{"id":"ead08872-60ab-4295-843e-222134f3a0d0","arxiv_id":"2504.18688","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The exoALMA survey obtained deep, high-resolution ALMA images of 15 protoplanetary disks and found gas and dust substructure in 14 of them, along with small velocity disturbances in all sources.","lead":"This paper presents the science goals, observing design, and first data release of the exoALMA Large Program, a survey of fifteen planet-forming disks with the ALMA telescope. The new images reveal intricate gas and dust structures in nearly every disk, opening a closer view of where and how planets form.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The paper's own tables contradict its headline: Table 2 lists no continuum substructure for PDS 66, the abstract says 'all but one,' yet Section 5 says all sources show substructure.","rationale":"The reader's weakest assumption identified the biased sample selection as the main threat to the generalization from fifteen selected disks to 'all disks.' That is a valid concern about external validity. However, my reading found a more immediate, internal problem: the manuscript makes three mutually incompatible statements about how many sources show substructure. Table 2 lists PDS 66 with no continuum substructure; the abstract says 'all but one disk'; and Section 5 says 'all exoALMA sources' exhibit dust continuum substructure. This is not a matter of interpretation or consensus; it is a factual inconsistency within the paper's own data products. If PDS 66 is the exception, the scientific headline overstates even the sample-level result. The paper's core value as a data release and technical description remains, and the inconsistency is correctable through rewording and a careful count of Table 2, so the verdict stays CONDITIONAL as the reader recommended. The condition should include reconciling these statements, not only softening the generalization to the broader population. I do not see a reason to move the verdict to REJECT: the observational program, calibration, imaging, and public data release are described in detail and appear credible, and the disputed claim is a summary sentence rather than the dataset itself. The companion papers cited as 'TBD' in Section 5 are a secondary verification concern: the summary of findings cannot be fully checked until those papers appear, but that does not change the verdict because the current paper's primary deliverable is the data products, not the final science interpretation. For the concrete test, a simple table cross-check settles the inconsistency: if PDS 66 is confirmed as the single exception, the correct phrasing is 'all but PDS 66' or '14 of 15,' and the abstract's final claim must be amended accordingly.","tokens_in":26015,"tokens_out":3412,"duration_ms":34234,"concrete_test":"Produce a source-by-source table cross-referencing Table 2's 'Continuum Substructure' column with the abstract's 'all but one disk' and Section 5's 'all exoALMA sources exhibit substructure in their dust continuum.' If PDS 66 is confirmed to have no entry in that column and no molecular substructure is reported for it in the companion analysis (Galloway-Sprietsma et al. 2024), then Section 5 and the abstract's final sentence must be revised to '14 of 15' or 'all but PDS 66.' If PDS 66 does show a substructure type omitted from Table 2, the table must be corrected and the 'all but one' phrasing reconciled.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—that 'all disks appear to exhibit physical and dynamical substructure'—requires that every source in the sample actually shows substructure in dust and/or gas. The manuscript's own data contradict this. In Table 2, the 'Continuum Substructure' column is blank for PDS 66, while every other source lists rings, asymmetries, spirals, inner disks, cavities, or warps. The abstract itself states 'Extensive substructure was found in all but one disk,' identifying PDS 66 as the exception. Yet Section 5, item 1, says 'all exoALMA sources exhibit substructure in their dust continuum (Curone et al. 2024)' and 'all sources also displayed extensive structure in their molecular emission.' This is an internal inconsistency, not a consensus dispute: the descriptive claim about the sample is contradicted by the sample's own catalog. If PDS 66 genuinely lacks continuum substructure, then the abstract's final sentence—'all disks appear to exhibit physical and dynamical substructure'—is false as written; the defensible statement is '14 of 15 disks' or 'all but PDS 66.' The sample-bias concern raised by the reader is real, but this inconsistency is more load-bearing because it undermines the claim even within the selected sample. The accompanying limitation paragraph in Section 3.1 acknowledges the sample is 'decidedly biased,' but that caveat does not repair the internal mismatch between the abstract, Table 2, and Section 5.