{"id":"d29dd954-3e61-4c70-8196-8f50e4725fb3","arxiv_id":"2608.06579","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":18,"one_line_summary":"MagAO-X polarimetry is characterized with a Mueller-matrix model; a dual rotating quarter-wave plate compensator raises average polarimetric efficiency to 87.4% and yields first-light i' images of the HR 4796 disk.","lead":"MagAO-X, a visible-light adaptive optics camera on the 6.5 meter Magellan telescope, has been equipped with a polarimetric mode that measures how much of the light from a dusty disk is polarized. The team calibrated the instrument, added a rotating waveplate compensator to boost polarimetric efficiency by 17.5%, and captured new images of the debris disk around HR 4796.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The headline efficiency/IP numbers exclude the telescope's M3, yet the abstract omits this caveat; the on-sky M3 model is an unvalidated idealized silver assumption.","rationale":"The reader's weakest-assumption analysis correctly identifies the unvalidated idealized M3 model as the most load-bearing concern. This is the one optical element that cannot be calibrated with the polarization generator, it directly enters the on-sky Mueller inversion, and the abstract's headline claim is missing the caveat that the efficiency and IP numbers exclude M3. The concern is real but known and addressable: the paper explicitly states plans to fit M3 with standard stars, and the qualitative DQWP improvement is visible directly in the single-difference curves (Figures 3 and 8) rather than depending only on the model. The on-sky HR 4796 detection is a demonstration, not a precision measurement. Therefore the conditional acceptance remains appropriate; no change to the reader's verdict is warranted.","tokens_in":947,"tokens_out":692,"duration_ms":87636,"concrete_test":"Observe an unpolarized standard star (e.g., HD 14069) through the full telescope plus instrument path including M3, and compare the measured Stokes Q and U against the model prediction that uses the idealized silver M3 matrix. If the residual polarization exceeds about 2% of the total intensity, the M3 assumption is inadequate and the HR 4796 Stokes images and the quoted efficiency/IP values must be re-derived.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central performance claim (87.4% efficiency, 8.3% IP, Table 2) is measured with the polarization generator injecting light after M3, so it describes the instrument alone. On-sky reductions (Section 5.1) instead use an idealized silver Mueller matrix for M3 (Eq. 10 in Section 3.3) because the generator cannot illuminate M3. This M3 model is never validated against any polarized or unpolarized standard star, and the authors list such measurements only as future work. If the real M3 diattenuation or retardance differs, the least-squares Mueller inversion of the HR 4796 data yields biased Stokes Q/U, and the end-to-end on-sky polarimetric efficiency and IP could be substantially worse than the abstract's headline numbers imply. The abstract presents these lab-only values without the 'excluding M3' caveat stated in the figures and Section 3.4, risking overstatement of the achieved on-sky performance.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper describes the visible-light polarimetric mode of MagAO-X, a high-contrast instrument at the Magellan Clay Telescope. The authors built a polarization generator to inject known polarization states after the telescope's tertiary mirror, fit a Mueller-matrix model to the resulting calibration data across r', i', and z' filters, and used the fitted model to compute polarimetric efficiency and instrumental polarization as functions of the k-mirror image rotator angle. They find that the image rotator introduces significant, angle-dependent inefficiency and polarization; to mitigate this, they designed and deployed a dual rotating quarter-wave plate (DQWP) compensator. After installation, they report an average polarimetric efficiency of 87.4% (up from 70.0%) and an average instrumental polarization of 8.3% (down from 13.7%). Finally, they present on-sky i'-band polarimetric images of the HR 4796 debris disk, showing the forward-scattering side of the disk, and conclude that lessons learned will inform future ELT polarimeters.","tokens_in":18815,"tokens_out":3387,"duration_ms":33579,"significance":"If the reported performance holds, this is a useful addition to the comparatively small set of visible-light high-contrast polarimeters on large telescopes. The paper's strengths are concrete: it includes a purpose-built polarization generator, a physically motivated Mueller-matrix model with low fit residuals (MSE of 0.1-0.3% in the calibration data), a novel DQWP compensation scheme that demonstrably flattens the image-rotator-induced efficiency curves, and a first on-sky demonstration that detects the polarized HR 4796 disk at small inner working angle. These elements are reproducible in principle and the work is relevant both to the MagAO-X upgrade path and to polarimetric design for GMT and ELT instruments. The main weakness is that the headline