{"id":"3f178ced-fdf5-4984-8e56-88c46ddc0612","arxiv_id":"1909.02591","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In a complete ALMA survey of embedded protostars in Ophiuchus, 38% are detected in dust polarization, and most detections are consistent with dust self-scattering in inclined, optically thick disks rather than magnetic fields.","lead":"Astronomers mapped dust polarization around all 37 embedded protostars in the Ophiuchus cloud with ALMA at 35 au resolution. Most detections trace light scattering inside disks, not magnetic fields, which changes how disk magnetism should be studied.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Self-scattering classification hinges on the β<0.5 optical-depth proxy, which the paper's own Appendix C admits is uncalibrated and can be flattened by temperature gradients, cold dust, or scattering.","rationale":"The paper's central interpretation is that most disk-scale polarization in embedded protostars arises from self-scattering, not magnetic alignment. The discriminator in Section 4.2 is the dust opacity index β: sources with β<0.5 are deemed optically thick and assigned to self-scattering, while β>0.5 sources are candidates for magnetic alignment. This mapping is load-bearing because Section 4.2.2 excludes magnetic alignment for thick disks on the authority of Yang et al. (2017), and Table 9 then yields the key 9/14 statistic. The paper is commendably open about the proxy's limits in Appendix C, citing temperature gradients, grain properties, and scattering as flattening agents, and it explicitly states that a true opacity determination requires multi-wavelength modeling beyond the paper. But the concern is not merely academic: several β values sit close to the 0.5 boundary (VLA 1623A: 0.45±0.29; VLA 1623B: 0.48±0.08; IRS 63: 0.35±0.19), and IRS 37-A is classified as scattering without any β measurement. A Rayleigh-Jeans correction at realistic dust temperatures will raise the inferred β for the 233–350 GHz pairs; if any of the nine 'thick' sources shift above 0.5 under that correction, the optical-depth gate fails and the classification loses its basis. The proposed recomputation is simple and would settle the point. I therefore agree with the reader that the paper should be CONDITIONAL pending this check, and I do not move the verdict; the survey itself is a valuable data release with rigorous calibration, debiasing, and reproducibility, and the concern is about the interpretation, not the data.","tokens_in":60081,"tokens_out":13393,"duration_ms":138723,"concrete_test":"Recompute β for the nine sources marked 'thick' in Table 9 using the same 233/349 GHz flux pairs listed in Table C1, replacing the RJ assumption α=β+2 with a full Planck-function fit at T=20 K and T=10 K. If any source's temperature-corrected β rises above 0.5, the β<0.5→thick mapping in §4.2.1 is not robust and the self-scattering assignments for those sources should be revisited.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that 9/14 detected sources are consistent with dust self-scattering in optically thick disks (abstract; Table 9) is anchored by the mapping β<0.5→optically thick, stated in §4.2.1. Appendix C explicitly acknowledges that β is only a proxy: grain properties, disk temperature gradients (Shetty et al. 2009), and dust scattering can flatten the spectral index without the emission being optically thick. Several inferred β values for sources marked 'thick' sit close to the 0.5 threshold: VLA 1623A (0.45±0.29), VLA 1623B (0.48±0.08), and IRS 63 (0.35±0.19). IRS 37-A has no β measurement at all, yet is classified 'yes' in Table 9 based on morphology and inclination alone. A Rayleigh-Jeans (RJ) correction for plausible dust temperatures (the paper itself uses T=20 K for mass estimates) lowers the effective spectral slope between 233 and 350 GHz, which would raise the inferred β; for the near-threshold sources, this could push β above 0.5. If the optical-depth proxy fails, the gate in §4.2.2 excluding magnetic and other alignment mechanisms for thick disks (Yang et al. 2017) is no longer closed, and the conclusion that 'dust polarization is not a good tracer of magnetic fields on ≲100 au scales on average' loses its statistical foundation. This is a correctness risk, not a stylistic one: the survey data, calibration, debiasing, and reproducibility are strong, but the physical interpretation is not uniquely determined without an independent optical-depth calibration.