{"id":"d5ed0083-f0d7-4969-b6a7-c4e9a95f8588","arxiv_id":"2505.06044","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":9,"one_line_summary":"A shadow and horizon fitting framework estimates the tilt, orientation, and surface height of misaligned inner and outer disks in transition disks; for HD 100453 it gives a 70-degree misalignment and an inner disk aspect ratio of 0.17.","lead":"Astronomers built a new way to read scattered-light images of transition disks, using shadow shapes and the visible edge of the outer disk to measure the tilt and height of the disks. Applied to HD 100453, it finds the inner disk is misaligned by about 70 degrees with a maximum aspect ratio of 0.17, matching earlier infrared interferometry.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No synthetic recovery test: the claimed simultaneous inference of ii, PAi, hr, and the outer surface profile is only validated on one real object, so biases from the assumed Eq. (6) surface family and vertically symmetric inner disk (Eq. 18) are not quantified.","rationale":"The reader's conditional verdict is appropriate. The forward model is coherent, and the agreement of the recovered inner-disk orientation with the independent GRAVITY result is real supporting evidence, as is the analytic derivation of the horizon (Eq. 13) and the shadow quartic (Appendix B). The soft spot is not internal inconsistency but model-misspecification bias: the parameterized surface family and vertical symmetry are acknowledged to be empirical or approximate, and the single real-data demonstration cannot distinguish a correct inversion from a biased one that happens to match one object. The only direct way to establish the central claim is to show that known inputs are recovered from synthetic data. This is feasible because the forward model is analytic and the gradient extraction is deterministic. Absent that test, the headline numbers — especially hr = 0.17 with uncertainty 0.001 and the outer-disk H proportional to R conclusion — are conditional on the assumed functional forms. Thus the verdict should remain CONDITIONAL rather than ACCEPT or REJECT; if the injection-recovery test fails, the framework would need major revision. This is a standard validation gap for a method paper, not a claim of any defect in intent or effort.","tokens_in":25415,"tokens_out":10103,"duration_ms":113539,"concrete_test":"One check settles this: injection-recovery on synthetic images. Generate SPHERE-like polarized-intensity images (or, minimally, synthetic azimuthal and radial gradient maps) from the analytic model with known parameters, using a surface height profile deliberately different from Eq. (6) — for example a flared power law H = H0 (R/R0)^1.3 or a profile with a local height bump — and with a vertically asymmetric inner disk (upper surface hr plus delta, lower surface hr minus delta). Add realistic noise, run the same CLAHE/Canny edge extraction and parallel-tempered MCMC pipeline, and ask whether the input ii, PAi, hr, Hc, and alpha fall inside the reported credible intervals across multiple noise realizations.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim — that shadow boundaries plus the horizon determine ii, PAi, hr, and the outer surface height profile — requires that the forward model faithfully represent the real system. That model is built on two structural premises: the outer disk surface is single-valued, axisymmetric, and of the empirical form Eq. (6), which Appendix A explicitly states is not derived from first principles; and the inner disk shadow is cast by a vertically symmetric cone with a single maximum aspect ratio hr (Eq. 18). Section 4.3 itself acknowledges that MHD winds can make the inner disk vertically asymmetric, and Appendix A concedes that the outer profile is only a sufficient approximation in the inner region. If either premise is violated, the analytic shadow and horizon curves are displaced, and the MCMC posterior will shift ii, PAi, hr, Hc, and alpha to compensate. The paper applies the pipeline to HD 100453 and reports hr = 0.170 ± 0.001, but these error bars are statistical only; no synthetic-image recovery experiment is presented, so the size of the structural bias is unknown. The agreement of ii and PAi with VLTI/GRAVITY is genuine support, but it does not validate hr or the outer surface profile, which are the quantities most sensitive to the parametric family. The flared model in Section 4.2 frees beta and finds beta close to 1, partially addressing the built-in H proportional to R asymptote of Eq. (6), but it remains within the same single-valued axisymmetric family and introduces the extra correlated parameter Rout.