performance numbers are model-derived and exclude the telescope's M3 mirror, while the on-sky reduction uses an unvalidated idealized silver model for M3; the abstract presents the lab-only numbers without this caveat.","major_comments":[{"comment":"The headline values 'average increase in polarimetric efficiency of +17.5% (to 87.4%) and a reduction in instrumental polarization of -5.4% (to 8.3%)' are derived from the Mueller-matrix model fitted to calibration data taken with the polarization generator injecting light after the telescope's M3 mirror. As the captions of Figures 4 and 9 state, these values are 'based on the Mueller-matrix model excluding M3.' The abstract omits this caveat, presenting the numbers as measured instrument performance without qualification. Because M3's diattenuation and retardance are not measured (the paper lists standard-star calibration as future work), the end-to-end on-sky efficiency and IP could differ substantially. The abstract and conclusions should explicitly state that these are instrument-only values, not end-to-end telescope-plus-instrument values.","section":"Abstract; Sec. 3.4; Table 2"},{"comment":"The average improvement across filters masks a lack of improvement in r': post-DQWP r' efficiency is 67.9% (a change of only +0.3%), and r' instrumental polarization remains 22.4%. The statement in the abstract that the DQWP 'increased polarimetric efficiency... and reduced instrumental polarization... across all filters' is technically true only because i' and z' improve dramatically, while r' is essentially unchanged. Since the r' band is one of the three science filters and is the filter most affected by PBS leakage, the paper should prominently report filter-by-filter values in the abstract or at least in the conclusions, rather than relying on the average.","section":"Sec. 4.2; Table 2"},{"comment":"The on-sky calibration assumes an idealized silver-mirror Mueller matrix for M3 because the polarization generator cannot inject light through the telescope. This assumed M3 model is never validated against polarized or unpolarized standard stars; the authors explicitly state that such measurements are future work. Since the least-squares Mueller inversion of the HR 4796 data (Eqs. 17-18) uses this unvalidated M3 term, the calibrated Stokes Q and U images inherit unknown systematic errors from any discrepancy between the real M3 and the idealized silver model, including in the reported on-sky efficiency and IP values. This is a load-bearing uncertainty for the paper's central demonstration, and it should be discussed quantitatively (e.g., a sensitivity analysis of plausible M3 diattenuation/retardance values on the final Q_phi image).","section":"Sec. 5.1; Eq. (10)"},{"comment":"The efficiency and IP values in Table 2 are quoted to one decimal place without any uncertainties. The Mueller-model fit has small MSE (0.1-0.3%), but the derived efficiency/IP values are nonlinear functions of the fitted parameters, and their uncertainties are not estimated (e.g., via covariance propagation or bootstrap). Without error bars, it is impossible to assess whether the reported improvements are statistically significant, particularly for i' and z' where the post-DQWP values are close to 98% and 96%. Adding uncertainties to Table 2 and to the abstract's headline numbers would materially strengthen the claims.","section":"Sec. 3.4; Table 2"}],"minor_comments":[{"comment":"The phrase 'two radial-basis-function (RBF) Gaussian processes' is imprecise; Gaussian process regression with an RBF kernel is the standard terminology, and the text should be clarified to distinguish the kernel from the process.","section":"Sec. 4.1"},{"comment":"The 'optimized offset angle was 1.9' should specify the units (degrees) and define the sign convention for the offset relative to the azimuthal angle theta in Eq. (17).","section":"Sec. 5.2"},{"comment":"The symbol X is used both as the vertical-minus-horizontal flux difference and later as Stokes Q or U in the derivation; this overloaded notation is confusing. Suggest using a different symbol for the raw difference, such as D, and reserving X for the Stokes parameter in Eq. (16).","section":"Sec. 2.1; Eq. (7)"},{"comment":"The matrix product in Eq. (11) lists components from PBS back to telescope, but the text says 'matrix multiplication from right to left' and then gives M = M_PBS * M_Inst * ... * M_Tel; the ordering is correct, but a sentence explicitly noting that this corresponds to light propagating from telescope to detector would help the reader avoid confusion.","section":"Sec. 3.3, after Eq. (11)"},{"comment":"The sentence 'we converted the data to units of e-/s using the camera conversion gain, EM gain, and integration time' would benefit from stating the numerical values of the conversion gain and EM gain for each camera, as these affect the absolute calibration of the Stokes images.","section":"Sec. 5.1"}],"recommendation":"major_revision","confidential_remarks":"The paper is within scope for an instrumentation journal, and the authors have done a substantial amount of careful calibration work. The central issue is a mismatch between the abstract's unqualified performance claims and the actual, M3-excluding, model-derived nature of those numbers. This can be fixed by revising the abstract and conclusions, adding uncertainties to Table 2, and adding a sensitivity discussion of the M3 assumption in the on-sky analysis. If the authors address these points, the paper would be suitable for publication; the current version, however, overstates the achieved on-sky polarimetric performance."