{"id":"10be6884-d96c-4f55-b1f0-4469b98a189d","arxiv_id":"2411.19662","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":12,"one_line_summary":"Dayside spectroscopy of WASP-189b with CRIRES+ is best explained by a 4.4 km/s day-to-night wind, with no significant equatorial jet.","lead":"Astronomers used the CRIRES+ spectrograph to observe the ultra-hot Jupiter WASP-189b and detected carbon monoxide and iron lines redshifted by about 6 km/s. By fitting the shapes of those lines with wind models, they conclude the planet is best described by a 4.4 km/s day-to-night wind rather than a fast equatorial jet.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Wind detection is statistically marginal: the retrieved vday-night = 4.4+1.8-2.2 km/s is consistent with zero at ~2σ, and the paper offers no Bayesian model comparison to substantiate the abstract's 'best fitted' claim.","rationale":"The reader's verdict is CONDITIONAL and I agree. The most load-bearing concern is not the specific line-profile assumptions (hot spot, uniform wind) but the statistical strength of the detection itself. The paper is honest about the ~2σ consistency in Sect. 5.2, but the abstract and the strongest claim overstate the result by omitting this caveat and by using 'best fitted' without a formal comparison. A Bayes factor would settle whether the data actually favor the wind model over a simpler alternative. The missing hot-spot simulations are a secondary reproducibility issue, but the statistical concern is more fundamental because it applies even if the model is perfectly correct. Since the paper already discloses the weakness, the verdict remains CONDITIONAL (with requested revisions), not REJECT.","tokens_in":19211,"tokens_out":4345,"duration_ms":33244,"concrete_test":"Compute the Bayesian evidence (or BIC as a proxy) for the wind-pattern retrieval of Sect. 5.1 versus a nested no-wind model with vday-night = 0, vjet = 0, and veq free (as in Sect. 4), using identical data, SYSREM filtering, and the likelihood of Eq. (3). If the log-evidence difference is < 3 (Bayes factor < ~20), the abstract's 'best fitted by a day-to-night wind' should be replaced with 'consistent with both wind and wind-free scenarios at ~2σ.'","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that the line profile favors a day-to-night wind of 4.4 km/s over a wind-free scenario—rests on a marginal detection. From Table C.1, vday-night = 4.4+1.8-2.2 km/s, so the 2σ lower bound is ~0 km/s. The paper itself states (Sect. 5.2): 'the retrieved posteriors of the day-to-night wind speed are consistent with a wind-free scenario to within ~2σ.' Despite this, the abstract presents the wind as the best-fit scenario without the caveat. No formal model comparison (e.g., Bayesian evidence, Bayes factor, or even BIC) is provided between the wind-pattern retrieval and a no-wind model (e.g., the Sect. 4 retrieval with veq free and ϵ=1). The degeneracy between vday-night and voffset, acknowledged in Sect. 5.2, further complicates the separation: a wind-free model with a larger voffset may fit similarly. Since the wind detection is the paper's primary new result, the absence of a quantitative model comparison is the most load-bearing weakness.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents CRIRES+ K-band dayside emission spectroscopy of the ultra-hot Jupiter WASP-189b. After removing stellar and telluric lines with SYSREM, the authors detect CO and Fe with a velocity offset of about 6 km/s relative to the expected planetary rest frame. They then perform a Bayesian atmospheric retrieval with petitRADTRANS, and extend it with a disk-integrated line-profile model that includes tidally locked rotation, a uniform day-to-night wind, and a Gaussian-profile equatorial jet. They report a day-to-night wind speed of 4.4+1.8-2.2 km/s and an equatorial jet velocity consistent with zero, together with a reduced residual velocity offset, an inverted temperature-pressure profile, and equilibrium-chemistry C/O and metallicity constraints. The central interpretive claim is that the dayside circulation of WASP-189b is dominated by day-to-night flow rather than by a fast zonal jet as predicted by Lee et al. (2022).","tokens_in":19479,"tokens_out":6358,"duration_ms":56867,"significance":"If the wind detection is statistically robust, the result is an important empirical constraint on circulation in ultra-hot Jupiters, favoring day-to-night flows over zonal