{"id":"a426c3c0-29a4-40a2-a4ea-42a20f98eac3","arxiv_id":"2509.10451","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"A Jones-matrix calibration model for CHARA's MIRC-X and MYSTIC instruments reproduces instrumental diattenuation and retardance and claims residual accuracies of 3.4% (H band) and 5.9% (K band) in visibility ratio and 1.4 to 2.4 degrees in differential phase.","lead":"This paper builds a Jones-matrix model of the polarization distortions in the CHARA optical interferometer and calibrates it using observations of the unpolarized star Upsilon Andromedae with the MIRC-X and MYSTIC instruments. It reports residual calibration accuracy of about 3 to 6 percent in visibility ratio and 1.4 to 2.4 degrees in differential phase, and argues CHARA can now measure intrinsic polarization of resolved dusty stars and disks.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The quoted calibration accuracies are in-sample residuals from a model whose transport-module simplification is admitted to leave systematic differential-phase residuals; the accuracy claim needs an independent test.","rationale":"The paper's Jones-matrix framework is a sensible and publishable step, and the W2 fixed-mirror and LiNbO3 misalignment attributions are concrete and independently plausible. The public mircxpol package is real supporting evidence for reproducibility. However, the headline accuracy numbers are in-sample fit residuals, and the model itself is acknowledged to omit polarization effects in the transport module. Because those effects, if present, enter through rotations that depend on the source position on the sky, the υ And residual RMS cannot by itself establish the accuracy for future targets. Re-fitting with a full Jones transport matrix is the direct, low-cost test that can show whether the simplification actually matters; if it does not change residuals or parameters, the concern is retired. This does not change the reader's CONDITIONAL verdict: the paper should be published after the calibration accuracy is demonstrated on an independent source or with a proper calibrator cycle.","tokens_in":33628,"tokens_out":10437,"duration_ms":97020,"concrete_test":"Refit the October 2022 υ And dataset with Eq. 27 modified so that the Transport Module rotation R(γ) is replaced by a full per-telescope 2×2 Jones matrix fMtrans (using the released mircxpol package), then compare the differential-phase residual RMS and the inferred fMCoudé/fMLab parameters with Figures 5–8. If the differential-phase RMS decreases materially (e.g., by more than about 0.5°) or any fitted fM parameter shifts by more than its quoted uncertainty, the quoted accuracies are model-dependent and the transport-module simplification is load-bearing; a null result would retire this specific concern.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim in the Abstract (±3.4% visibility-ratio and ±1.4° differential-phase accuracy for MIRC-X; ±5.9% and ±2.4° for MYSTIC) is derived from the RMS scatter of residuals after fitting Eq. 27 to a single unpolarized calibrator, υ And, on three nights, with tuned systematic-error terms (§6.1), night-to-night offset subtraction (§6.2), and some data excluded (§4.1). In §3.2 the Transport Module M8–M10 is modeled as a pure rotation R(γ=39.85°), with no diattenuation or retardance. Section 6.4 explicitly concedes that 'residual systematic deviations in the differential phase data suggest that unmodeled transport module effects may be present' and that M7–M10 diattenuation/retardance would produce off-diagonal Jones elements absent from Eq. 27. If such effects exist, the fitted fMAT, fMCoudé, and fMLab values become effective parameters that absorb orientation-dependent transport effects; the residual RMS measured on υ And then does not bound the calibration error for a science target observed at different hour angles or parallactic angles. The capability claim therefore rests on an admitted model simplification and has not been validated on an independent source.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents a Jones-matrix model of the polarization transfer function of the CHARA Array, spanning the telescope optics, Coudé path, transport mirrors, delay lines, and the MIRC-X and MYSTIC beam combiners. The model is fit to three nights of observations of the unpolarized standard star υ Andromedae, yielding per-telescope diattenuation and retardance parameters for grouped mirror sets (fMAT, fMCoudé, fMLab), a detection of a large Coudé-path phase shift for telescope W2 (attributed to a fixed aluminum M4 mirror), and evidence for misaligned LiNbO3 compensator plates on S1 and W2. The paper reports post-fit calibration accuracies of ±3.4% in visibility ratio and ±1.4° in differential phase for MIRC-X, and ±5.9% and ±2.4° for MYSTIC, and concludes that CHARA can now deliver high-accuracy spectropolarimetric measurements of resolved sources such as AGB stars and