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript introduces an analytic framework for inferring the three-dimensional geometry of transition disks with misaligned inner disks. The outer disk is modeled as an axisymmetric, vertically single-valued surface with an empirical height profile (Eq. 6), and the inner disk shadow is modeled as a cone with a single maximum aspect ratio (Eq. 18). The horizon of the outer disk is defined by NS·ez=0 and yields Eq. (13); the shadow boundary is obtained as the solution of a quartic equation (Appendix B). The model is fitted to VLT/SPHERE J-band polarized-intensity images of HD 100453 using MCMC, yielding ii=50.5°, PAi=88.3°, misalignment ~70°, hr=0.17, and an outer surface that converges to H∝R. A flared extension (Section 4.2) gives beta≈1. The paper claims the method resolves parameter degeneracies and provides new constraints on the vertical structure of both disks.","tokens_in":25706,"tokens_out":8974,"duration_ms":94273,"significance":"If the framework works as claimed, it would be a useful addition to the transition-disk toolbox, because it extracts geometric information from scattered-light shadows without full radiative-transfer modeling. The analytic expressions for the horizon and shadow boundary are a genuine improvement over the thin-disk shadow model of Min et al. (2017). The recovered inner-disk inclination is consistent with an independent VLTI/GRAVITY measurement, which is a valuable external check. The main limitation is that the central inversion is validated on only one real object and never on synthetic images with known input geometry; the empirical outer-surface family (Eq. 6/A1) and the symmetric-cone inner disk (Eq. 18) are structural assumptions whose systematic effect on hr and the outer surface profile is not quantified. These issues, not the analytic geometry, are what stand between the current manuscript and a robust method paper.","major_comments":[{"comment":"The likelihood used in the MCMC fit is never defined. The text states only that the model minimizes \"the distance between these curves and the fast gradient points,\" but does not specify the data-point selection, the distance metric, the noise model, or the treatment of outliers. Without these definitions, the quoted posterior uncertainties (e.g., hr=0.170±0.001, PAi=88.3±0.5) and the claimed breaking of the PA–i degeneracy cannot be reproduced or interpreted. Please provide an explicit likelihood and data model.","section":"§2.4, Table 1"},{"comment":"The flared model changes the recovered inner-disk orientation far beyond the statistical errors of the fiducial model: ii shifts from 50.5°±0.8° to 44.5°±1.0° and PAi from 88.3°±0.5° to 83.4°±0.8°. The text in §4.2 calls these results consistent, but the shifts are several times the reported uncertainties. This indicates that the inner-disk geometry is sensitive to the assumed outer-disk surface parameterization, and the current error bars do not capture the dominant systematic model uncertainty. The authors should reconcile this discrepancy or explicitly report the model-dependent range of ii and PAi.","section":"§4.2, Tables 1 and 2"},{"comment":"The inner disk is modeled as vertically symmetric with a single maximum aspect ratio hr, and Section 4.3 itself notes that MHD winds can make the two surfaces asymmetric. If the upper and lower surfaces differ, the shadow boundary curves are displaced and the fitted ii, PAi, and hr will be biased. The paper does not test this sensitivity with synthetic images or with an asymmetric two-surface model. At minimum, the authors should quantify how much the inferred parameters shift when the front and back surfaces are allowed to have different heights, or when h(r) is replaced by a non-conical profile, before claiming that hr is constrained.","section":"§2.2, §4.3, Eq. (18)"},{"comment":"No synthetic-image recovery experiment is presented. The paper's central claim—that shadow boundaries plus the horizon simultaneously determine ii, PAi, hr, and the outer surface profile (Re, Hc, alpha)—requires demonstrating that the inversion recovers known input values when applied to synthetic scattered-light images. This is especially important because Eq. (6)/(A1) is explicitly empirical and not derived from first principles; if the true outer surface deviates from this family, the analytic curves will be displaced and the MCMC will compensate by shifting other parameters. The GRAVITY check validates the inclination, but it does not validate hr or the outer surface height profile. Please add injection-recovery tests for at least the fiducial and flared families, and ideally for a surface generated by a different functional form.","section":"§3, Appendix A"}],"minor_comments":[{"comment":"Equation (1) defines the outer-disk normal vector but the text says \"where ii is the inclination of the outer disk\"; this should be io, and the notation conflicts with the inner-disk inclination used in Eq. (14).","section":"§2.1, Eq. (1)"},{"comment":"The reproducibility of the pipeline is incomplete: the CLAHE and Canny parameters, the definition of the mask used to isolate shadows, and the MCMC chain settings (number of walkers, steps, burn-in, convergence