheres harder to detect in the optical/UV.","The inferred escape rate (~2–4×10^8 g/s) is far below the ~5×10^9 g/s threshold that would strip a 1.5% primordial atmosphere over the system's age, so LHS 1140b's atmosphere can persist.","The non-detection for LHS 1140c, which sits on the other side of the cosmic shoreline, supports the idea that irradiation drives the divide between airless and atmosphere-bearing rocky planets."],"fun_headline_variants":["Rocky exoplanet's helium escape vanishes a year later","Helium seen escaping habitable-zone rocky world—then gone","Variable helium outflow from rocky exoplanet LHS 1140b","Rocky planet's helium escape detected, then absent in recheck","Time-variable helium escape on rocky habitable-zone planet"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The result depends on the unmeasured extreme-ultraviolet spectrum of LHS 1140 being accurately represented by scaling two similar M dwarfs' spectra to the star's X-ray flux; if the true EUV flux or its spectral shape differs, the retrieved hydrogen-to-helium ratio and the helium-dominated conclusion could change.","fun_headline_variants_meta":{"raw":{"variants":["Rocky exoplanet's helium escape vanishes a year later","Helium seen escaping habitable-zone rocky world—then gone","Variable helium outflow from rocky exoplanet LHS 1140b","Rocky planet's helium escape detected, then absent in recheck","Time-variable helium escape on rocky habitable-zone planet"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000207,"raw_usage":{"total_tokens":1222,"prompt_tokens":715,"completion_tokens":507,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":459,"completion_tokens_details":{"reasoning_tokens":431}},"tokens_in":459,"tokens_out":507,"duration_ms":5232,"temperature":1.0,"reasoning_tokens":431,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T02:25:37.127584+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A direct EUV observation of LHS 1140 (e.g., 100–911 Å) showing a flux more than about ten times the scaled analogue-star values would falsify the hydrogen-depleted outflow, because hydrogen would then absorb the ionizing flux and suppress metastable helium; conversely, a 2025-style non-detection coinciding with a measured drop in stellar XUV would support the variability interpretation.","supporting_citations":[],"review_version":1}