YSOs.","tokens_in":33995,"tokens_out":3876,"duration_ms":35595,"significance":"If the calibration accuracy claim holds, this work would be an important step toward routine long-baseline near-infrared spectropolarimetry, enabling resolved studies of dust scattering in AGB envelopes and YSO inner disks. The paper has clear strengths: the Jones-formalism framework is standard and clearly presented; the model successfully identifies specific hardware anomalies (the W2 fixed mirror and LiNbO3 plate misalignments) that are plausible and partly corroborated by independent hardware knowledge; and the authors release the fitting code as the mircxpol package, which supports reproducibility. The main weakness is that the headline accuracy numbers are in-sample residual scatters from a fit to a single unpolarized target, with systematic error terms tuned to force reduced chi-square near unity, and with an admitted simplification of the transport module. The central capability claim therefore rests on internal consistency rather than independent validation.","major_comments":[{"comment":"The quoted calibration accuracies (±3.4% visibility ratio, ±1.4° differential phase for MIRC-X; ±5.9%, ±2.4° for MYSTIC) are the RMS residuals from fitting Eq. (27) to the same υ And dataset, after tuning the systematic error terms σ_flux,sys, σ_vis,sys, and σ_phase,sys so that each observable's reduced chi-square approaches unity. This makes the residuals in-sample scatter, not an independent accuracy estimate: the tuned error terms can absorb model deficiencies, so the near-Gaussian histograms in Fig. 8 partly reflect the fitting construction. Please provide an out-of-sample test, for example fitting two nights and testing the third, or observing an independent unpolarized calibrator and reporting its residuals without re-tuning the systematic terms, before the abstract-level accuracy claim is made.","section":"§6.1, Fig. 8, Abstract"},{"comment":"The Transport module (mirrors M8–M10) is modeled as a pure rotation R(γ=39.85°) with no diattenuation or retardance. Section 6.4 explicitly concedes that 'residual systematic deviations in the differential phase data suggest that unmodeled transport module effects may be present' and that M7–M10 diattenuation/retardance would create off-diagonal Jones elements absent from Eq. (27). If the transport module does introduce such effects, the fitted fMAT, fMCoudé, and fMLab parameters become effective parameters that absorb orientation-dependent transport effects, and the residual RMS measured on υ And at specific hour angles may not bound the calibration error for a science target observed at different parallactic angles. Please quantify the sensitivity to this assumption, for example by re-fitting with a full fMtrans Jones matrix and comparing the calibrated residuals, or by clearly qualifying the accuracy claim as conditional on the transport-module simplification.","section":"§3.2, Eq. (27), §6.4"},{"comment":"The manuscript excludes data with differential-phase uncertainties exceeding 10° and, for October 21, observations at HA ≤ −4.9 hr are excluded from the fit due to 'anomalous behavior' while being retained in the figures. Because the model is fit to this single unpolarized target, these post-hoc exclusions can bias the derived instrumental parameters and the residual statistics used for the calibration-accuracy claim. Please report the effect of including the excluded data (or justify the cut with an objective, reproducible criterion) and state how the quoted accuracies change when the exclusions are lifted.","section":"§4.1"},{"comment":"Night-to-night correlated errors are removed by subtracting each night's mean parameter offset across wavelengths, which reduces the scatter in diattenuation from ±3.70% to ±0.52% and in phase from ±1.47° to ±0.36°. The manuscript does not state clearly whether the Figure 8 residual histograms and the abstract calibration accuracies include this nightly-offset correction, nor how a future science observation would receive the same correction without a suitable same-night calibrator. Please clarify this and, if the quoted numbers are post-correction, also report the raw per-night accuracies before the empirical offset subtraction.","section":"§6.2 and §6.1, Fig. 8"}],"minor_comments":[{"comment":"The sentence 'the differential intrinsic polarization of spatially resolved sources, such as AGB stars and YSOs, typically greater than these instrumental uncertainties' is missing a verb; it should read '... is typically greater than ...'