jets in WASP-189b and providing a direct test of general circulation models. The paper has several genuine strengths: the SYSREM filtering effect on line shapes is validated with injection tests in Appendix B, the number of SYSREM iterations is selected objectively following Cheverall et al. (2023), the reduced data are publicly available, and the authors openly discuss degeneracies and modelling limitations. The main significance is currently limited by the marginal statistical significance of the wind signal and by the simplified kinematic model used to interpret the line profile.","major_comments":[{"comment":"The headline result that the line profile is 'best fitted' by a day-to-night wind is not supported by any quantitative model comparison. The retrieved vday-night = 4.4+1.8-2.2 km/s is consistent with zero at about the 2-sigma level, as the paper itself states, and voffset is a free parameter that can absorb part of the velocity shift. The authors should compare the wind retrieval against a no-wind model (vday-night fixed to zero, with voffset left free) using Bayesian evidence, BIC, or a likelihood-ratio test, and report the resulting significance. Until this is done, the abstract and Section 6 overstate the result.","section":"Section 5.2 and Table C.1"},{"comment":"The line-profile model assumes a uniform day-to-night wind over the entire visible hemisphere, no sub-stellar hot spot, and a linear limb-darkening law with a free coefficient epsilon. The retrieved epsilon is unconstrained, so the brightness weighting of the disk is largely free. The statement that 'a hot spot alone can not result in the retrieved line profile' is load-bearing for the wind interpretation, but the supporting simulations are not shown or referenced. Please include these simulations or an equivalent injection test with a hot-spot model, and quantify how much the retrieved vday-night and voffset shift when a hot spot or a spatially varying wind field is included.","section":"Sections 5.1-5.2"},{"comment":"The posteriors of sigma_jet and epsilon fill the entire prior range, and vjet is consistent with zero within uncertainties. This means the only constrained wind parameter is vday-night, while the two unconstrained parameters add flexibility to the line-profile model. The paper should state this limitation more prominently and, ideally, repeat the wind retrieval with sigma_jet and epsilon fixed to fiducial values to demonstrate that the vday-night constraint is stable under plausible choices of these nuisance parameters.","section":"Appendix C and Fig. C.1"}],"minor_comments":[{"comment":"The abstract should quote the 1-sigma uncertainties on the retrieved wind speeds, since their precision is central to the claim.","section":"Abstract"},{"comment":"The prior range for log10 kappa_IR differs between the initial retrieval ([-15, 4]) and the wind-pattern retrieval ([-5, 5]); please clarify whether this difference is intentional and whether it affects the comparability of the retrieved T-p parameters.","section":"Tables 3 and C.1"},{"comment":"The appendix title refers to 'Kp-vsys maps', but the plotted quantity is voffset as defined in Eq. (2); retitle the appendix to 'Kp-voffset maps' for consistency.","section":"Appendix A"},{"comment":"The sentence 'We were not able to place constrains on the limb darkening coefficient' contains a typo: 'constrains' should be 'constraints'.","section":"Section 5.2"},{"comment":"The claim that the CO and Fe signals are 'stable across a wider range of iterations' is not documented; a supplementary figure or a brief quantitative statement of the iteration range would make this assertion checkable.","section":"Section 3.4"}],"recommendation":"major_revision","confidential_remarks":"The paper is likely publishable after the requested model comparison is added. If the authors prefer not to add a formal Bayesian model comparison, the abstract and conclusions must be weakened to say that the data are marginally consistent with a day-to-night wind rather than that the line profile is 'best fitted' by one. The unsupported hot-spot simulation claim should also be either shown or removed."