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a credible instrument paper that delivers what it promises. The Mueller-matrix fits to the calibration data are good (MSE ~0.1-0.3%), the DQWP compensator clearly improves the polarimetric response, and the HR 4796 i'-band image is a real on-sky demonstration of the mode. It is the first detailed characterization of MagAO-X's polarimetric mode and the first visible-light high-contrast PDI images from this instrument. That is a legitimate contribution, with the DQWP being a useful engineering example for other high-contrast polarimeters.\n\nThe soft spots are real but not fatal. The headline efficiency and IP numbers (87.4% and 8.3%) are measured with the polarization generator that injects light after the telescope M3, so they describe the instrument alone. The on-sky reduction uses an ideal-silver Mueller matrix for M3, which is stated in Section 5.1 but never validated against a polarization standard. If M3's actual diattenuation/retardance differs, the Stokes images would be biased. The abstract does not mention this caveat, which is a minor overstatement. Also, the quoted efficiencies and IP values have no error bars; given that the model is fitted to the same data, they are model-derived quantities. The r' channel remains poor (67.9% efficiency) due to PBS leakage, and the authors acknowledge a planned new PBS. No data or code are released, which limits independent checking but is common for instrument papers.\n\nFor a serious referee, I would send it out. The central claim—that MagAO-X has a working PDI mode with a demonstrable on-sky disk image—holds up. The missing pieces are addressable: add error bars, state the M3 caveat in the abstract, and perhaps include a polarized standard-star check as future work. The citation list looks appropriate, including VAMPIRES and SPHERE work. This is a paper for the high-contrast polarimetry community; it is not a paradigm shift but a useful capability step.","headline":"A solid instrument paper with a real on-sky result; the headline numbers are instrument-only and the M3 model is an assumption, but the limitations are mostly stated in the text.","tokens_in":19466,"tokens_out":2408,"would_cite":true,"duration_ms":22822,"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":"MagAO-X's visible-light polarimeter is characterized with a polarized-light generator and a fitted Mueller-matrix model, and a dual rotating quarter-wave plate compensator raises average polarimetric efficiency from 70.0% to 87.4%, cuts…","keywords":["polarimetric differential imaging","instrumental polarization","Mueller matrix","dual rotating quarter-wave plate","high-contrast imaging","debris disk","extreme adaptive optics","MagAO-X"],"falsifier":"Observe a grid of polarized and unpolarized standard stars at several parallactic angles and telescope altitudes, and fit the M3 Mueller matrix from the on-sky data alone; if the best-fit M3 diattenuation and retardance depart from the idealized silver-mirror values by more than the model residuals, the internal-calibration numbers (87.4% efficiency, 8.3% instrumental polarization) and the HR 4796 Stokes images are systematically biased.","tokens_in":18360,"feed_emoji":"🔭","tokens_out":8157,"duration_ms":67026,"temperature":0.7,"pith_summary":"MagAO-X, the visible-light extreme-adaptive-optics instrument on the 6.5 m Magellan Clay Telescope, has been upgraded into a polarimetric differential imaging (PDI) camera in the r', i', and z' bands. This paper is trying to establish that the instrument's polarimetric response can be captured by a Mueller-matrix model calibrated with a purpose-built polarization generator, and that the dominant defect—a dynamic loss of efficiency tied to the k-mirror image rotator—can be actively corrected with a dual rotating quarter-wave plate (DQWP) compensator. It reports that the compensator raises average polarimetric efficiency from 70.0% to 87.4% and lowers average instrumental polarization from 13.7% to 8.3%, and demonstrates the end-to-end pipeline on sky with i'-band imaging of the HR 4796 debris disk. A sympathetic reader would care because PDI is one of the cleanest ways to suppress unpolarized starlight and see faint dust-scattered light around other stars, and the same calibration-plus-compensation architecture is the path for polarimeters on next-generation extremely large telescopes.","feed_headline":"Dual rotating wave plates lift MagAO-X polarimetry to 87.4%","feed_subtitle":"Compensator fixes k-mirror crosstalk and delivers a close-up polarized view of the HR 4796 debris ring.","key_machinery":"The object that carries the argument is the