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper presents ALMA 1.3 mm full-polarization observations of 37 young stellar objects in Ophiuchus, including all embedded protostars in the cloud, at a common resolution of about 35 au. The survey detects polarized emission in 14 of 37 YSOs, and the authors classify the polarization morphologies as uniform (nine sources), azimuthal (four sources), or complex (two sources). On the basis of morphology, inclination, and a dust opacity index β estimated from archival fluxes, they conclude that the majority of detections (9/14) are consistent with dust self-scattering in optically thick disks, while the remaining sources are candidate tracers of magnetic fields with inferred poloidal or hourglass field geometries. The paper also reports disk masses and sizes, clump-scale comparisons, multiplicity statistics, galaxy contamination estimates, and it releases the continuum and polarization maps. The central interpretive claim is that dust polarization is not, on average, a good tracer of magnetic fields on ≲100 au scales.","tokens_in":60377,"tokens_out":6828,"duration_ms":77403,"significance":"If the interpretation holds, this is the first large, homogeneous, unbiased sample showing that disk-scale submillimeter polarization around embedded protostars is dominated by self-scattering rather than magnetic grain alignment, and it provides quantitative guidance for future polarization studies of embedded disks. The strengths of the paper are substantial: a uniform observing setup and reduction, a careful debiasing treatment in Section 2.4, an explicit validation of off-axis polarization in Appendix B, a quantitative galaxy contamination model in Section 5.5, and a public release of the data products. The authors are also transparent about many uncertainties, including the non-uniqueness of source classifications and the proxy nature of β. The main scientific risk is that the central 9/14 classification is conditioned on the β<0.5 optical-depth threshold, which the paper itself acknowledges to be an uncalibrated proxy.","major_comments":[{"comment":"In Table C1, for sources without quoted flux errors the text says 'we assume 10% errors,' but several rows (e.g., WL 17, Elias 29, Oph-emb-6) quote errors that are smaller than 10% of the flux. Please clarify which errors are adopted in those cases, or apply the 10% assumption uniformly as stated.","section":"Table C1"}],"minor_comments":[{"comment":"In Conclusion 7, 'protoplanetery' is a typo for 'protoplanetary.'","section":"Section 6"}],"recommendation":"major_revision","confidential_remarks":"This is a strong observational survey paper with careful data reduction, public data release, and a clear presentation of the main morphological results. My major concerns are all fixable within the scope of the manuscript: the β-proxy robustness test, the IRS 37-A classification, and the quantification of the azimuthal morphology. I would not reject the paper, and I do not think the issues require new observations; they require a more careful accounting of the uncertainties that the authors themselves already identify. If the suggested sensitivity test is performed and the claims are appropriately qualified, the paper would be a valuable contribution to the field."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The thing to know: this is the largest homogeneous ALMA dust polarization survey of embedded protostars to date, it comes with a full data release, and it will be a reference dataset whether or not the physical interpretation holds. The survey-level results—38% detection rate, the uniform-versus-azimuthal morphology split, and the inclination trend (uniform for i > 60°, azimuthal for i < 60°)—are genuinely new and cleanly presented.\n\nThe data work is careful: debiasing is handled properly, the off-axis validation in Appendix B is a real service to the community, and the non-detection upper limits let others use the sample statistically. The galaxy contamination estimates and the multiplicity census are useful secondary products. I ran out of reasons to distrust the measurements.\n\nThe soft spot is the step from morphology to mechanism. The paper classifies 9 of 14 detections as self-scattering using β < 0.5 as the optically thick threshold, and Appendix C honestly says β is a proxy that can be flattened by temperature gradients, grain properties, or scattering. That is more than a stylistic caveat: several sources sit right at the threshold (VLA 1623A at 0.45±0.29, VLA 1623B at 0.48±0.08, IRS 63 at 0.35±0.19), and IRS 37-A is classified as scattering with no β measurement at all. A Rayleigh–Jeans correction using the paper’s own 20 K dust temperature would steepen the inferred β and could push the near-threshold sources over 0.5. If that happens, the gate in §4.2.2 that excludes magnetic alignment for thick disks no longer closes, and the strong statement that dust polarization is “not a good tracer of magnetic fields on ≲100 au scales on average” loses its statistical footing. The magnetic-field subset is already small—five sources, really four once GY 91 is set aside—so the no-toroidal-field conclusion is correspondingly fragile.\n\nThat said, the inclination–morphology relation is empirical and does not depend on β. Uniform polarization along the minor axis in highly inclined disks is what scattering models predict and is not the expected signature of poloidal or toroidal magnetic alignment, so the self-scattering interpretation has independent support beyond the β proxy. The paper just needs to be honest about how much weight each leg carries.\n\nMinor point: GY 91 is treated as detected in Table 5 but as non-detected in Table 7; the text calls it marginal, which is fine, but the tables should be consistent.