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Genuinely new here is the full line-profile retrieval that tries to separate a day-to-night wind from an equatorial jet on WASP-189b's dayside. Yan et al. (2022) already measured the ~6 km/s redshift; this paper goes further and asks what velocity pattern produces the line shape. The method is thoughtful: a 2D disk model with tidal rotation plus a uniform day-to-night wind and a Gaussian zonal jet, with SYSREM filtering validated in Appendix B, which matters for any line-shape claim. The retrieval gives v_day-night = 4.4+1.8-2.2 km/s and v_jet ~1 km/s, consistent with no jet. That is a plausible observational counterpoint to the Lee et al. (2022) GCM prediction of a fast equatorial jet.\n\nThe paper is also honest about its limits. It states plainly that the day-to-night wind is consistent with a wind-free scenario at ~2σ, and it discusses the degeneracy with voffset, the possible contributions from ephemerides, and a hot spot.\n\nThe soft spots are real but not fatal. The abstract's 'best fitted' overstates a posterior that includes zero within 2σ, and there is no Bayesian model comparison—no evidence or BIC—between the wind model and a wind-free model with a larger voffset. That is the load-bearing weakness. If a wind-free model fits nearly as well, the headline result becomes an upper limit rather than a detection. The missing hot-spot simulations, mentioned in Sect. 5.2 but not shown, are a minor issue by comparison. The unconstrained jet width and limb-darkening parameter are consistent with a weak jet, but they also mean the model may be insensitive to jet structure, so 'no jet' should be read with caution.\n\nI would send this to a serious referee. The data are public, the methodology is careful, and the question—whether the dayside circulation of an ultra-hot Jupiter is dominated by day-to-night flow or a jet—is exactly what a single well-observed object can illuminate, even if the answer here is provisional. The referee should ask for a formal model comparison and an abstract that states the 2σ caveat. With those changes the paper is a solid contribution regardless of whether the wind detection survives.\n\nWho gets value: anyone working on high-resolution emission retrievals of hot Jupiters, and theorists comparing GCMs to line-shape observations.","headline":"A careful, honest retrieval paper whose headline wind detection is statistically marginal; worth a serious referee, but the abstract oversells and a model comparison is missing.","tokens_in":20100,"tokens_out":2904,"would_cite":true,"duration_ms":22975,"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 dayside wind of ultra-hot Jupiter WASP-189b is best explained by a day-to-night flow of about 4.4 km/s, with no detectable equatorial jet.","keywords":["exoplanet atmospheres","ultra-hot Jupiter","WASP-189b","high-resolution spectroscopy","atmospheric circulation","day-to-night wind","equatorial jet","emission spectroscopy"],"falsifier":"Take high-resolution K-band dayside spectra over a wider orbital-phase range around secondary eclipse and measure the CO and Fe line-centre shift as a function of sub-stellar longitude. If the shift tracks the rotating hot region, the day-to-night wind interpretation is supported; if it stays constant near 2 km/s, the remaining offset is dominated by orbital-clock or systematic effects and the wind speed would be overestimated. A hot-spot-only retrieval, with the same grid but no wind, would settle whether the extended red wing truly requires a wind; the paper states such a comparison but does not show it.","tokens_in":19003,"feed_emoji":"🌬️","tokens_out":13071,"duration_ms":100899,"temperature":0.7,"pith_summary":"This paper tries to establish how winds circulate in the upper atmosphere of WASP-189b, an ultra-hot gas giant with a dayside hot enough to melt iron. Using high-resolution K-band emission spectroscopy, the authors detect carbon monoxide and iron lines from the planet's dayside and fit their shapes with a model that separates three motions: the planet's tidally locked rotation, a uniform day-to-night flow, and a zonal equatorial jet. The best-fitting day-to-night wind is $4.4^{+1.8}_{-2.2}$ km/s, while the equatorial jet is only $1.0^{+0.9}_{-1.8}$ km/s and is consistent with zero. If correct, the roughly 6 km/s red-shift measured in the planet's lines is mostly gas streaming from the heated dayside toward the nightside, and the circulation is dominated by day-to-night flow rather than the fast equatorial jet predicted by some global circulation models.","feed_headline":"WASP-189b's dayside wind blows day-to-night at 4.4 km/s","feed_subtitle":"A retrieval of CO and Fe emission lines finds the equatorial jet