{"id":"0b06fa66-888d-4409-884f-247fe9c6f4ba","arxiv_id":"2504.18717","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Channel maps of 15 exoALMA disks reveal kinematic deviations in 13, planet-wake-like kinks in six, and partial CO freeze-out in seven midplanes.","lead":"An ALMA study of 15 planet-forming disks finds wiggling gas patterns in 13 of them, and in six disks the patterns look like the wakes of giant planets 1 to 5 times Jupiter's mass. The same data show that carbon monoxide only partially freezes in disk midplanes and re-evaporates in outer regions.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Planet-wake interpretation is not uniquely supported because the models assume vertical isothermality, ruling out VSI by construction; a no-planet radiation-hydro comparison is needed.","rationale":"The reader's weakest assumption identifies the translation of velocity kinks into single-planet wakes as the key unproven step, which matches my read. I sharpen this into a specific model-construction flaw: the phantom simulations are vertically isothermal, which by design suppresses the vertical shear instability, one of the most plausible non-planetary generators of both localized velocity perturbations and filamentary structures in T Tauri disks. Because the null hypothesis (no planet, realistic thermal structure) is never simulated, the observed kinks cannot be uniquely attributed to planets; the paper's own text acknowledges this. A second, related gap is that the quantitative planet masses and depletion factors are derived from a small, by-eye model grid, not a formal fit. These issues do not undermine the robust observational results (deviations in 13/15 disks, midplane CO emission in the favorable 7 disks), but they do mean the planet claim is a plausible interpretation, not a demonstrated detection. The proposed test directly targets the central concern by comparing VSI-active no-planet models to the same channel maps using a statistical metric. This preserves the reader's CONDITIONAL verdict: accept the paper's observational census, but keep the embedded-planet claim contingent on a direct test against the leading alternative physics.","tokens_in":16476,"tokens_out":6896,"duration_ms":74325,"concrete_test":"Run global 3D radiation-hydrodynamical simulations (with vertical temperature gradient; e.g., PLUTO with fargOCA or the code used by Barraza-Alfaro 2025) for SY Cha and LkCa 15 with the exoALMA disk parameters and no embedded planet, post-process with mcfost to produce 12CO J=3-2 channel maps at matching resolution and sensitivity, and compare the kink channels to both the no-planet maps and the 5 MJup planet map of Fig. 7 using a quantitative metric such as residual chi-squared or normalized cross-correlation. If the no-planet, VSI-active model matches the observed kinks as well as or better than the planet model, the evidence for embedded planets in these disks is not established. The same setup can also test whether varying disk aspect ratio changes the inferred planet mass.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that six disks harbor embedded planets rests on interpreting localized velocity kinks as planet wakes, yet the only hydrodynamic models tested are single-planet simulations with a vertically isothermal equation of state. Vertical isothermality removes the vertical temperature gradient that drives the vertical shear instability (VSI), a leading non-planetary source of non-Keplerian motions and filamentary emission in irradiated T Tauri disks. The paper itself notes that the observed filaments resemble VSI/MRI predictions and calls the isothermal assumption a major limitation, but no no-planet, VSI-capable simulation is ever compared to the kink channels. Moreover, the planet mass and depletion constraints come from a by-eye comparison of four mass steps (Fig. 7) and from two disks only (Fig. 8), with no exploration of disk scale height, viscosity, or thermal structure, so the 1-5 MJup estimates and the 10^-3-10^-2 depletion factor are indicative, not demonstrated. The qualitative detections of kinematic deviations and midplane CO emission are robust; the planet interpretation is not uniquely supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper analyzes 12CO J=3-2 channel maps of the 15 exoALMA protoplanetary disks. The authors report kinematic deviations from Keplerian rotation in 13 of 15 disks, classifying them as