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper is the overview paper for the exoALMA ALMA Large Program, describing the selection of fifteen protoplanetary disks, the observational setup (12CO, 13CO, CS J=7-6, and 330 GHz continuum), the data products released (fiducial, high-resolution, low-resolution, and RML image sets; moment maps; derived source properties), and a summary of initial results from companion papers. The central scientific claim is that all disks in the sample exhibit physical and dynamical substructure, indicating ongoing dynamical processing by embedded planets, instabilities, or winds.","tokens_in":26297,"tokens_out":4412,"duration_ms":38220,"significance":"If the data release is as described, it is a valuable public resource for the community, providing deep, high-resolution observations of gas and dust in a sample of disks. The benchmarking against synthetic data, the detailed description of multiple imaging sets, and the explicit public release plans are commendable and represent a solid contribution. However, the paper's headline claim of ubiquity of substructure is undermined by both the sample selection and internal inconsistencies regarding PDS 66, so the scientific conclusion requires revision before the paper can be accepted.","major_comments":[{"comment":"The manuscript contains a direct internal contradiction about PDS 66. The abstract states 'Extensive substructure was found in all but one disk', Table 2 leaves the 'Continuum Substructure' column blank for PDS 66, and Section 3.1 says 'all but PDS 66 exhibit gap-like or cavity-like structures in their continuum emission'. Yet Section 5, bullet 1 asserts that 'all exoALMA sources exhibit substructure in their dust continuum (Curone et al. 2024)' and that 'all sources also displayed extensive structure in their molecular emission'. Section 4.2.4 also says 'most sources were found to exhibit structure in their radial emission profiles aside from PDS 66'. These statements cannot all be correct; the text must be revised so that the abstract, Table 2, and Section 5 are mutually consistent, for example by saying '14 of 15 disks' or 'all but PDS 66'.","section":"Abstract, Section 5, Table 2, Section 3.1, Section 4.2.4"},{"comment":"The concluding sentence of the abstract, 'From this sample it is clear that, when observed in detail, all disks appear to exhibit physical and dynamical substructure', overgeneralizes from a sample that the paper itself describes in Section 3.1 as 'decidedly biased'. The selection criteria required large gas extent, low inclination, no envelope contamination, and (with the exception of PDS 66) prior knowledge of gap- or cavity-like dust structures. Even after fixing the PDS 66 inconsistency, this sentence should be explicitly restricted to the exoALMA sample rather than phrased as a claim about protoplanetary disks in general, or it should be accompanied by the stated caveat that generalization is limited by selection.","section":"Abstract, final sentence; Section 3.1"},{"comment":"The claim that 'all sources also displayed extensive structure in their molecular emission' is not demonstrated in this paper and appears to conflict with the abstract's 'all but one' statement. The caption of Figure 1 also asserts 'all sources showing evidence of gas substructures' without qualification. Since the text points to companion papers for details, the authors should clarify whether molecular emission substructure is claimed for all 15 sources or for 14, and ensure the figure caption, abstract, and Section 5 are uniformly worded.","section":"Section 5, bullet 1; Figure 1 caption"}],"minor_comments":[{"comment":"In the description of the high-resolution image sets, 'the sensitives dropped by a factor of ~5' should read 'the sensitivities dropped by a factor of ~5'.","section":"Section 4.2.1"},{"comment":"The caption begins with a stray equals sign: '=The spatial resolution and channel spacing...' should begin with 'The spatial resolution and channel spacing...'.","section":"Figure 2 caption"},{"comment":"The table note says 'zeorth moment maps'; this should be 'zeroth moment maps'.","section":"Table 2 note"},{"comment":"Many companion-paper references are listed with 'TBD' as the journal (e.g., Bae et al. 2024, Curone et al. 2024, Izquierdo et al. 2024). This may be standard for a special issue, but the manuscript should include a note to readers about the publication status of these papers.","section":"References"},{"comment":"The author list includes Charles H. Gardner, but reference 'Gardener et al. 2024' is misspelled with an extra 'e'; please make the spelling consistent.","section":"Author list and references"},{"comment":"In the caption of the appendix figures, '100 ms s−1' should read '100 m s−1'.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The paper depends heavily on companion papers cited as 'TBD', which makes it difficult to independently verify the summarized results. The internal inconsistency about PDS 66 suggests that the companion analyses may not all be finalized; the editor may wish to confirm that the companion papers are available before final acceptance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Thanks for the notes. Practical take: this is a real resource—15 disks with 12CO, 13CO, CS J=7-6 at 0.15 arcsec and ~100 m/s, plus 330 GHz continuum at 90 mas, all scheduled for public release. The team benchmarked the analysis tools against synthetic data (Bae et al. 2024), and the description of the three imaging sets plus the RML products is detailed and usable. This is the kind of paper that will be cited as \"the exoALMA data\" for years. The authors also deserve credit for stating plainly in Section 3.1 that the sample is \"decidedly biased\"—large, bright, low-inclination disks, mostly with known gaps or cavities.