{"id":"9965a611-b2b4-46b4-9726-6d7c965a0cb5","arxiv_id":"2505.20588","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":6,"one_line_summary":"Continuous JWST observations of WASP-107b reveal metastable helium absorption beginning 1.5 hours before ingress, evidence of an extended ellipsoidal thermosphere, with spot-corrected water abundance log10 H2O = -2.5 ± 0.6.","lead":"Using JWST NIRISS-SOSS observations of the exoplanet WASP-107b, the authors report continuous helium absorption that starts about 1.5 hours before the planet crosses its host star and lasts until after transit. This is the first space-based detection of pre-transit helium escape and reveals an outflow stretching tens of planetary radii.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Pre-transit helium detection may be susceptible to a time-dependent wavelength calibration drift; the single-visit dataset cannot rule this out, so the central claim needs a trace-stability check.","rationale":"The reader's weakest assumption correctly identifies the lack of an independent epoch and the possibility of instrume/ntal or stellar systematics shaping the pre-transit signal. My concern sharpens this into a specific, testable mechanism: a time-dependent wavelength calibration drift that differentially affects the narrow helium bin relative to the broad continuum bands. This is the single most load-bearing issue because, if it lands, the central claim of first detection of pre-transit helium absorption from space collapses entirely. The two-pipeline agreement does not mitigate it, because both pipelines share the same raw data and are susceptible to the same drift. The paper's own arguments (flat far bins, stellar helium upper limit) do not directly address a dispersion-axis shift. The proposed test is feasible with existing data and would distinguish a genuine extended thermosphere from a systematic. Because the detection is otherwise supported by a clear in-transit and post-transit signal and the consistency with previous ground-based amplitudes, the reader's CONDITIONAL verdict remains appropriate; my concern does not move it but reinforces the need for the trace-stability check before acceptance as a firm discovery.","tokens_in":48593,"tokens_out":5487,"duration_ms":64188,"concrete_test":"Using the same NIRISS-SOSS visit, measure the centroid of the order-1 spectral trace (or the cross-correlation shift of each frame against a high-S/N template) as a function of time, and test for a monotonic drift during the 1.5 h pre-transit window. Then re-extract the helium light curve after registering all frames to a common wavelength solution (shifting by the measured trace position). If the pre-transit slope in the helium bin persists after registration, the signal is robust; if it weakens or disappears, it is a wavelength-calibration artifact. As a secondary check, repeat the same continuum-subtracted light curve on a narrow stellar photospheric line of similar width: a matching pre-transit slope would indicate a common systematic.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline claim of pre-transit helium absorption rests on the differential light curve constructed in Section 2.3 by subtracting the average of two broad continuum bands (1070–1080 and 1090–1100 nm) from the narrow bin at 1.0835 μm. This measurement is vulnerable to a time-varying wavelength solution: NIRISS-SOSS time series are known to exhibit small trace shifts along the dispersion axis. A drift of even a fraction of a pixel would alter the flux in the narrow helium bin, where the spectral slope across the line is steep and changes sign, while leaving the broad continuum bands nearly unaffected. Such a systematic would produce a smooth pre-transit slope exactly in the helium bin, mimicking the reported 17σ excess. The two independent reduction pipelines (exoTEDRF and NAMELESS) operate on the same raw frames, so a common trace drift would survive both reductions and would not be flagged by their mutual agreement. The argument that the 'furthest bins' show flat light curves does not exclude this mechanism, because the drift affects the line bin far more than bins displaced by ~1 nm, and the continuum bands are even wider. The paper reports no check of the trace centroid or of the temporal stability of the wavelength solution, and the observation is a single 6.2-hour visit with no second epoch for an independent control. If a monotonic drift occurred during the pre-transit window, the central claim of planetary absorption would be an artifact.