large-scale arcs, localized velocity kinks, and filamentary structures. For six disks (AA Tau, SY Cha, J1842, J1615, LkCa 15, HD 143006) they interpret the kinks as planet wakes, and for five of them they run phantom/mcfost single-planet simulations, concluding that planet masses of 1-5 MJup at orbital radii of 80-310 au best match the observations. In the seven disks with favorable inclination for separating the upper and lower emitting surfaces, they consistently detect midplane CO emission and infer a partial freeze-out depletion factor of roughly 10^-3 to 10^-2 relative to the warm molecular layer, together with CO desorption in the outer regions. The paper explicitly acknowledges that the morphological classification is subjective and that the modeling is a restricted, preliminary exploration.","tokens_in":16653,"tokens_out":7376,"duration_ms":78298,"significance":"The observational catalog of kinematic substructures and the systematic detection of midplane CO emission are valuable contributions, and the paper strengthens them with concrete cross-checks: reimaging without continuum subtraction, regularized maximum likelihood imaging, and consistency checks against dynamite and discminer surface parameters. The public release of synthetic models and data products through the cited Dataverse is a further strength that supports reproducibility. The wake-location overlay in Figure 5 provides a falsifiable prediction that can be tested with deeper or higher-resolution data. If the planetary interpretation is confirmed by a dedicated test against non-planet mechanisms, the inferred population of 1-5 MJup planets at large orbital radii would be an important result for planet formation. As it stands, the robust claims are the qualitative detections of kinematic deviations and midplane CO; the quantitative mass and depletion values are model-dependent and require stronger support.","major_comments":[{"comment":"The planet mass estimates are based on a by-eye comparison to a grid of 1, 2, 5, and 10 MJup at fixed disk parameters, with no quantitative goodness-of-fit metric or uncertainty budget. The abstract and conclusions present 'masses between 1 and 5 MJup' as a result, but this range is not demonstrably constrained: the paper does not explore variations in disk scale height, viscosity, CO abundance, or planet accretion radius. Either a quantitative fitting procedure over a denser grid should be added, or the mass range should be explicitly described as illustrative rather than as an estimate.","section":"4.2, Fig. 7"},{"comment":"The attribution of the velocity kinks to embedded planets is not tested against the main non-planet alternatives. All new hydrodynamic runs adopt a locally isothermal equation of state and contain exactly one planet, while the observed filamentary structures in SY Cha, J1615, LkCa 15, J1852, DM Tau, and HD 34282 are said to resemble VSI/MRI predictions, as the authors themselves note. A no-planet, vertically stratified or MHD simulation should be compared to the same channel maps to show that instabilities such as VSI cannot produce the observed kink morphology; without such a control, the planetary interpretation is one viable hypothesis rather than a demonstrated detection. I would caution against phrasing the issue as 'vertical isothermality removes VSI', since locally isothermal baroclinic disks can in fact host VSI; the real gap is the absence of a no-planet comparison run.","section":"4.1, 5, Fig. 3"},{"comment":"The depletion factor of 10^-3 to 10^-2 is derived from only two disks (LkCa 15 and J1615) and from a single hydrodynamic realization, with no exploration of correlations with disk density, thermal structure, or UV radiation, yet the abstract applies this range to all seven disks where midplane CO is detected. The qualitative conclusion that CO is not fully frozen out in the midplane is well supported, but the quantitative depletion range should either be restricted to the two modeled sources or be supported by similar fits to the other five disks.","section":"3.2, 4.2, Fig. 8"},{"comment":"Several quoted quantities are fitted parameters rather than independent measurements: the planet orbital radii are set by the visually identified kink locations, the disk masses are tuned to reproduce the separation between emitting layers, and the CO depletion factor is adjusted to match the observed brightness. This fitting procedure is appropriate