\n\nThe soft spots are real, though. The stress-test note lands: Table 2 lists no continuum substructure for PDS 66, the abstract says \"all but one,\" and Section 5 says \"all exoALMA sources exhibit substructure in their dust continuum.\" That is a direct internal contradiction, not a matter of interpretation. The final sentence of the abstract—\"all disks appear to exhibit physical and dynamical substructure\"—is simply false as written if PDS 66 is in the sample. It should read \"14 of 15\" or \"all but PDS 66.\" And the sample bias compounds this: even if every source showed substructure, you cannot infer ubiquity across the disk population from a sample preselected for large, structured disks. The authors know this, but the abstract doesn't reflect it.\n\nI'd also flag that most substantive claims are in companion papers with \"TBD\" citations (Izquierdo et al. 2024, Longarini et al. 2024, Curone et al. 2024). That's normal for a survey paper of this kind, but it means the first-look results are a preview, not a final word. The data release and technical content stand on their own.\n\nBottom line: the flaws are fixable and localized. The paper deserves a serious referee, and I'd send it to review with the expectation of a revision that aligns the abstract and Section 5 with the paper's own Table 2 and softens the population-level language. The disk community will want this data release.","headline":"A valuable ALMA survey data-release paper whose headline ubiquity claim contradicts its own Table 2 and needs a careful revision.","tokens_in":27062,"tokens_out":3350,"would_cite":true,"duration_ms":31018,"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":"A deep ALMA survey finds that planet-forming disks are full of structure, in gas, dust, and motion.","keywords":["protoplanetary disks","planet formation","ALMA observations","molecular line emission","disk kinematics","disk substructure","submillimeter interferometry","exoALMA survey"],"falsifier":"Observe a sample of protoplanetary disks chosen without regard to size, brightness, or known structure at the same spatial and spectral resolution; if a substantial number show smooth molecular emission and purely Keplerian velocity fields, the claim of universal substructure would be refuted. Equivalently, a single disk with no detectable dust or gas substructure and no kinematic perturbation at the 10 m/s level, observed at this depth, would falsify the paper's blanket conclusion.","tokens_in":25803,"feed_emoji":"🪐","tokens_out":9451,"duration_ms":88941,"temperature":0.7,"pith_summary":"The exoALMA Large Program set out to test whether the substructure seen in dust around young stars is matched by structure in the gas and in the gas motions. It observed fifteen large, bright, moderately inclined protoplanetary disks with the Atacama Large Millimeter/submillimeter Array, imaging 12CO, 13CO, and CS line emission at roughly 0.15 arcsecond and 100 m/s resolution plus 330 GHz continuum at 90 mas. The paper reports that all but one disk shows extensive substructure in both dust and molecular line emission, and that every source shows kinematic perturbations in its rotation pattern, with velocity residuals up to about 15 percent of the local rotation. If this holds, planet-forming disks are not smooth rotating gas bodies but are being sculpted by embedded planets, instabilities, or winds, and the publicly released dataset gives the community a way to test and refine that picture.","feed_headline":"Deep ALMA survey finds substructure in 14 of 15 planet-forming disks","feed_subtitle":"Public data reveal rings, gaps, and 10 m/s velocity wiggles pointing to young planets and disk winds at work.","key_machinery":"The load-bearing instrument is the survey design itself: a sample of fifteen large (1 to 7 arcseconds), bright, low-inclination disks and a spectral setup that pushes ALMA Band 7 to a velocity resolution of about 27 m/s while imaging three molecular lines and 0.9 mm continuum. The 12CO and 13CO lines trace the disk atmosphere at different heights, while the heavier CS molecule probes closer to the midplane and is less affected by thermal broadening, making it sensitive to non-thermal motions. From these cubes the team extracts emission surfaces, two-dimensional temperature maps, and rotation curves whose precision reveals 10 m/s-level kinematic wiggles; those wiggles are the diagnostic connecting observed substructure to dynamical causes.","core_discovery":"The paper's central claim is that when protoplanetary disks are observed at sufficient spatial and spectral resolution, physical and dynamical substructure is the norm rather than the exception. For fifteen carefully selected disks, the survey finds that molecular line emission shows rings, gaps, clumps, and azimuthal asymmetries in all but one source, and that maps of the projected gas velocity contain non-Keplerian perturbations in all fifteen. The velocity measurements reach uncertainties of order 10 m/s, and deviations from a simple Keplerian rotation model are attributed to local pressure variations, self-gravity of the disk, and unresolved dynamical processes. The paper frames these observations as evidence of ongoing dynamical processing by young embedded planets, magneto-hydrodynamical instabilities, or winds, and presents the full data release as the evidence base for that conclusion.","pith_inferences":["If