diagnostics) are not reported.","section":"§2.3–2.4"},{"comment":"The four roots of the quartic in Eq. (B4) are classified by signs, but the manuscript does not describe how the correct roots are tracked across the parameter space during the MCMC walk; a brief description or a figure showing all four candidate curves would help.","section":"Appendix B"},{"comment":"The statement that beta≈1 \"suggests that the disk is not significantly flared\" should be tempered, because the fiducial model in Eq. (6) already asymptotes to H∝R; the flared model is nested, so beta close to 1 is partly built into the parameterization and is not an independent empirical result.","section":"§4.2"},{"comment":"The caption contains the typo \"azimutal\" instead of \"azimuthal.\"","section":"Figure 8 caption"},{"comment":"The manuscript does not state whether the MCMC code and data products will be made available; given the method's complexity, a public implementation or a detailed pseudocode appendix would materially aid reproducibility.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper is within the scope of a specialized astrophysics journal and the analytic framework is worth publishing after revision. The main risk is overclaiming: the model-dependent shifts in Tables 1–2 and the absence of synthetic recovery tests should be addressed before the paper can serve as a reliable methodological reference. I do not see grounds for rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this deserves a serious referee. The geometry is clean, the new bits are real, and the application to HD 100453 is honest about its own limitations.\n\nWhat is actually new: Min et al. (2017) treated the inner disk as infinitesimally thin, which left a wide degeneracy in inclination and position angle. This paper adds a finite maximum aspect ratio hr, lets the outer disk surface height vary radially through an empirical profile (Eq. 6), and fits the horizon and shadow boundaries simultaneously. That is a genuine technical extension, and it resolves a known degeneracy. The recovered inner disk orientation (ii ~ 50.5 deg, PAi ~ 88.3 deg) is consistent with the independent VLTI/GRAVITY measurement from Bohn et al. (2022), which is real supporting evidence, not just internal consistency.\n\nThe soft spots are the ones the stress-test note flags. There is no synthetic-image recovery experiment. The outer disk PA and inclination are fixed from prior work without propagating their uncertainty. The outer surface profile is explicitly empirical, and Appendix A concedes it is not derived from first principles. The inner disk is modeled as vertically symmetric, and Section 4.3 itself acknowledges that MHD winds can break that symmetry. The reported error bars are statistical only. The flared model in Section 4.2 partially addresses the H-proportional-to-R asymptote by freeing beta, and finds beta close to 1, but it stays in the same parametric family, so it does not fully answer the question of structural bias.\n\nNone of this is fatal for a proof-of-concept applied to one object. The paper does not hide its limitations, and the geometric derivation can be checked from the text. But the central claim is that shadow boundaries plus the horizon determine ii, PAi, hr, and the outer surface profile. To support that claim as a general method, the authors need an injection-recovery test with synthetic images, including at least one vertically asymmetric inner disk and one outer surface that deviates from Eq. (6). Until then, I would trust the orientation parameters more than hr or the outer surface details.\n\nWho is this for: anyone working on transition disk shadows, inner disk misalignment, or scattered-light disk geometry. It is a useful method paper to cite. I would bring it to a reading group, mostly for the clean analytic formulation and the cautionary note on structural assumptions.\n\nRecommendation: send to peer review. The derivation is checkable, the application is honest, and the missing synthetic test is a discussable revision, not a desk-reject reason.","headline":"Solid incremental method paper: the new geometry is real, the orientation check against GRAVITY is genuine, and the missing synthetic recovery test is the one thing to fix before this framework becomes a general tool.","tokens_in":26279,"tokens_out":1915,"would_cite":true,"duration_ms":21377,"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":"A single near-infrared image of a shadowed transition disk can reveal the misalignment, thickness, and height profile of both its inner and outer disks.","keywords":["transition disks","protoplanetary disks","disk shadows","scattered light","disk geometry","inner disk misalignment","disk surface height","HD 100453"],"falsifier":"Take a transition disk whose outer surface is known from independent data, such as resolved molecular-line emission or radiative-transfer fits, and whose inner disk is vertically asymmetric or warped; if