for testing a hypothesis, but the paper should consistently label the resulting masses, radii, and abundances as model-inferred values, not as free derivations from the data, to avoid giving the impression of independent predictions.","section":"Table 2, 4.1"}],"minor_comments":[{"comment":"The caption lists the synthetic model grid as '1, 2, 3 and 5 M Jup', while the text in §4.1 states masses of 1, 2, 5, or 10 MJup; please reconcile this inconsistency.","section":"Fig. 7 caption"},{"comment":"The word 'corrresponds' should be corrected to 'corresponds'.","section":"3.1"},{"comment":"The abbreviation 'SPHERE DPI' is not expanded at first use; please define 'DPI' (dual-polarization imaging) in the caption or in the text.","section":"Fig. 5 caption"},{"comment":"Velocity units appear inconsistently as 'km/s', 'm/s', and 'm s^-1' across Table 1 and the text; please adopt one consistent notation, preferably 'm s^-1' for channel offsets and 'km s^-1' for systemic velocities.","section":"Table 1 and text"},{"comment":"The sentence noting that 'the majority of disks exhibiting evidence of desorption also show potential signs of planets' and calling this 'likely a coincidence' would benefit from either a brief statistical argument or removal, as presented it is an unsupported aside.","section":"3.2"}],"recommendation":"major_revision","confidential_remarks":"This is a useful and honest survey letter, and the public release of data products is commendable. My main concern is that the abstract and conclusions state quantitative planet masses and depletion factors that the modeling section itself describes as indicative. I would like the revision to either strengthen the modeling support (at minimum a no-planet control simulation for one or two disks, and a more quantitative mass comparison) or soften the headline claims accordingly. The paper's own limitation statements are candid, but they need to be reflected in the abstract's wording."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague—\n\nHere's the short version: this is a genuinely useful survey paper. It applies the channel-map kink technique to a uniform 15-disk sample, finds kinematic deviations in 13/15, and identifies six disks with kinks that look like planet wakes. The observational census is the real contribution; the planet and abundance numbers are plausible but soft.\n\nWhat's new: the uniform application to a large sample, the systematic detection of midplane CO emission in seven disks, and the CO desorption at large radii. The kink method itself goes back to Pinte et al. 2018b/2020 and Calcino et al. 2022, and the HD 143006 kink was already known. But the survey-level statistics and the CO abundance findings are new. The paper is careful: they reimaged without continuum subtraction, checked with RML, and the data and models are public. That's reproducible work and it shows.\n\nWhere it's soft: the planet masses come from visually comparing one channel to a grid of 1, 2, 5, 10 MJup models, no uncertainties, and the models are gas-only, vertically isothermal phantom runs. Vertical isothermality kills the vertical shear instability by construction, and the paper itself notes that the filamentary structures resemble VSI/MRI predictions. No no-planet, VSI-capable run is shown. So the six planet candidates are best read as \"kinematic signatures consistent with wakes,\" not as detected planets. The authors actually say this—they use \"consistent with,\" \"potential,\" \"preliminary\"—so the headline isn't overclaimed. The depletion factor of 10^-3 to 10^-2 comes from a single model realization and two disks; again, the qualitative detection of partial freeze-out is solid, the number is indicative.\n\nThe circularity concern: yes, masses and depletion are fitted to match the data, not predictions. That's fine for a survey letter as long as it's labeled as such, and it is. The kinks themselves are independent observables.\n\nBottom line: this deserves a serious referee. The observational results will be heavily cited, and the interpretations are honestly bounded. I'd send it to review and expect it to come back with the same caveats but sharper language. The reader's conditional verdict is about right; I'd put the emphasis on the CO result being the sturdier finding.