.","section":"Abstract"},{"comment":"The notation ⟨E_m + E_n⟩² is ambiguous; the expression should be written as ⟨|E_m + E_n|²⟩ to make the ensemble average and the modulus explicit.","section":"Eq. (6)"},{"comment":"The scalar factors f and e^{iφ} in Eq. (27) are introduced but not explicitly defined in the adjacent text; please define them as the net transmission and common phase of the beam path, respectively.","section":"§5, Eq. (27)"},{"comment":"The citation 'Gardner et al. 2025' appears in the text without a full bibliographic entry; please add the complete reference.","section":"Reference list"},{"comment":"The y-axis labels in the MYSTIC differential-phase figures are rendered as 'H V(deg)' rather than 'Δψ(H−V) (deg)'; please correct the axis labels for consistency with Figures 4, 20, and 21.","section":"Figures 22–24"}],"recommendation":"major_revision","confidential_remarks":"The paper's central claim is phrased more strongly than the evidence supports: the accuracy numbers are in-sample residuals after fitted systematic-error inflation and explicit nightly-offset subtraction, and the transport-module simplification is admitted in §6.4. I do not see this as a reason for rejection, because the formalism and the hardware-anomaly identifications are valuable and the path to a valid accuracy claim is clear: an independent calibration test (or at minimum a leave-one-night-out analysis and a sensitivity study of the transport-module assumption) would materially raise the paper's credibility. I would encourage the editor to request that validation rather than settle for softened wording alone."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this is a genuinely useful calibration paper for CHARA polarimetry, but the quoted accuracy numbers are in-sample residuals, so the capability claim is plausible rather than proven.\n\nWhat's new: it's the first full Jones-matrix model of the MIRC-X/MYSTIC beam trains, fit to an unpolarized calibrator, and it produces concrete hardware diagnoses—the W2 fixed-mirror phase shift and the misaligned LiNbO3 plates on S1 and W2. The formalism is standard but carefully applied, and the paper is unusually honest about what it can't explain. The diattenuation values are physically consistent with aluminum-coated reflections, which gives me some confidence the model is capturing real effects rather than just absorbing noise. The code release (mircxpol) is a plus.\n\nThe soft spots. The headline ±3.4% and ±1.4° are RMS scatter of residuals from the same dataset the model was fit to. That's a measure of self-consistency, not external accuracy. The systematic-error terms are tuned to force reduced chi-square to unity, which makes the Gaussian residual histograms in Figure 8 partly circular. There is also a post-hoc exclusion of early-hour-angle data on Oct 21, and only one calibrator over three nights. The bigger issue is the transport module: Sections 3.2 and 6.4 treat M8–M10 as a pure rotation, then concede that unmodeled diattenuation/retardance there may be responsible for residual differential-phase systematics. If that's true, the fitted fM parameters and the quoted accuracies are effective values that may not transfer to a science target observed at different hour angles. The authors acknowledge this and propose a full Jones matrix in future work, but the abstract still overstates what the current data demonstrate.\n\nIs the central claim wrong? Not necessarily. The fits are good, the hardware anomalies are independently plausible, and the transfer-function approach is standard. But the paper would be much stronger if the calibration accuracy were demonstrated on a second unpolarized source, or if the claims were softened to 'repeatability on the calibrator.' I'd recommend sending it to a competent referee rather than desk rejecting, with a request for one independent check or a clear rewrite of the accuracy claim.\n\nThis is a paper for the interferometry/polarimetry instrumentation community. I'd bring it to reading group and would cite it for the CHARA polarization model, while caveating the accuracy numbers.\n\nRecommendation: publish after revision, conditional on an independent validation or a moderated claim.","headline":"A solid CHARA polarization calibration with real hardware findings, but the headline accuracy numbers are in-sample residuals and the transport-module simplification may limit transferability.","tokens_in":34529,"tokens_out":2793,"would_cite":true,"duration_ms":26282,"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":"The CHARA Array can now calibrate its instrumental polarization well enough to measure intrinsic polarization in resolved AGB stars and young stellar objects.","keywords":["optical interferometry","spectropolarimetry","instrumental polarization calibration","Jones matrix","CHARA Array","MIRC-X","MYSTIC","circumstellar dust"],"falsifier":"Observe a second zero-polarization standard star at various hour angles and fit the same model; if the fitted lab-module parameters differ from the Upsilon Andromedae fit by more than the quoted ±0.52 percent / ±0.36 degrees, or if replacing the transport rotation $R(\\gamma)$ with a full Jones matrix shifts the W2 Coudé phase by more than its ~2 degree uncertainty, the transport-module assumption — and the derived calibration accuracy — fails. A direct laboratory measurement