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyzes a single JWST NIRISS-SOSS time-series observation of WASP-107b (678 integrations spanning 6.2 hours) and reports: (i) continuous metastable helium absorption before, during, and after transit, claimed as the first pre-transit helium detection from space at 17 sigma; (ii) an ellipsoidal thermosphere model produced with the EvE code that reproduces the light curve with an elongation of 10-18 planetary radii ahead of the planet; (iii) a broad 0.6-2.8 micron transmission spectrum with water at log10 H2O = -2.5 +/- 0.6, a blueward slope attributed to unocculted stellar spots rather than haze, and a 2-sigma upper limit on potassium; and (iv) formation and evolution inferences, including a possible recent migration scenario. The data are reduced with two independent pipelines (exoTEDRF and NAMELESS) whose spectra agree at the 0.76-sigma level, and retrievals are run with SCARLET, petitRADTRANS, Pyrat Bay, and TauREx.","tokens_in":48838,"tokens_out":6446,"duration_ms":71210,"significance":"If the pre-transit helium signal is real, this is a significant result: it would be the first space-based detection of continuous helium absorption before transit, a direct probe of an extended thermosphere on both sides of WASP-107b, and it would demonstrate the value of long-baseline JWST time series for escape studies. The paper has clear strengths: two reduction pipelines, careful discussion of known systematics, sensitivity tests around the retrievals, an explicit treatment of the transit light source effect, and an unusually candid statement of the degeneracies in the helium model. However, the central claim rests on a differential light curve from a single visit, and the paper does not perform the trace-stability check needed to rule out a time-varying wavelength solution. The stated onset time of the pre-transit absorption is also inconsistent with the observing window. These issues are addressable with the existing data, so the result is defensible in principle, but the manuscript needs revision before the headline claim can be accepted.","major_comments":[{"comment":"The helium light curve is constructed by subtracting the average of two broad continuum bands (1070-1080 nm and 1090-1100 nm) from the narrow 1.0835 micron line bin. NIRISS-SOSS time series can exhibit slow trace shifts along the dispersion axis, and a monotonic shift of even a fraction of a pixel would create a smooth ramp in the line bin, where the spectral slope across the helium feature is steep and changes sign, while leaving the broad continuum bands nearly unaffected. Because exoTEDRF and NAMELESS reduce the same raw frames, their agreement does not rule out a common wavelength drift, and the flatness of the bins immediately adjacent to the line is not a discriminating test because the effect is much stronger in the line bin. The manuscript does not report a trace-centroid or wavelength-solution stability check, and the observation is a single 6.2-hour visit with no second-epoch control. Please measure the trace position as a function of time (for example, from the centroid of the cross-dispersion profile or from cross-correlation of successive extracted spectra) and quantify the wavelength shift required to reproduce the observed pre-transit ramp; this check is necessary to validate the 17-sigma claim.","section":"Section 2.3 and Figure 5"},{"comment":"The abstract and Section 2.3 state that the pre-transit absorption begins approximately 1.5 hours before the planet's ingress. The observation includes 2.33 hours of pre-transit baseline, and with a transit duration of 2.75 hours, ingress occurs at T0 - 1.375 hours; 1.5 hours before ingress is therefore T0 - 2.875 hours, which is before the start of the observation at T0 - 2.33 hours. The onset cannot have been measured at that time. The sentence either contains a typo (for example, 1.5 hours before mid-transit) or the absorption is already present at the start of the observation; the text should be corrected and the actual observable onset time stated. Relatedly, Section 2.3 defines the helium baseline as T0 - 3 hours, which is also outside the observed window; this definition should be clarified.","section":"Abstract and Section 2.3"},{"comment":"The paper explicitly acknowledges in Section 2.5 that the mass-loss rate and upper-atmosphere temperature are not definitively constrained by the helium data, that the H/He ratio is fixed to solar, and that no definitive conclusions about the upper-atmosphere density structure can be drawn. Nevertheless, Section 2.8 uses the thermospheric mass-loss rate of order 1-10 Earth masses per Gyr, derived from this same model, to estimate a primordial metallicity of 4x stellar and to argue that in-situ formation is