dual rotating quarter-wave plate (DQWP) compensator: two zero-order quarter-wave plates (design wavelength 780 nm) on stepper-controlled rotation mounts, placed in the collimated beam after the tweeter deformable mirror, with a tracking law that reorients the input polarization so it reaches the polarizing beamsplitter in the instrument's eigenpolarization, cancelling the elliptical retardance and diattenuation of the k-mirror and periscope. The measurement side of the argument is a Mueller-matrix model of the whole optical chain, fit by Nelder-Mead minimization of mean-squared error against normalized single-difference fluxes taken with and without the injection polarizer; the fitted model yields polarimetric efficiency and instrumental polarization as functions of image-rotator angle and filter, and later supplies the least-squares Mueller-matrix inversion used for the HR 4796 Stokes images. The one component that cannot be measured with the generator, the telescope tertiary mirror M3, is inserted in the on-sky model as an idealized silver mirror.","core_discovery":"The central claim is that MagAO-X can be made a calibrated, high-contrast visible-light polarimeter, and that its main weakness—polarimetric efficiency and instrumental polarization that swing strongly with the k-mirror image rotator angle—is a hardware problem with a hardware solution. Using a polarization generator that injects effectively 100% linearly polarized light, the authors measure normalized single-difference fluxes for every HWP and image-rotator angle in each filter, then fit a Mueller-matrix model that chains the telescope tertiary mirror (M3), the half-wave plate, the M4 fold mirror, the image rotator, a generic periscope-dominated instrument term, and the polarizing beamsplitter, with separate diattenuations for the two cameras. The fit quantifies efficiency as the average recovered linear polarization for Stokes Q and U and instrumental polarization as the polarized fraction generated from unpolarized input. The newly installed DQWP dynamically reorients incoming polarization to the instrument's eigenpolarization, raising average efficiency from 70.0% to 87.4% (i' from 82.0% to 97.9%, z' from 59.9% to 96.3%) and cutting average instrumental polarization from 13.7% to 8.3%; r' stays limited by beamsplitter leakage. On sky, least-squares Mueller-matrix inversion of double-differenced i' images of HR 4796 yields a Stokes $Q_\\phi$ image of the bright forward-scattering near side of the disk at one of the closest inner working angles yet.","pith_inferences":["Editorial inference: the DQWP is effectively an analog pre-compensator that diagonalizes the instrument's Mueller matrix in real time; the same tracking-law idea could be retrofit to any Nasmyth high-contrast polarimeter whose k-mirror angle changes during an observation.","Editorial inference: because M3 is excluded from the internal calibration, the 87.4% efficiency and 8.3% instrumental polarization are upper limits on true on-sky performance; standard-star measurements will likely revise them by a few percent in one direction or the other.","Editorial inference: the residual speckle noise and dispersion smearing seen around HR 4796 suggest the polarimetric contrast floor is currently set by the atmosphere-dispersion control law, not by the polarimeter; closing the ADC loop could improve inner working angle without any new polarimetric optics.","Editorial inference: the observed asymmetry in the r' response is a clean signature of PBS leakage rather than mirror retardance, and the same Mueller-model fit could be used to predict the polarimeter's performance after the PBS swap before re-running the full calibration."],"forward_implications":["In i' and z', where the DQWP tracking law pushes efficiency to 97.9% and 96.3%, MagAO-X can now run PDI close to its photon-noise limit rather than being crippled by image-rotator crosstalk.","The r' band remains the weak link: polarizing-beamsplitter leakage below roughly 650 nm caps its efficiency, and the 65/35 wavefront-sensor beamsplitter adds about 22% instrumental polarization, so the planned November 2026 PBS replacement should be the decisive fix.","The HR 4796 result shows the full calibration chain—double-differencing, Mueller-matrix inversion, ad-hoc IP removal, and $Q_\\phi$ optimization—works on a real debris disk and resolves the forward-scattering near side at small inner working angle.","Because rotating the DQWP shifts the pupil image on the wavefront sensor by about 0.5 pixels over 180 degrees of plate rotation, a per-QWP lookup table for the pupil-alignment loop is required before the compensator can be used for deep, long-exposure high-contrast observations.","The DQWP-plus-Mueller-model architecture is explicitly the template for GMT and ELT polarimeters, whose Nasmyth optics will introduce even larger and more dynamic polarization effects."],"supporting_citations":[{"why":"Describes the concept, implementation, and on-sky commissioning of the MagAO-X polarimetric module that this paper characterizes and upgrades.","marker":"[23]"},{"why":"Establishes that Nasmyth reflections