\n\nRecommendation: accept after minor revision, or at minimum send to a competent referee.","headline":"A solid survey-level census of kinematic deviations in 15 disks with carefully hedged planet candidates; the CO midplane detection is the sturdiest new result.","tokens_in":17433,"tokens_out":1947,"would_cite":true,"duration_ms":19134,"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":"Six of 15 planet-forming disks show gas kinks consistent with embedded giant planets, and seven show only partial CO freeze-out.","keywords":["protoplanetary disks","planet-disk interaction","CO freeze-out","velocity kinks","channel maps","disk kinematics","wide-orbit planets","submillimeter astronomy"],"falsifier":"A high-contrast image at the deprojected kink locations in AA Tau, SY Cha, J1842, J1615, LkCa 15, or HD 143006 that shows no companion, combined with a non-planet hydrodynamic simulation that reproduces the same kinks, would falsify the planet interpretation; a resolved line-ratio measurement of midplane CO in one of the seven disks implying depletion below $10^{-3}$ of the warm-layer value would falsify the partial freeze-out claim.","tokens_in":16251,"feed_emoji":"🪐","tokens_out":10664,"duration_ms":93551,"temperature":0.7,"pith_summary":"The paper analyzes velocity-channel maps of $^{12}$CO $J=3-2$ emission from 15 planet-forming disks and finds that 13 depart from pure Keplerian rotation, with kinks, arcs, and filaments. In six disks — AA Tau, SY Cha, J1842, J1615, LkCa 15, and HD 143006 — the localized velocity kinks match the predicted wakes of single planets of 1 to 5 Jupiter masses on orbits of 80 to 310 au. In seven disks viewed near 45 degrees, the vertical CO snowline is always detected and the midplane CO is only partially frozen out, depleted by about $10^{-3}$ to $10^{-2}$ relative to the warm molecular layer, with desorption in the outer regions. The paper argues that wide-orbit giant planets may be common in large disks and that CO chemistry in disk midplanes is more complex than full freeze-out.","feed_headline":"Possible planets found in 6 of 15 planet-forming disks","feed_subtitle":"Velocity kinks in CO channel maps also reveal midplane gas only partly frozen out.","key_machinery":"The workhorse is the velocity channel map: a single-velocity image of the line emission, with isovelocity curves showing where the projected Keplerian velocity equals that channel. A 'velocity kink' is a localized distortion of an isovelocity curve, and the paper uses the result that each time a planet's wake crosses an isovelocity curve it creates a kink, so multiple kinks across channels do not rule out a single planet. To turn kinks into planet properties, the paper runs gas-only smoothed-particle hydrodynamics simulations post-processed with 3D radiative transfer to produce synthetic channel maps for planets of 1, 2, 5, and 10 $M_\\mathrm{Jup}$, comparing shapes and amplitudes by eye. For the chemistry, a CO abundance prescription with freeze-out below 20 K, photodissociation, and photodesorption is used to model the midplane depletion and outer-disk desorption.","core_discovery":"The central claim is that individual channel maps of $^{12}$CO $J=3-2$ in the 15-disk sample reveal a rich zoo of kinematic deviations from Keplerian rotation, and that in six of those disks the deviations take the form of velocity kinks consistent with the wakes of embedded planets. Comparing the data to gas-only hydrodynamical plus radiative-transfer models with a single planet, the paper infers planet masses of about 1 to 5 $M_\\mathrm{Jup}$ at orbital radii of 80 to 310 au. A separate but equally central discovery is that in every disk where the upper and lower emission surfaces can be separated (seven disks), there is non-zero emission between the surfaces, implying the vertical CO snowline is present but freeze-out is partial: the midplane $^{12}$CO abundance is only $10^{-3}$ to $10^{-2}$ of the warm-layer value, and CO desorbs in the outer disk. Both results are presented as preliminary, with the planet masses limited by the simplicity of the model grid.","pith_inferences":["If midplane CO is only partially depleted in a wide range of disks, then CO snowline positions derived from emission gaps may overestimate the true freeze-out radius; mapping midplane emission at higher resolution could test this.","The prevalence of wide-orbit planet candidates in this sample could connect to direct-imaging surveys: many of these planets would be directly imageable with next-generation instruments if they are real companions.","Comparing these