improbable. This propagates an unquantified model degeneracy into a formation conclusion. Please either remove or soften the formation inference, or propagate the full degeneracy in mass-loss, temperature, and H/He ratio through the calculation so that the evolutionary claims do not inherit free parameters that are explicitly stated to be indeterminate.","section":"Sections 2.4, 2.5, and 2.8"}],"minor_comments":[{"comment":"The significance levels quoted in the abstract and Section 1 (17 sigma, 19 sigma, 36 sigma) are never defined; please specify the statistic used (for example, weighted mean of the excess absorption divided by its uncertainty) and whether red noise or time-correlated systematics are included.","section":"Throughout"},{"comment":"The sentence describing the pre-transit increase in absorption is duplicated in the main text; one occurrence should be removed.","section":"Section 1"},{"comment":"The caption statement that the helium line 'maintains baseline flux values until T0 < 3 hours' is unclear; it should state the actual time range used for the baseline.","section":"Figure 5"},{"comment":"The typesetting of target and code names is inconsistent (for example, 'W ASP-107 b', 'T auREx', 'Pyrat Bay'); this should be harmonized before publication.","section":"Throughout"},{"comment":"The paper states that data and code will be available upon request; for a benchmark result of this kind, depositing the reduced spectra and light curves in a public repository would be more appropriate.","section":"Data and code availability"}],"recommendation":"major_revision","confidential_remarks":"The key question for this paper is whether the pre-transit ramp survives a trace-stability analysis. The two-pipeline agreement is valuable but does not address a common time-varying wavelength solution. I would ask for the trace-centroid diagnostic and a significance calculation that accounts for any measured drift before further consideration. I would also ask the authors to correct the onset-time inconsistency and to decouple the formation discussion from the admittedly degenerate mass-loss rate. The broad transmission-spectrum and retrieval results appear solid and are a useful contribution regardless of the helium timing claim."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe genuinely new thing here is a continuous JWST NIRISS-SOSS light curve showing helium absorption starting about 1.5 hours before ingress, not just the usual post-transit tail. If that pre-transit signal is real, it is a benchmark for atmospheric escape studies: short-baseline ground-based helium observations have been underestimating thermospheric extent. The paper also earns credit for comparing four retrieval codes, running two independent reductions (exoTEDRF and NAMELESS) that agree, and being upfront that mass-loss and temperature are degenerate in their EvE modeling.\n\nThe soft spot is the pre-transit measurement itself. The helium light curve is built by subtracting the average of two broad continuum bands from the single 1.0835 μm bin. NIRISS-SOSS has known trace position drifts along the dispersion axis. A time-varying drift would affect that narrow bin far more than the wide bands, producing a smooth pre-transit slope that looks like the reported signal. The two pipelines share the same raw frames, so their agreement does not rule this out. The paper reports no check of the trace centroid or wavelength solution stability, and the data are a single 6.2-hour visit with no second epoch. That is a load-bearing gap for the 'first pre-transit detection' claim.\n\nThe rest of the paper is in better shape. The transmission spectrum and retrievals are broadly consistent with prior HST/JWST results; the spot-contamination interpretation of the short-wavelength slope is statistically favored and plausible, though not independently verified. The EvE model is admittedly simplified (ellipsoidal geometry, Parker wind density, reduced XUV flux by a factor of 50), and the authors do not oversell the derived mass-loss rate. The lack of a data/code deposit is a minor but real annoyance for a benchmark claim.\n\nBottom line: this deserves peer review. The observational claim is important enough for a referee to engage with it, but the requested revisions should be concrete: a trace-stability analysis, a search for time-correlated systematics at the pixel level, and ideally a second epoch or simultaneous high-resolution confirmation. Without that, the pre-transit detection is suggestive, not established.