create instrumental polarization and state mixing, motivating the calibration approach.","marker":"[17]"},{"why":"Supplies the Mueller-matrix fitting and least-squares inversion methods, plus the first-light HR 4796 ring that the on-sky result is compared against.","marker":"[18]"},{"why":"Defines the polarimetric-efficiency and instrumental-polarization estimators and the IP-correction framework adopted here.","marker":"[19]"},{"why":"Demonstrates the same instrument-polarization characterization approach on the SCExAO/VAMPIRES polarimeter.","marker":"[21]"},{"why":"Introduces the dual-rotating-waveplate compensation technique that the DQWP is built on.","marker":"[25]"},{"why":"Reports the VAMPIRES visible-light high-contrast polarimetry that this instrument and its DQWP calibration inherit.","marker":"[27]"},{"why":"Provides the optical polarized phase function of the HR 4796 dust ring used to validate the on-sky polarized image.","marker":"[30]"}],"fun_headline_variants":["MagAO-X polarimetry efficiency jumps to 87.4% with DQWP","DQWP boosts MagAO-X polarimetric efficiency to 87.4%","MagAO-X polarimeter reaches 87.4% efficiency with DQWP","MagAO-X reveals HR 4796 ring at close inner working angle","Fixing k-mirror crosstalk lifts MagAO-X polarimetry to 87.4%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The on-sky calibration rests on the assumption that the telescope's tertiary mirror M3 behaves exactly like an idealized silver mirror, because the polarization generator injects light only after M3 and cannot measure it; if the real M3 Mueller matrix differs from the silver model, the reported efficiency, instrumental polarization, and HR 4796 Stokes images inherit a systematic error.","fun_headline_variants_meta":{"raw":{"variants":["MagAO-X polarimetry efficiency jumps to 87.4% with DQWP","DQWP boosts MagAO-X polarimetric efficiency to 87.4%","MagAO-X polarimeter reaches 87.4% efficiency with DQWP","MagAO-X reveals HR 4796 ring at close inner working angle","Fixing k-mirror crosstalk lifts MagAO-X polarimetry to 87.4%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001788,"raw_usage":{"total_tokens":7170,"prompt_tokens":1191,"completion_tokens":5979,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":807,"completion_tokens_details":{"reasoning_tokens":5870}},"tokens_in":807,"tokens_out":5979,"duration_ms":43719,"temperature":1.0,"reasoning_tokens":5870,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T04:16:20.013111+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe a grid of polarized and unpolarized standard stars at several parallactic angles and telescope altitudes, and fit the M3 Mueller matrix from the on-sky data alone; if the best-fit M3 diattenuation and retardance depart from the idealized silver-mirror values by more than the model residuals, the internal-calibration numbers (87.4% efficiency, 8.3% instrumental polarization) and the HR 4796 Stokes images are systematically biased.","supporting_citations":[{"cited_title":"Concept, implementation, and on-sky commissioning of the AO polarimetric module on MagAO-X,","cited_arxiv_id":null,"evidence_quote":"Describes the concept, implementation, and on-sky commissioning of the MagAO-X polarimetric module that this paper characterizes and upgrades."},{"cited_title":"Accurate optical polarimetry on the nasmyth platform,","cited_arxiv_id":null,"evidence_quote":"Establishes that Nasmyth reflections create instrumental polarization and state mixing, motivating the calibration approach."},{"cited_title":"Polarimetry with the gemini planet imager: methods, performance at first light, and the circumstellar ring around hr 4796a,","cited_arxiv_id":null,"evidence_quote":"Supplies the Mueller-matrix fitting and least-squares inversion methods, plus the first-light HR 4796 ring that the on-sky result is compared against."},{"cited_title":"Polarimetric imaging mode of VLT/SPHERE/IRDIS - II. Characterization and correction of instrumental polarization effects,,","cited_arxiv_id":null,"evidence_quote":"Defines the polarimetric-efficiency and instrumental-polarization estimators and the IP-correction framework adopted here."},{"cited_title":"Characterizing the instrument polarization of SCExAO V AMPIRES,","cited_arxiv_id":null,"evidence_quote":"Demonstrates the same instrument-polarization characterization approach on the SCExAO/VAMPIRES polarimeter."},{"cited_title":"The V AM- PIRES instrument: imaging the innermost regions of protoplanetary discs with polarimetric interferometry,","cited_arxiv_id":null,"evidence_quote":"Introduces the dual-rotating-waveplate compensation technique that the DQWP is built on."},{"cited_title":"Visible-light high-contrast imaging and polarimetry with SCExAO/V AMPIRES,","cited_arxiv_id":null,"evidence_quote":"Reports the VAMPIRES visible-light high-contrast polarimetry that this instrument and its DQWP calibration inherit."},{"cited_title":"Optical polarised phase function of the HR 4796A dust ring,","cited_arxiv_id":null,"evidence_quote":"Provides the optical polarized phase function of the HR 4796 dust ring used to validate the on-sky polarized image."}],"review_version":1}