the pattern of ubiquitous substructure survives a less biased sample, smooth-disk assumptions in planet formation population synthesis would need revision, since most disks would begin with radial structure that sets where planets can form.","A direct test would be to apply the same 0.15 arcsecond, 100 m/s observing strategy to a sample of disks selected without any prior knowledge of dust gaps; the exoALMA result predicts that nearly all will show gas substructure and kinematic perturbations.","The velocity-amplitude versus planet-mass relation implied by these observations could be calibrated by targeting disks with known directly imaged planets, such as PDS 70, to turn the kinematic amplitude into a planet mass estimator.","Comparing the three molecular tracers' kinematic signatures offers a height-resolved view of instabilities; a future analysis could test whether the perturbation amplitude changes with height as expected for midplane-driven planets versus vertically extended instabilities."],"forward_implications":["Kinematic perturbations become a practical tool for finding young planets: the observed 100 m/s-scale kinks in channel maps and velocity residuals can be matched to planet-driven spiral wakes in disks at typical ALMA distances.","Pressure maxima coincident with dust rings act as dust traps, so the correlation between velocity residuals and continuum rings strengthens the case that millimeter-sized grains are dynamically concentrated where gas surface density peaks.","Rotation curves that deviate from a simple Keplerian model by self-gravity yield dynamical disk masses for ten sources, providing an independent, chemistry-free mass measurement that can calibrate CO-based estimates.","The public release of calibrated data, image cubes, and moment maps lets the community reanalyse all fifteen disks with other methods, turning a single survey into a reusable benchmark for disk modelling codes."],"supporting_citations":[{"why":"Documents the ubiquity of dust rings and gaps that motivates searching for matching gas structure.","marker":"Andrews 2020"},{"why":"Reviews planet-disk interaction signatures and the dynamical processes invoked to explain substructure.","marker":"Bae et al. 2023"},{"why":"Introduces the kink channel-map signature used to link kinematic perturbations to embedded planets.","marker":"Pinte et al. 2018a"},{"why":"Demonstrates that CO line centroids can be measured to a few m/s, justifying the 100 m/s channel sampling and the 10 m/s velocity precision claims.","marker":"Teague et al. 2018a"},{"why":"Its CO isotopologue images of five disks set the target 0.1 arcsecond resolution for gas morphology.","marker":"Öberg et al. 2021"},{"why":"Provides the ring and gap identification method and radial profile tools used to catalogue continuum substructure.","marker":"Huang et al. 2018a"},{"why":"Supplies the parallax distances used for physical scales and stellar mass derivation.","marker":"Gaia Collaboration et al. 2023"},{"why":"Classifies each source's continuum substructure types listed in Table 2 and analysed across the sample.","marker":"Curone et al. 2024"},{"why":"Describes the discminer modelling that yields inclinations, position angles, and the rotation curves used for the kinematic perturbation claims.","marker":"Izquierdo et al. 2024"}],"fun_headline_variants":["Nearly all disks show substructure and velocity anomalies","10 m/s precision reveals non-Keplerian motions in all disks","exoALMA: substructure is the norm in planet-forming disks","Deep ALMA data: substructure and wiggles in 14 of 15 disks","All 15 disks show non-Keplerian velocity perturbations"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The conclusion that all disks exhibit substructure when observed closely rests on a sample the authors themselves describe as decidedly biased: the disks were preselected to be large, bright, moderately inclined, free of envelope contamination, and, with one exception, already known to host gap- or cavity-like dust structures.","fun_headline_variants_meta":{"raw":{"variants":["Nearly all disks show substructure and velocity anomalies","10 m/s precision reveals non-Keplerian motions in all disks","exoALMA: substructure is the norm in planet-forming disks","Deep ALMA data: substructure and wiggles in 14 of 15 disks","All 15 disks show non-Keplerian velocity perturbations"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001818,"raw_usage":{"total_tokens":7212,"prompt_tokens":1062,"completion_tokens":6150,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":678,"completion_tokens_details":{"reasoning_tokens":6067}},"tokens_in":678,"tokens_out":6150,"duration_ms":36048,"temperature":1.0,"reasoning_tokens":6067,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T10:11:53.849099+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe a sample of protoplanetary disks chosen without regard to size, brightness, or known structure at the same spatial and spectral resolution; if a substantial number show smooth molecular emission and purely Keplerian velocity fields, the claim of universal substructure would be refuted. Equivalently, a single disk with no detectable dust or gas substructure and no kinematic perturbation at the 10 m/s level, observed at this depth, would falsify the paper's blanket conclusion.","supporting_citations":[],"review_version":1}