\n\nWho this is for: atmospheric escape and JWST time-series people will want it on their radar. I would bring it to a reading group, but I would not cite the pre-transit claim as established fact until the systematic is addressed.","headline":"Plausible space-based pre-transit helium detection, but the single-visit differential light curve needs a trace-stability check before the headline claim is solid.","tokens_in":49541,"tokens_out":3618,"would_cite":false,"duration_ms":34174,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":["97.82.-k"],"model":"deepseek-v4-flash","headline":"This paper reports the first space-based detection of continuous helium absorption beginning 1.5 hours before the transit of WASP-107b, with an ellipsoidal thermosphere model implying an outflow reaching tens of planetary radii.","keywords":["WASP-107b","helium escape","exoplanet atmospheres","transit spectroscopy","JWST NIRISS-SOSS","metastable helium","atmospheric escape","stellar contamination"],"falsifier":"Observe a second transit of WASP-107b with the same NIRISS-SOSS setup; if the helium bin does not show the same roughly 1.5-hour pre-transit rise at the same orbital phase, or shows a signal that tracks the telescope roll angle or detector position rather than the planet, the planetary-outflow interpretation fails.","tokens_in":48349,"feed_emoji":"🪐","tokens_out":7220,"duration_ms":66096,"temperature":0.7,"pith_summary":"This paper reports a JWST/NIRISS-SOSS time series of the warm super-Neptune WASP-107b that shows excess absorption in the metastable helium line at 1.083 microns not only during transit but also for about 1.5 hours before ingress and continuously after egress. The authors argue this is the first space-based detection of pre-transit helium absorption for any exoplanet, at 17 sigma significance in the pre-transit phase and 36 sigma at maximum transit depth. If the interpretation holds, the escaping atmosphere is not a small comet-like tail but a large ellipsoidal thermosphere confined near the planet and extending roughly 10-18 planetary radii ahead of it. The paper further claims a water detection, attributes the blueward spectral slope to unocculted stellar spots rather than haze, and places a 2 sigma upper limit on potassium abundance consistent with a super-solar metallicity atmosphere.","feed_headline":"JWST detects helium 1.5 hours before WASP-107b's transit begins","feed_subtitle":"The same observation maps an ellipsoidal thermosphere tens of planetary radii wide around the ultra-light planet.","key_machinery":"The load-bearing tool is the metastable helium triplet at 1.083 microns observed at R~700, with the helium light curve isolated by subtracting neighboring continuum bins; this isolates the outflow from the solid-body transit. The paper's model of the thermosphere is an ellipsoidal Parker-wind outflow generated with an evaporating-exoplanet code, in which escaping atoms are launched over a 3D surface confined near the planet; varying the ellipse elongation reproduces the pre-transit slope for leading sizes of 10-18 planetary radii. A second key element is the transit light source effect: accounting for unocculted stellar spots in atmospheric retrievals changes the inferred water abundance and explains the blueward slope.","core_discovery":"The central discovery claim is that WASP-107b's metastable helium absorption is continuous across the full observed orbit segment: significant absorption begins approximately 1.5 hours before ingress, reaches a maximum transit depth of 2.395% +/- 0.01% near the helium triplet, and persists through 1 hour after egress. Previous ground-based observations had only seen the post-transit tail; the pre-transit rise was missed because their out-of-transit reference spectra were taken during the very phase where the absorption is already present. The paper models the thermosphere as an ellipsoidal outflow that remains confined close to the planet, with a leading elongation of 10-18 planetary radii and similar trailing extent, and shows that this geometry reproduces the pre-transit slope and the continued post-transit absorption. The authors also retrieve a water abundance of log10 H2O = -2.5 +/- 0.6, show that the short-wavelength slope is best explained by unocculted stellar spots (5.2 sigma) rather than haze, and derive a potassium upper limit below 75 times stellar abundance at 2 sigma.","pith_inferences":["A single-epoch detection cannot distinguish a permanent ellipsoidal thermosphere from a time-variable outflow shaped by the current stellar wind; a second NIRISS-SOSS transit would test whether the pre-transit rise repeats at the same orbital phase.","If pre-transit helium absorption is common among evaporating planets, published mass-loss rates that assumed symmetric transit and