channel maps with non-planet simulations (e.g., vertical shear instability or buoyancy spirals) at the same resolution would separate planet wakes from other velocity perturbations; the paper's public data products make such a comparison possible."],"forward_implications":["If the six planet candidates are real, wide-orbit giant planets (1–5 $M_\\mathrm{Jup}$ at 80–310 au) are present in 6 of 15 large, bright disks, suggesting such planets are not rare when the disk is big enough.","The candidate planets sit just outside the dust continuum emission, which would mean they are truncating the dusty disk at large radii.","Partial midplane freeze-out ($10^{-3}$–$10^{-2}$ depletion) and outer-disk desorption must be built into disk chemical models, and CO-based gas mass estimates that assume full freeze-out may need revision.","The background of other kinematic deviations sets the current detection floor at about 1 $M_\\mathrm{Jup}$, so lower-mass planets could be hidden in the noise.","Additional physics (vertical temperature gradients, multiple planets, or disk instabilities) is likely needed to explain all observed deviations, especially the filamentary structures."],"supporting_citations":[{"why":"Established that localized velocity kinks in CO emission trace embedded planets.","marker":"Pinte et al. 2018b"},{"why":"Showed that a planet wake creates a kink each time it crosses an isovelocity curve, justifying multi-kink detections.","marker":"Bollati et al. 2021"},{"why":"Supplied the CO freeze-out and photodesorption prescription and the earlier IM Lupi midplane emission detection this paper extends.","marker":"Pinte et al. 2018a"},{"why":"Provided the smoothed-particle hydrodynamics code used to simulate planet wakes.","marker":"Price et al. 2018"},{"why":"Provided the radiative-transfer code used to compute synthetic CO channel maps from the hydrodynamical models.","marker":"Pinte et al. 2006, 2009"},{"why":"Benchmarked the combined hydrodynamics and radiative-transfer modeling used for the planet-mass grid.","marker":"Bae et al. 2025"},{"why":"Previously modeled HD 143006 with a planet, the source whose kink detection is revisited here.","marker":"Ballabio et al. 2021"},{"why":"Independently derived a midplane CO depletion factor of about 1/300 in HD 163296, bracketing the paper's $10^{-3}$–$10^{-2}$ range.","marker":"Qi & Wilner 2024"}],"fun_headline_variants":["Six of 15 disks show possible planet wakes","CO maps: planet wakes in 6 disks, freeze-out in 7","Planet clues from CO channel maps in 6 disks","Velocity kinks hint at planets in 6 of 15 disks","Partial CO freeze-out and planet wakes in ALMA disks"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument rests on assuming each velocity kink is the wake of a single embedded planet rather than another disk motion (instability, thermal wave, or companion), and that the planet masses come out right from a simplified gas-only, vertically isothermal, single-planet model with a fixed viscosity and a chosen CO abundance.","fun_headline_variants_meta":{"raw":{"variants":["Six of 15 disks show possible planet wakes","CO maps: planet wakes in 6 disks, freeze-out in 7","Planet clues from CO channel maps in 6 disks","Velocity kinks hint at planets in 6 of 15 disks","Partial CO freeze-out and planet wakes in ALMA disks"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000276,"raw_usage":{"total_tokens":1680,"prompt_tokens":1011,"completion_tokens":669,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":627,"completion_tokens_details":{"reasoning_tokens":584}},"tokens_in":627,"tokens_out":669,"duration_ms":6693,"temperature":1.0,"reasoning_tokens":584,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:10:52.443758+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A high-contrast image at the deprojected kink locations in AA Tau, SY Cha, J1842, J1615, LkCa 15, or HD 143006 that shows no companion, combined with a non-planet hydrodynamic simulation that reproduces the same kinks, would falsify the planet interpretation; a resolved line-ratio measurement of midplane CO in one of the seven disks implying depletion below $10^{-3}$ of the warm-layer value would falsify the partial freeze-out claim.","supporting_citations":[{"cited_title":"2025, ApJL, exoALMA","cited_arxiv_id":null,"evidence_quote":"Benchmarked the combined hydrodynamics and radiative-transfer modeling used for the planet-mass grid."}],"review_version":1}