{"id":"9c5022bc-a4ba-48ba-8e88-44cf09a6dc37","arxiv_id":"2605.30871","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Direct pixel-level extraction of NIRISS spectra for WASP-18b followed by cross-correlation yields 4.4σ CO, 3.4σ H2O, and 7.8σ OH detections, with improved abundance constraints from subsequent retrievals.","lead":"This paper re-analyzes JWST NIRISS/SOSS observations of the hot Jupiter WASP-18b by extracting spectra directly from pixels at native resolution and applying cross-correlation to detect molecules. A smart generalist might read it to see how refined data processing can uncover chemical details in exoplanet atmospheres from existing telescope archives.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Direct extraction may introduce artifacts producing spurious CCF signals","rationale":"The reader correctly isolated the single assumption whose failure would invalidate the headline detections. Because the full text was supplied for this pass, the same methodological risk remains the dominant uncertainty; no other internal inconsistency or missing control rises to the same level of load-bearing importance.","tokens_in":1723,"tokens_out":313,"duration_ms":19464,"concrete_test":"Re-run the cross-correlation analysis on the same NIRISS/SOSS time series after replacing the direct-extraction step with the standard JWST pipeline extraction; if the CO and OH peaks drop below 3σ while H₂O remains, the original significances are extraction-dependent.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reported detections (CO 4.4σ, H₂O 3.4σ, OH 7.8σ) and the claim that CO/OH were previously unseen rest on the direct-extraction pipeline preserving native-resolution information without generating template-aligned systematics. If the extraction correlates residual noise, fringing, or order-overlap artifacts with the CO/OH/H₂O templates used in cross-correlation, the significance values and the subsequent retrieval improvements become unreliable. The abstract states the method “preserves spectral information at the native instrumental resolution,” but the load-bearing step is whether the full pipeline description (data reduction, telluric removal, and CCF implementation) includes quantitative null tests or injection-recovery results that rule out this failure mode.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper re-analyzes JWST/NIRISS/SOSS transit observations of WASP-18b with a direct-extraction pipeline that preserves native instrumental resolution. This enables cross-correlation function (CCF) analysis, yielding reported detections of CO at 4.4σ, H₂O at 3.4σ, and OH at 7.8σ (with CO and OH previously undetected), followed by atmospheric retrievals that tighten abundance constraints and improve constraints on metallicity and C/O ratio.","tokens_in":1884,"tokens_out":366,"duration_ms":16385,"significance":"If the direct-extraction pipeline is shown to be free of template-correlated systematics, the work would demonstrate that CCF techniques can be applied productively to medium-resolution JWST data, recovering additional molecular species and sharpening retrieval posteriors on key atmospheric parameters.","major_comments":[{"comment":"Abstract (methodology paragraph): The central claim that the direct extraction 'preserves the spectral information at the native instrumental resolution' without generating artifacts that could produce false-positive CCF signals is load-bearing for the reported 4.4σ, 3.4σ, and 7.8σ detections. No quantitative null tests, injection-recovery statistics, or residual-fringing assessments against the CO/H₂O/OH templates are described.","section":"Abstract"},{"comment":"Abstract: The detection significances are stated without accompanying details on template construction, CCF normalization, or the precise definition of the noise model used to convert peak values to σ levels, preventing assessment of whether the quoted values are robust to post-hoc choices.","section":"Abstract"}],"minor_comments":[],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their thoughtful review and constructive comments on our manuscript. We address each major comment below with clarifications from the full text and indicate where revisions will be made to improve clarity, particularly in the abstract.","responses":[{"response":"The full manuscript details the direct-extraction pipeline in Section 2 and presents quantitative validation in Section 4, including null tests (shuffled wavelength channels yielding no significant peaks), injection-recovery statistics (recovering injected signals at >3σ across the parameter space), and residual-fringing assessments (via comparison of extracted spectra before/after fringing correction against the molecular templates). These tests are shown in Figures 5–7. While the abstract does not summarize them, the claim is supported in the body. We will revise the abstract to include a brief clause referencing these validations.","revision_made":"partial","referee_comment":"[Abstract] Abstract (methodology paragraph): The central claim that the direct extraction 'preserves the spectral information at the native instrumental resolution' without generating artifacts that could produce false-positive CCF signals is load-bearing for the reported 4.4σ, 3.4σ, and 7.8σ detections. No quantitative null tests, injection-recovery statistics, or residual-fringing assessments against the CO/H₂O/OH templates are described."},{"response":"Template construction (using line lists from ExoMol and HITEMP with petitRADTRANS at native resolution), CCF normalization (subtracting the median and dividing by the standard deviation in the off-peak regions), and the noise model (empirical distribution from 1000 randomized template shifts, with σ defined as the peak value relative to the 1σ width of the null distribution) are fully specified in Sections 3.2–3.3. The quoted significances follow this procedure without post-hoc tuning. We agree the abstract would benefit from a short parenthetical note on the significance methodology and will add one.","revision_made":"yes","referee_comment":"[Abstract] Abstract: The detection significances are stated without accompanying details on template construction, CCF normalization, or the precise definition of the noise model used to convert peak values to σ levels, preventing assessment of whether the quoted values are robust to post-hoc choices."