of the diattenuation and retardance of mirrors M7–M10 at 45 degrees incidence would settle it independently.","tokens_in":2001,"feed_emoji":"🔭","tokens_out":2798,"duration_ms":75289,"temperature":0.7,"pith_summary":"This paper aims to prove that the CHARA optical interferometer can now separate the polarizing effects of its own mirrors and instruments from the polarization of the light arriving from a star. The authors build a Jones-matrix model of the full beam train, fit it to observations of an unpolarized calibrator star, and report that after correcting for night-to-night offsets the instrumental polarization can be calibrated to about ±3.4 percent in visibility ratio and ±1.4 degrees in differential phase in the H band, and ±5.9 percent and ±2.4 degrees in the K band. These uncertainties sit below the differential polarization expected from resolved AGB dust shells and young-stellar-object disks, so the claim is that CHARA is now a viable spectropolarimetric instrument. The argument succeeds only if the model's simplifying assumption about one mirror group is valid; the paper itself flags where that assumption may fail.","feed_headline":"Polarization calibration at CHARA reaches 3 percent accuracy","feed_subtitle":"A Jones-matrix model of the beamtrain makes resolved spectropolarimetry of dust shells and disks feasible.","key_machinery":"The central object is the product of 2×2 Jones matrices in Equation 27, which maps the sky-frame electric field to the detector's horizontal/vertical basis through a sequence of rotations and mirror matrices. Rotations account for the parallactic angle $q$, altitude $a$, azimuth $A$, the half-wave plate angle, and the fixed 39.85-degree transport rotation $\\gamma$; the mirror matrices $fM_{\\mathrm{AT}}$, $fM_{\\mathrm{Coudé}}$, and $fM_{\\mathrm{Lab}}$ each carry a fitted diattenuation $A^2$ and phase retardance $\\psi$. The model's observables are flux ratios $f_H/f_V$ per telescope, visibility ratios $V_H/V_V$ per baseline, and differential phases $\\Delta\\psi_{H-V}$, all of which are insensitive to overall intensity fluctuations and can be fit globally across baselines and nights. This cascade is what converts raw fringe measurements into per-element instrumental polarization parameters, and inverting it would recover the sky coherency matrix of a science target.","core_discovery":"Using Jones matrices, the paper models every reflection group in the CHARA path — telescope, Coudé path, transport, and lab — as products of rotations and two-parameter mirror matrices carrying a diattenuation $A^2$ and a retardance $\\psi$. Fitting this model to three nights of calibrated observations of the unpolarized star Upsilon Andromedae yields per-telescope parameters for both MIRC-X (H band) and MYSTIC (K band). The fit recovers the expected roughly 4–6 percent diattenuation from multiple 45-degree aluminum reflections and identifies specific hardware anomalies: telescope W2's fixed aluminum mirror (instead of a deformable mirror) adds roughly 22–26 degrees of Coudé-path phase, and two lithium-niobate compensator plates are misaligned, producing chromatic phase slopes. After applying nightly offset corrections, the residual RMS calibration accuracy is ±3.4 percent in visibility ratio and ±1.42 degrees in differential phase for MIRC-X, and ±5.9 percent and ±2.36 degrees for MYSTIC, which the authors argue is enough to detect intrinsic polarization from spatially resolved dust structures around AGB stars and YSOs.","pith_inferences":["If the transport module is later shown to have real diattenuation or retardance, the quoted calibration accuracies are optimistic, but the fix is straightforward: replace the rotation $R(\\gamma)$ with a full Jones matrix and re-fit the same calibrator data.","The identified lithium-niobate plate misalignments suggest a hardware remedy — retune the plates to the primary visibility maximum — that could flatten the chromatic phase slopes without additional modeling.","The roughly 3 percent flux-ratio accuracy will keep net-polarization measurements of faint sources out of reach; the competitive science channel is polarized differential visibility, not total polarization.","The same Jones-matrix calibration framework, with adjusted rotation angles and mirror groupings, could be adapted to other long-baseline arrays that lack field rotators."],"forward_implications":["CHARA data can now be calibrated for H- and K-band differential polarization without assuming the instrument is polarization-free.","Resolved observations of AGB stars and YSO inner disks can recover local intrinsic polarization in the 10–30 percent range, well above the reported 3–6 percent visibility-ratio uncertainty.","The W2 fixed-mirror anomaly identified in the 2022 data should disappear after the May 2024 deformable-mirror upgrade, improving polarization symmetry across the array.","Adopting a unified calibration matrix and routine