the fitted shadow curves cannot reproduce the observed shadow edges to within the gradient uncertainties, or if the inferred $h_r$ disagrees with the surface height measured at the same radius by an independent tracer, the central assumption of a symmetric, single-valued surface is contradicted. A cheaper check is to measure the two shadow boundary widths in HD 100453 separately, since a vertically symmetric slab predicts nearly equal front and back widths while an MHD wind predicts a measurable difference.","tokens_in":25146,"feed_emoji":"🪐","tokens_out":6107,"duration_ms":59226,"temperature":0.7,"pith_summary":"This paper presents an analytic framework that reads the geometry of a transition disk system from the shadow cast by a misaligned inner disk onto the outer disk and from the horizon, the line on the outer surface separating visible from hidden surface. The method models the outer surface as an axisymmetric optically thick surface with an empirical height profile and the inner disk as a slab of finite thickness, then derives analytic curves for the shadow boundaries and the horizon. Fitting these curves to the gradient structure of a single scattered-light image simultaneously recovers the inner disk's inclination, position angle, and maximum aspect ratio, plus the outer disk's inner edge radius, surface height, and curvature. Applied to HD 100453, it yields a ~70-degree misalignment and an inner-disk aspect ratio of 0.17, a value several times the gas pressure scale height that the authors attribute to vertical lofting by turbulence or an MHD wind. The significance is that one image can now constrain both disks' 3D structure, breaking degeneracies that limited earlier shadow analyses.","feed_headline":"Shadow curves reveal a 70-degree misaligned disk","feed_subtitle":"A new analytic fit turns one scattered-light image into orientation, thickness, and surface height for both disks.","key_machinery":"The load-bearing objects are two families of analytic curves on the outer disk surface. The horizon is defined by the condition $N_S \\cdot \\hat{e}_z = 0$, where $N_S$ is the surface normal of the outer disk, and takes the closed form $\\phi_{o,\\mathrm{Hor}} = \\arcsin\\big(\\frac{dR}{dH} \\frac{\\hat{n}_o\\cdot\\hat{e}_z}{\\hat{m}_o\\cdot\\hat{e}_z}\\big)$ together with its supplement. The shadow boundary is the intersection of the outer surface with the inner disk's optically obscured slab $S_i = r(\\cos\\phi_i\\,\\hat{l}_i + \\sin\\phi_i\\,\\hat{m}_i) \\pm h_r r\\,\\hat{n}_i + c_i$; eliminating $r$ and $\\phi_i$ reduces this intersection to a quartic in $x = \\sin(\\phi_o)$ with coefficients $b_0, b_1, b_2$. These curves are fit to the image's gradient features, where radial gradients trace the horizon and azimuthal gradients trace the shadow edges, using parallel-tempered Markov chain Monte Carlo sampling. The framework converts a scattered-light image into a parameter estimation problem over the outer-disk center, inner edge, surface height and curvature, and the inner-disk inclination, position angle, and maximum aspect ratio.","core_discovery":"The central claim is that the boundary curves of a shadow and the horizon of the outer disk surface contain enough information to determine, simultaneously, the orientation and thickness of a misaligned inner disk and the full height profile of the outer disk's scattering surface. Earlier analytic shadow models neglected inner-disk thickness, producing infinitesimally narrow shadows and degeneracy between inclination and position angle, while treating the outer scattering surface as having a constant aspect ratio displaced the shadow positions. By modeling the inner disk as an optically thick slab of maximum aspect ratio $h_r$ and the outer disk as an axisymmetric surface with an empirical height profile $R(H)$ given by Eq. (6), the paper derives analytic expressions for both curve families: the horizon via $N_S \\cdot \\hat{e}_z = 0$, and the shadow boundary as the intersection of the two surfaces, which reduces to a quartic equation in $x = \\sin(\\phi_o)$. For HD 100453 the fit gives an inner inclination of $50.5^\\circ$, position angle $88.3^\\circ$, misalignment of about $70^\\circ$, maximum aspect ratio $h_r = 0.17$, outer inner-edge radius $R_e = 16.8$ au, surface height $H_c = 6.6$ au at $R_c = 25$ au, and $\\alpha \\approx 4$, so the outer surface converges rapidly to $H \\propto R$. The derived inner-disk orientation is consistent with independent interferometric measurements, and the $h_r$ value is roughly five times the gas pressure scale height, implying a dust scattering surface lofted well above the isothermal scale height.","pith_inferences":["Applying the same framework to the other known shadow-casting disks, such as HD 142527, CQ Tau, DoAr 44, and J1604, would test whether $h_r \\sim 0.17$ is a characteristic of wind-lofted surfaces or specific to HD 100453; this extension is not carried out in the paper.","Because