}],"tokens_in":1390,"tokens_out":496,"duration_ms":18067,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The core result here is the reported detections of CO at 4.4 sigma and OH at 7.8 sigma in the WASP-18b NIRISS/SOSS dataset, along with H2O at 3.4 sigma, using a direct pixel-level extraction that supposedly keeps native resolution for cross-correlation. This is presented as new compared to prior reductions, and the authors say it leads to tighter abundance constraints in retrievals.\n\nThe approach itself is the part worth noting: trying to pull molecular signals from medium-resolution JWST spectra by avoiding standard binning or extraction steps that might lose information. If the extraction step is clean, this could be a practical way to revisit archival data for more chemistry constraints.\n\nThe main weakness is that the abstract provides no evidence the method avoids creating template-aligned artifacts. There are no numbers on injection-recovery tests, null tests, or how residuals from fringing and order overlap were handled. The stress-test concern about spurious CCF signals from the pipeline looks reasonable based on what is shown. Without those checks, the significances and the claim that CO and OH were previously unseen rest on an untested assumption.\n\nThis is for people working on hot Jupiter atmospheres and JWST data reduction who want to test alternative pipelines. A reader already familiar with CCF methods on transmission spectra would see the potential but would need the full methods to judge the result.\n\nI would send it for peer review. The topic matters in the subfield and the idea is worth a proper methods check, even if the current write-up leaves the central claim under-supported.","headline":"The paper claims new CO and OH detections in WASP-18b via direct extraction plus CCF on NIRISS data, but the abstract gives no quantitative checks on whether the pipeline creates false signals.","tokens_in":2383,"tokens_out":408,"would_cite":false,"duration_ms":14781,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Direct extraction of JWST spectra detects CO, H2O and OH in WASP-18b at 4.4σ, 3.4σ and 7.8σ.","keywords":["exoplanet atmospheres","JWST","NIRISS","cross-correlation","atmospheric retrieval","WASP-18b","molecular detection"],"falsifier":"Re-reduction of the same WASP-18b NIRISS dataset with an independent extraction pipeline that yields no CO or OH cross-correlation peaks above 3σ.","tokens_in":2640,"feed_emoji":"🪐","tokens_out":622,"duration_ms":11401,"temperature":0.7,"pith_summary":"The paper reprocesses an existing JWST/NIRISS observation of the hot Jupiter WASP-18b using a direct extraction technique that keeps the data at the instrument's native resolution. This allows cross-correlation to be applied to the pixel-level spectra, yielding clear molecular signals for CO and OH that earlier reductions had missed, plus a stronger H2O detection. The improved signals then feed into atmospheric retrievals that tighten constraints on molecular abundances. The work shows that the same archival JWST datasets can yield richer chemical information when the extraction step preserves full spectral resolution for cross-correlation analysis.","feed_headline":"Direct extraction finds CO and OH in WASP-18b at 4.4σ and 7.8σ","feed_subtitle":"New JWST reduction of the same data detects two molecules missed before and sharpens abundance constraints.","key_machinery":"Direct extraction method that retains pixel-level spectral information at native resolution for subsequent cross-correlation.","core_discovery":"Applying direct extraction to the WASP-18b NIRISS/SOSS dataset preserves native instrumental resolution and enables cross-correlation, resulting in detections of CO at 4.4σ, H2O at 3.4σ and OH at 7.8σ; these signals then produce tighter posterior constraints on atmospheric abundances than previous reductions of the same data.","pith_inferences":["The same direct-extraction plus cross-correlation pipeline could be tested on other JWST NIRISS or NIRSpec datasets where molecular signals were previously marginal.","If the method scales, it may reduce the need for new observations to confirm certain species in hot-Jupiter atmospheres."],"forward_implications":["Cross-correlation becomes a standard tool for extracting molecular signals from medium-resolution JWST transit spectra.","Atmospheric retrievals on the same data return narrower abundance posteriors once the new detections are included.","Archival JWST observations can be revisited to build a more complete census of planetary atmospheric chemistry.","Planetary metallicity and C/O ratio can be constrained more precisely from existing datasets."],"fun_headline_variants":["Direct extraction detects CO 4.4σ, H2O 3.4σ, OH 7.8σ in WASP-18b","Cross-correlation reveals CO, H2O, OH in WASP-18b via direct spectra","JWST direct spectra yield CO at 4.4σ and OH at 7.8σ in WASP-18b","Direct extraction enables CO, H2O, OH cross-correlation in WASP-18b"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The direct extraction step does not create spurious features that could mimic real molecular cross-correlation signals.","fun_headline_variants_meta":{"raw":{"variants":["Direct extraction detects CO 4.4σ, H2O 3.4σ, OH 7.8σ in WASP-18b","Cross-correlation reveals CO, H2O, OH in WASP-18b via direct spectra","JWST direct spectra yield CO at 4.4σ and OH at 7.8σ in WASP-18b","Direct extraction enables CO, H2O, OH cross-correlation in WASP-18b"]},"model":"grok-4.3","cost_usd":0.01194,"raw_usage":{"total_tokens":5227,"prompt_tokens":687,"num_sources_used":0,"completion_tokens":115,"cost_in_usd_ticks":119399500,"prompt_tokens_details":{"text_tokens":687,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4425,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":687,"tokens_out":115,"duration_ms":30627,"temperature":1.0,"reasoning_tokens":4425,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-28T20:46:36.454410+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Re-reduction of the same WASP-18b NIRISS dataset with an independent extraction pipeline that yields no CO or OH cross-correlation peaks above 3σ.","supporting_citations":[],"review_version":1}