post-transit baselines could be systematically biased, and reanalysis with full phase coverage may change population-level escape statistics.","The model's need to reduce stellar XUV flux by a factor of 50 to form a smooth tail points to a sensitive dependence on high-energy stellar input; simultaneous X-ray/EUV monitoring of WASP-107 could turn this from a tuning knob into a testable prediction.","The stellar-spot interpretation predicts that the strength of the short-wavelength slope should vary with the stellar rotation phase; photometric monitoring across the 17-day rotation period could independently confirm the spot explanation."],"forward_implications":["Short-baseline helium observations of other planets likely underestimate the spatial extent of thermospheres; longer JWST baselines will be needed to measure full outflow geometry.","Ground-based out-of-transit reference spectra built during the pre-transit slope are biased, so existing high-resolution absorption amplitudes and line shapes for WASP-107b may need revision.","Ignoring stellar contamination in retrievals overestimates the water abundance by about a factor of 40, so future combined JWST spectra should include spotted-star models.","The inferred mass-loss rate (~1-10 Earth masses per Gyr) and super-solar metallicity make in-situ formation improbable and support a migration origin, possibly with ongoing tidal heating.","Continuous phase coverage provides a template for coordinated space and ground campaigns to break the mass-loss/temperature degeneracy."],"supporting_citations":[{"why":"First detection of helium in WASP-107b's extended atmosphere with HST; supplies the baseline detection that this work extends to continuous phase coverage.","marker":"[2]"},{"why":"High-resolution ground confirmation of the extended helium atmosphere; provides the comparison for absorption amplitude and the stellar helium spectrum used in modeling.","marker":"[12]"},{"why":"Post-transit tail detection at 10830 Angstroms; the paper's post-transit absorption extends this tail continuously into the observed window.","marker":"[13]"},{"why":"Evaporating exoplanet code used to simulate synthetic transits of the ellipsoidal thermosphere; carries the geometric and atmospheric forward modeling.","marker":"[16]"},{"why":"Simulations of stellar wind confinement of evaporating atmospheres; motivates the pre-transit outflow shape and the model's leading extension.","marker":"[19]"},{"why":"Defines the transit light source effect; the retrieval framework's stellar-spot treatment used to explain the blueward slope is built on this.","marker":"[31]"},{"why":"Introduces the metastable helium 10830 Angstrom triplet as a probe of escaping exoplanet atmospheres; justifies the line choice.","marker":"[55]"},{"why":"Hydrodynamic prediction of a hybrid bubble-stream structure for WASP-107b's outflow; matches the ellipsoidal dense stream used in the simplified model.","marker":"[97]"}],"fun_headline_variants":["JWST sees helium around WASP-107b from pre- to post-transit","WASP-107b helium envelope detected 1.5h before transit starts","JWST maps WASP-107b's extended thermosphere in helium","Continuous helium absorption spans WASP-107b transit","Pre-transit helium reveals WASP-107b's far-reaching outflow"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The pre-transit helium signal is interpreted as planetary gas rather than a time-varying instrumental or stellar effect, but it comes from a single night of data with no independent second epoch to check.","fun_headline_variants_meta":{"raw":{"variants":["JWST sees helium around WASP-107b from pre- to post-transit","WASP-107b helium envelope detected 1.5h before transit starts","JWST maps WASP-107b's extended thermosphere in helium","Continuous helium absorption spans WASP-107b transit","Pre-transit helium reveals WASP-107b's far-reaching outflow"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1670,"prompt_tokens":1086,"completion_tokens":584,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":702,"completion_tokens_details":{"reasoning_tokens":487}},"tokens_in":702,"tokens_out":584,"duration_ms":6322,"temperature":1.0,"reasoning_tokens":487,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T13:50:20.075706+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe a second transit of WASP-107b with the same NIRISS-SOSS setup; if the helium bin does not show the same roughly 1.5-hour pre-transit rise at the same orbital phase, or shows a signal that tracks the telescope roll angle or detector position rather than the planet, the planetary-outflow interpretation fails.","supporting_citations":[],"review_version":1}