\n\nWho is this for? anyone observing or modeling dust polarization in star-forming regions, especially with ALMA. It deserves a serious referee and, with a sensitivity analysis of the β threshold and a clearer separation of the empirical trends from the interpretive classification, it will be a solid anchor paper. I would engage with it and would bring it to a reading group.","headline":"A valuable, well-executed survey with a plausible but proxy-dependent central interpretation; the inclination–morphology trend should outlive the self-scattering classification, but the β threshold needs a sensitivity check.","tokens_in":60968,"tokens_out":3051,"would_cite":true,"duration_ms":38757,"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":"Most disk-scale dust polarization around embedded protostars traces scattering, not magnetic fields.","keywords":["dust polarization","protostars","Ophiuchus","ALMA","dust self-scattering","magnetic fields","protoplanetary disks","millimeter continuum"],"falsifier":"Resolve the nine candidate self-scattering disks at two ALMA bands, such as 0.87 mm and 3 mm, and measure spectral index and polarization maps. If beta < 0.5 is an optical-depth indicator, the short-wavelength emission should be optically thick and the polarization pattern should match scattering models in both bands; if beta is flattened by temperature gradients or scattering, the inferred optical depth and the self-scattering assignment would not survive, and magnetic-alignment alternatives would remain viable.","tokens_in":59872,"feed_emoji":"🔭","tokens_out":6408,"duration_ms":64098,"temperature":0.7,"pith_summary":"This paper reports the largest homogeneous ALMA survey of 1.3 mm dust polarization toward embedded protostars, covering all the Class 0 and Class I objects in the Ophiuchus cloud at 35 au resolution. It finds that only 14 of 37 young stellar objects are detected in polarization, and that most of those detections, 9 of 14, have morphologies and disk properties consistent with dust self-scattering in optically thick disks rather than with grain alignment by magnetic fields. If this reading is right, dust polarization on scales below about 100 au is usually not a magnetic-field tracer, and inferred field morphologies from such data must be treated cautiously. The survey also proposes that highly inclined disks, with inclinations above about 60 degrees, are the best targets for future scattering studies because their polarization patterns are uniform and do not require the disk to be fully resolved.","feed_headline":"Nine of 14 young disks show scattering, not magnetic fields","feed_subtitle":"A complete ALMA survey of Ophiuchus protostars finds dust polarization is not a reliable field tracer on disk scales.","key_machinery":"The load-bearing object is the predicted polarization pattern of dust self-scattering: Rayleigh scattering by roughly 200 micron grains in an optically thick disk produces e-vectors aligned with the disk minor axis for inclinations above about 60 degrees and an azimuthal pattern with central depolarization for near-face-on disks. The paper classifies each detection by comparing the observed morphology to that template and by estimating optical depth from the dust opacity index beta, obtained from the millimeter spectral index alpha = beta + 2 using archival fluxes; beta < 0.5 is taken to mean optically thick emission, where magnetic alignment is suppressed and scattering can dominate. The same template is used in reverse: sources with beta > 0.5 and non-scattering morphology are interpreted as magnetic-field tracers, with the field direction recovered by rotating the polarization e-vectors by 90 degrees.","core_discovery":"The central claim is that, in a complete, unbiased sample of embedded protostars, dust polarization on disk scales is dominated by dust self-scattering rather than by magnetically aligned grains. Nine of the fourteen detected sources have uniform or azimuthal polarization patterns that match scattering predictions for optically thick disks, while the remaining sources are inconsistent with scattering and may instead trace magnetic fields; rotating those polarization vectors by 90 degrees yields mainly poloidal or hourglass-shaped fields, with no evidence of a strong toroidal component toward any disk. The paper therefore concludes that dust polarization is not a good tracer of magnetic fields on scales below about 100 au on average, in contrast to envelope-scale polarization surveys where nearly all sources are detected.","pith_inferences":["If the self-scattering interpretation holds, millimeter dust polarization of embedded disks becomes a probe of grain growth and disk structure rather than of field geometry, and magnetic-field studies on these scales will need other tracers such as molecular-line polarization.","The inclination-morphology dichotomy predicts that deeper observations of moderate-inclination disks should reveal azimuthal patterns once the disk is resolved, while unresolved disks at lower inclinations should appear depolarized, which could explain many of the non-detections.","Multi-wavelength ALMA observations