near zero on this ultra-hot Jupiter.","key_machinery":"The argument is carried by a disk-integrated velocity-profile model. The projected dayside is divided into a fine grid of cells, each assigned a radial velocity from the tidally locked rotation, a constant day-to-night wind, and a Gaussian-shaped equatorial jet whose width is a free parameter; each cell is also brightness-weighted by a linear limb-darkening law. Histogramming the cell velocities weighted by brightness produces a line profile, which is convolved onto synthetic emission spectra in logarithmic wavelength space and fitted to the data after stellar and telluric lines are removed with a principal-component filter. A model-filter step reproduces the distortions that line removal introduces, and an injection test shows that the wind-induced broadening and the red-shifted wing survive that filtering. This replaces the simpler rigid-rotation profile with a wind screen, turning a velocity-offset measurement into a wind measurement.","core_discovery":"The central discovery is that the line shape of WASP-189b's dayside emission favours a wind pattern. The paper reports that the observed CO and Fe lines are best fitted when gas moves from the sub-stellar point toward the nightside at $4.4^{+1.8}_{-2.2}$ km/s across the visible hemisphere, while the retrieved equatorial jet velocity of $1.0^{+0.9}_{-1.8}$ km/s is consistent with the absence of a jet. Adding this wind reduces the unexplained velocity offset from about 4.7 km/s to $1.9^{+1.6}_{-1.5}$ km/s, so the wind explains most but not all of the red-shift; a stronger wind would produce more line broadening than observed. The same retrieval finds an inverted temperature-pressure profile and, under equilibrium chemistry, a C/O ratio of $0.32^{+0.41}_{-0.14}$ and a metallicity of $[M/H] = 1.40^{+1.39}_{-0.60}$, and the paper argues that a bright sub-stellar hot spot alone cannot reproduce the measured red-shifted wing.","pith_inferences":["A testable extension would be phase-resolved spectroscopy: if day-to-night flow dominates, the line-centre shift should track the sub-stellar longitude as the planet rotates, whereas a fixed ephemeris offset would not vary with phase.","The wind speed is mainly carried by the CO line shape because CO has the strongest signal; a retrieval on a wavelength setting with dominant Fe lines would test whether the wind geometry is uniform across species.","Because the model omits a bright sub-stellar spot and the paper does not show the hot-spot-only simulation, a natural next test is to add a spot brightness map; such a spot would also produce a red-shift, so part of the retrieved 4.4 km/s wind could trade against a spot-induced shift.","Since the posterior is consistent with no wind at about the 2-sigma level, the wind claim leans on the line-broadening side of the fit; a higher-signal-to-noise spectrum would decide between a 4 km/s wind plus small offset and a no-wind atmosphere with a large offset."],"forward_implications":["A red-shift of a few km/s in a hot Jupiter's dayside spectrum can be produced by a day-to-night wind, so future detections of such offsets do not by themselves require unseen orbital or systematic errors.","Most of the roughly 6 km/s offset in WASP-189b's CO and Fe lines is wind, with a remaining offset of about 2 km/s that could come from orbital-clock uncertainties or a sub-stellar hot spot.","A near-zero equatorial jet supports the idea that extreme irradiation suppresses zonal jets in ultra-hot Jupiters, leaving day-to-night flow as the dominant circulation.","An inverted temperature-pressure profile and a stellar-to-super-solar metallicity, if confirmed, indicate a strongly irradiated, heavy-element-enriched atmosphere with thermal dissociation shaping the upper layers.","Combining transmission and dayside observations in one retrieval would use opposite wind shifts on the two hemispheres to constrain the velocity offset and wind speed more tightly than either data set alone."],"supporting_citations":[{"why":"Prior dayside emission detection of Fe and the temperature-pressure profile that the adopted model is tuned to resemble.","marker":"Yan et al. (2020)"},{"why":"Reported the earlier ~4.5 km/s CO red-shift that this work's ~6 km/s offset confirms and attributes to day-to-night wind.","marker":"Yan et al. (2022)"},{"why":"Transmission spectroscopy showing species-dependent blue-shifts on the limb that motivate day-to-night