calibrator–science–calibrator observing should reduce the remaining correlated night-to-night errors.","Calibrated differential visibilities open the door to polarized aperture-synthesis imaging of circumstellar dust, for example with existing reconstruction tools."],"supporting_citations":[{"why":"Supplies the MIRC-X polarimetric mode (half-wave plates, LiNbO3 compensators, Wollaston prism) that the model describes.","marker":"Setterholm et al. 2020"},{"why":"Provides the VLTI Jones/Mueller polarization-transfer calibration approach this model is patterned on, along with the accuracy benchmark.","marker":"GRAVITY Collaboration et al. 2024"},{"why":"Documents the CHARA light-path layout and the fixed 39.85-degree field rotation used by the transport-module model.","marker":"ten Brummelaar 1997"},{"why":"Supplies the 'golden rule of imaging interferometers' argument for homologous reflections that justifies treating the transport module as a pure rotation.","marker":"Traub 1986"},{"why":"Gives the Z-cut LiNbO3 plate birefringence model used to interpret the chromatic phase slopes in the lab module.","marker":"Lazareff et al. 2012"},{"why":"Establishes Upsilon Andromedae as a zero-polarization standard star, the assumption that lets the fit attribute observed polarization to the instrument.","marker":"Tinbergen 1979"},{"why":"Quantifies Upsilon Andromedae's negligible polarization, supporting the unpolarized-calibrator assumption.","marker":"Piirola 1977"},{"why":"Documents the CHARA telescope optics and M4 deformable mirror, the component whose absence in W2 explains the anomalous Coudé phase.","marker":"Che et al. 2013"}],"fun_headline_variants":["CHARA polarimetry error down to 3.4%","Jones matrix model tames CHARA beam polarization","W2 mirror anomaly decoded in CHARA polarization model","CHARA calibration enables resolved spectropolarimetry"],"cache_read_input_tokens":36480,"weakest_assumption_plain":"The model assumes the mirrors that route the beam from each telescope into the lab (M7 through M10) only rotate the polarization direction and do not dim one polarization component or delay it relative to the other; if they do either, the fitted telescope parameters and quoted accuracies are biased.","fun_headline_variants_meta":{"raw":{"variants":["CHARA polarimetry error down to 3.4%","Jones matrix model tames CHARA beam polarization","W2 mirror anomaly decoded in CHARA polarization model","CHARA calibration enables resolved spectropolarimetry"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000925,"raw_usage":{"total_tokens":4072,"prompt_tokens":1161,"completion_tokens":2911,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":777,"completion_tokens_details":{"reasoning_tokens":2848}},"tokens_in":777,"tokens_out":2911,"duration_ms":21041,"temperature":1.0,"reasoning_tokens":2848,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T15:54:23.260788+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Observe a second zero-polarization standard star at various hour angles and fit the same model; if the fitted lab-module parameters differ from the Upsilon Andromedae fit by more than the quoted ±0.52 percent / ±0.36 degrees, or if replacing the transport rotation $R(\\gamma)$ with a full Jones matrix shifts the W2 Coudé phase by more than its ~2 degree uncertainty, the transport-module assumption — and the derived calibration accuracy — fails. A direct laboratory measurement of the diattenuation and retardance of mirrors M7–M10 at 45 degrees incidence would settle it independently.","supporting_citations":[{"cited_title":"R., Monnier , J","cited_arxiv_id":null,"evidence_quote":"Supplies the MIRC-X polarimetric mode (half-wave plates, LiNbO3 compensators, Wollaston prism) that the model describes."},{"cited_title":"2024, , 681, A115, 10.1051/0004-6361/202347238","cited_arxiv_id":null,"evidence_quote":"Provides the VLTI Jones/Mueller polarization-transfer calibration approach this model is patterned on, along with the accuracy benchmark."},{"cited_title":"1997, The 3D Layout of the CHARA Array, Technical Report 48, Center for High Angular Resolution Astronomy, Mt Wilson, CA","cited_arxiv_id":null,"evidence_quote":"Documents the CHARA light-path layout and the fixed 39.85-degree field rotation used by the transport-module model."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the 'golden rule of imaging interferometers' argument for homologous reflections that justifies treating the transport module as a pure rotation."},{"cited_title":"1979, Astronomy and Astrophysics, Suppl","cited_arxiv_id":null,"evidence_quote":"Establishes Upsilon Andromedae as a zero-polarization standard star, the assumption that lets the fit attribute observed polarization to the instrument."},{"cited_title":"D., et al","cited_arxiv_id":null,"evidence_quote":"Documents the CHARA telescope optics and M4 deformable mirror, the component whose absence in W2 explains the anomalous Coudé phase."}],"review_version":2}