the model supplies the scattering angle at every point of the outer surface, it turns the disk into a calibrated phase-function laboratory; combining the geometry with total-intensity images would allow grain properties to be fit as a function of radius, an extension the authors sketch but do not execute.","A vertically asymmetric inner disk, as expected from MHD wind simulations, would make the front and back shadow boundaries differ in width; comparing the widths of the two fitted shadow curves in HD 100453's image is a direct test of the midplane-symmetry assumption that the paper's own limitation discussion invites."],"forward_implications":["For any transition disk with a resolved shadow, the same fits can deliver the inner disk orientation and thickness without interferometric imaging of the inner disk itself.","The method breaks the inclination–position-angle degeneracy that forced earlier shadow models to accept a wide range of inclinations, so the orientation estimates become usable inputs for dynamical and SED modeling.","An inner-disk aspect ratio of 0.17 in HD 100453, well above the gas pressure scale height, indicates that the dust scattering surface is vertically extended; if typical, shadow-width measurements probe turbulent or wind-driven lofting rather than the thermal scale height.","Outer disk surface height profiles inferred this way, with the HD 100453 surface converging to $H \\propto R$, provide a direct comparison point for continuum emission bumps seen at millimeter wavelengths, linking the scattering surface to radial dust trapping.","The horizon curve anchors the central position and inner-edge radius of the outer disk, which tightens the astrometric calibration of scattered-light images."],"supporting_citations":[{"why":"Provides the J-band polarized scattered-light image of HD 100453 whose shadow and horizon features are fitted.","marker":"Benisty et al. 2017"},{"why":"Supplies the earlier analytic shadow model that this framework extends by adding inner-disk thickness and a radially varying outer surface.","marker":"Min et al. 2017"},{"why":"Fixes the outer-disk inclination and position angle used in the fit and gives the VLTI interferometric inner-disk orientation used for consistency comparison.","marker":"Bohn et al. 2022"},{"why":"Provides the ALMA Band 7 continuum intensity profile used to compare the inferred surface height and dust distribution.","marker":"Rosotti et al. 2020"},{"why":"Documents the vertically asymmetric MHD wind structure that the paper identifies as a limitation of its symmetric inner-disk assumption.","marker":"Bai 2017"},{"why":"Supplies the MCMC sampler used for parameter exploration and uncertainty estimation.","marker":"Foreman-Mackey et al. 2013"}],"fun_headline_variants":["Shadow curves reveal 70-degree disk misalignment","Single scattered-light image maps both disk geometries","70-degree tilt measured from disk shadow shapes","Misaligned disk geometry from shadow curves","Shadows expose 70-degree tilt in HD 100453"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The outer disk is a single-valued, axisymmetric, optically thick surface whose height profile is fixed to the empirical family of Eq. (6), and the inner disk is a vertically symmetric slab with a single maximum aspect ratio; if either surface deviates from these shapes, the analytic shadow and horizon curves shift and all fitted parameters inherit the bias.","fun_headline_variants_meta":{"raw":{"variants":["Shadow curves reveal 70-degree disk misalignment","Single scattered-light image maps both disk geometries","70-degree tilt measured from disk shadow shapes","Misaligned disk geometry from shadow curves","Shadows expose 70-degree tilt in HD 100453"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000643,"raw_usage":{"total_tokens":3009,"prompt_tokens":1052,"completion_tokens":1957,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":668,"completion_tokens_details":{"reasoning_tokens":1886}},"tokens_in":668,"tokens_out":1957,"duration_ms":16539,"temperature":1.0,"reasoning_tokens":1886,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T22:49:52.113609+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a transition disk whose outer surface is known from independent data, such as resolved molecular-line emission or radiative-transfer fits, and whose inner disk is vertically asymmetric or warped; if the fitted shadow curves cannot reproduce the observed shadow edges to within the gradient uncertainties, or if the inferred $h_r$ disagrees with the surface height measured at the same radius by an independent tracer, the central assumption of a symmetric, single-valued surface is contradicted. A cheaper check is to measure the two shadow boundary widths in HD 100453 separately, since a vertically symmetric slab predicts nearly equal front and back widths while an MHD wind predicts a measurable difference.","supporting_citations":[],"review_version":1}