winds and provide the wind speed comparison.","marker":"Prinoth et al. (2022)"},{"why":"The global circulation model predicting a fast 5-6 km/s equatorial jet that the retrieved near-zero jet is explicitly set against.","marker":"Lee et al. (2022)"},{"why":"Supplies the planetary parameters, radius, equilibrium temperature, and ephemeris used to compute the expected orbital velocity and tidal rotation.","marker":"Lendl et al. (2020)"},{"why":"Supplies the normalization and filtering approach used to remove stellar and telluric lines while preserving the wind-sensitive line shapes.","marker":"Gibson et al. (2022)"},{"why":"Defines the weighted cross-correlation computation used to detect CO and Fe and to build the Kp-voffset maps.","marker":"Cont et al. (2022)"},{"why":"Provides the rotational-broadening formula that the velocity-profile model generalises when fitting wind structure.","marker":"Díaz et al. (2011)"},{"why":"Radiative transfer code generating the synthetic emission spectra used in both the cross-correlation and the retrieval.","marker":"Mollière et al. (2019)"},{"why":"Equilibrium-chemistry solver that converts the retrieved C/O ratio and metallicity into pressure-dependent molecular abundances.","marker":"Stock et al. (2018)"}],"fun_headline_variants":["WASP-189b's dayside wind blows day-to-night at 4.4 km/s","Line shapes reveal WASP-189b's day-to-night wind at 4.4 km/s","Equatorial jet near zero on ultra-hot Jupiter WASP-189b","WASP-189b's CO and Fe lines point to day-to-night flow","WASP-189b: fast day-to-night wind, no equatorial jet"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the visible dayside can be described by a single uniform day-to-night wind plus a Gaussian jet superimposed on a tidally locked, linearly limb-darkened disk with no sub-stellar hot spot; if the real velocity field contains a hot spot or spatially varying winds, the retrieved speeds and the residual velocity offset will be biased.","fun_headline_variants_meta":{"raw":{"variants":["WASP-189b's dayside wind blows day-to-night at 4.4 km/s","Line shapes reveal WASP-189b's day-to-night wind at 4.4 km/s","Equatorial jet near zero on ultra-hot Jupiter WASP-189b","WASP-189b's CO and Fe lines point to day-to-night flow","WASP-189b: fast day-to-night wind, no equatorial jet"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000716,"raw_usage":{"total_tokens":3321,"prompt_tokens":1150,"completion_tokens":2171,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":766,"completion_tokens_details":{"reasoning_tokens":2060}},"tokens_in":766,"tokens_out":2171,"duration_ms":13189,"temperature":1.0,"reasoning_tokens":2060,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T05:59:50.392357+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take high-resolution K-band dayside spectra over a wider orbital-phase range around secondary eclipse and measure the CO and Fe line-centre shift as a function of sub-stellar longitude. If the shift tracks the rotating hot region, the day-to-night wind interpretation is supported; if it stays constant near 2 km/s, the remaining offset is dominated by orbital-clock or systematic effects and the wind speed would be overestimated. A hot-spot-only retrieval, with the same grid but no wind, would settle whether the extended red wing truly requires a wind; the paper states such a comparison but does not show it.","supporting_citations":[{"cited_title":"2020, A&A, 640, L5","cited_arxiv_id":null,"evidence_quote":"Prior dayside emission detection of Fe and the temperature-pressure profile that the adopted model is tuned to resemble."},{"cited_title":"2022, A&A, 661, L6","cited_arxiv_id":null,"evidence_quote":"Reported the earlier ~4.5 km/s CO red-shift that this work's ~6 km/s offset confirms and attributes to day-to-night wind."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The global circulation model predicting a fast 5-6 km/s equatorial jet that the retrieved near-zero jet is explicitly set against."},{"cited_title":"2020, A&A, 643, A94","cited_arxiv_id":null,"evidence_quote":"Supplies the planetary parameters, radius, equilibrium temperature, and ephemeris used to compute the expected orbital velocity and tidal rotation."},{"cited_title":"P., Nugroho , S","cited_arxiv_id":null,"evidence_quote":"Supplies the normalization and filtering approach used to remove stellar and telluric lines while preserving the wind-sensitive line shapes."}],"review_version":1}