across the nine candidate scattering disks would test the mechanism directly: the polarization fraction and pattern should vary with wavelength following scattering models, whereas magnetic alignment would not show the same wavelength dependence.","The beta < 0.5 optical-depth proxy could be checked with resolved temperature and optical-depth maps; if beta is flattened by temperature gradients or scattering rather than true high optical depth, the number of genuine scattering detections among the 14 would shrink."],"forward_implications":["On scales below about 100 au, ALMA dust polarization detections around embedded protostars should not be assumed to trace magnetic fields; in an unbiased sample, most detections are scattering signatures.","Disks with inclinations above about 60 degrees are the best targets for self-scattering studies because their uniform polarization can be recovered without fully resolving the disk.","No strong toroidal magnetic field component is found toward any disk; the inferred fields are poloidal or hourglass-shaped, which bears on magnetic braking and disk formation.","Most disks in Ophiuchus are low-mass, below about 10 Jupiter masses, and compact, below about 30 au, and disk mass and size are correlated with a surface density relation of about r^-0.9 to r^-0.6, matching Class II disks.","Roughly half of the undetected sources have 3-sigma upper limits below 2 percent, indicating that many disks have intrinsically low polarization fractions, possibly from unresolved azimuthal structure or from limited large-grain populations."],"supporting_citations":[{"why":"Establishes the self-scattering mechanism by which large grains in an optically thick disk produce a few percent polarization.","marker":"Kataoka et al. 2015"},{"why":"Provides the scattering polarization morphologies versus disk inclination used to classify uniform versus azimuthal patterns.","marker":"Yang et al. 2016"},{"why":"Shows magnetic-alignment polarization is suppressed in optically thick emission, the basis for assigning thick disks to scattering.","marker":"Yang et al. 2017"},{"why":"ALMA detection of HL Tau polarization demonstrating scattering morphology and wavelength dependence used for comparison.","marker":"Stephens et al. 2017"},{"why":"Multi-wavelength 872 micron observations of VLA 1623 that confirm the 1.3 mm polarization angles and support the mechanism split.","marker":"Harris et al. 2018"},{"why":"Paper I, presenting the VLA 1623 polarization data on which several classifications in this survey rest.","marker":"Sadavoy et al. 2018a"},{"why":"The archival spectral index showing IRAS 16293B has beta near zero, used to mark that source as optically thick.","marker":"Chandler et al. 2005"},{"why":"Envelope-scale polarization survey whose high detection rate is the contrast for the claim that disk-scale polarization is not a good field tracer.","marker":"Hull et al. 2014"}],"fun_headline_variants":["Scattering dominates disk polarization in Ophiuchus protostars","ALMA survey: dust scattering, not fields, shapes disk polarization","Ophiuchus disks: self-scattering beats magnetic alignment","14 polarized disks: 9 scattering, 5 possibly magnetic"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The self-scattering classification rests on the assumption that a dust opacity index beta < 0.5, derived from the flux spectral index with alpha = beta + 2, reliably identifies optically thick emission; if scattering or disk temperature gradients flatten the spectral index without true high optical depth, several of the nine thick-disk assignments would weaken.","fun_headline_variants_meta":{"raw":{"variants":["Scattering dominates disk polarization in Ophiuchus protostars","ALMA survey: dust scattering, not fields, shapes disk polarization","Ophiuchus disks: self-scattering beats magnetic alignment","14 polarized disks: 9 scattering, 5 possibly magnetic"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000736,"raw_usage":{"total_tokens":3343,"prompt_tokens":1053,"completion_tokens":2290,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":669,"completion_tokens_details":{"reasoning_tokens":2217}},"tokens_in":669,"tokens_out":2290,"duration_ms":17958,"temperature":1.0,"reasoning_tokens":2217,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T04:45:42.466029+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Resolve the nine candidate self-scattering disks at two ALMA bands, such as 0.87 mm and 3 mm, and measure spectral index and polarization maps. If beta < 0.5 is an optical-depth indicator, the short-wavelength emission should be optically thick and the polarization pattern should match scattering models in both bands; if beta is flattened by temperature gradients or scattering, the inferred optical depth and the self-scattering assignment would not survive, and magnetic-alignment alternatives would remain viable.","supporting_citations":[{"cited_title":"W., Girart, J","cited_arxiv_id":null,"evidence_quote":"Shows magnetic-alignment polarization is suppressed in optically thick emission, the basis for assigning thick disks to scattering."}],"review_version":1}