{"id":"45e01f56-0cf9-47e2-98d8-8ca1a83b36a9","arxiv_id":"2608.10751","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Multi-epoch spectropolarimetry of 6 symbiotic and 18 red giant stars shows frequent continuum polarization variability, with Hα line polarization detected in only one symbiotic star, UV Aur.","lead":"This paper presents a 26-month optical spectropolarimetric survey of 24 evolved stars, reporting that continuum polarization varies between epochs in many symbiotic and red giant systems. It offers a rare long-baseline dataset for probing the unresolved shapes of circumstellar envelopes.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The variability claim lacks quantitative significance testing: no error bars on continuum polarization and only two epochs, so 'significant variation' is not established.","rationale":"The reader's weakest assumption is that instrumental and interstellar polarization are constant offsets; I partially agree, but the more load-bearing gap is internal: no error bars or significance tests accompany the epoch-difference claims. The manuscript actually provides the exact formula needed to compute uncertainties, so the missing step is elementary and should have been included. The central claim in the abstract ('significant variation between observation epochs') is falsifiable only if the differences are compared with the uncertainties. Since the continuum measurements are medians of 200-pixel windows per order, a simple error propagation from the extracted flux and the 4 HWP positions is available. The absence of this analysis makes the strongest claim unverified rather than wrong. The paper remains a valuable dataset contribution and the qualitative statements are plausible, but the conditional verdict is appropriate: release the data and add error bars and significance tests. I would not reject because the raw data and literature anchors provide independent value; I would not accept as-is because the headline claim is not quantitatively supported.","tokens_in":26198,"tokens_out":3132,"duration_ms":35092,"concrete_test":"Re-plot Figures 5–7 with per-order uncertainty bars computed from the exposure-level S/N via the equation in Section 2.3 (sigma_P = 1/(sqrt(N)/2 (S/N))), and run a two-sample Welch t-test on the two epoch continuum polarization measurements for each target. Count the targets with |Delta P| > 3 sigma_P_combined; if the fraction is not a clear majority of the 24 targets, the abstract's 'significant variation' claim is not supported and should be downgraded to 'candidate variability pending error analysis'.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim—that continuum polarization of most symbiotics and red giants varies significantly between epochs—rests on a quantitative statement that the analysis never quantifies. Section 2.3 supplies the propagation formula sigma_P = 1/(sqrt(N)/2 (S/N)) and states that a precision of 0.1–0.3% is desired, but the continuum polarization plots (Figures 5–7) show neither error bars nor the SNR attained per order. With only two epochs per target, 'significant variation' can only mean the epoch difference exceeds the measurement uncertainty; the paper does not demonstrate this for any individual target. The instrumental-polarization and ISP stability assumptions flagged by the reader are systematic-offset concerns, but even if both are perfectly stable, the random uncertainties still need to be propagated. Without a statistical test, the variability claim is an assertion, not a result. This matters because the abstract's headline finding and the paper's legacy value as a variability dataset both depend on that comparison.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents multi-epoch optical spectro-polarimetric observations (March 2024 to May 2026) of 6 symbiotic stars and 18 red giants obtained with ProtoPol on the PRL 2.5 m telescope. It reports line-polarization behavior across Hα and the Raman-scattered 6830/7088 Å features, together with continuum polarization measurements at two or more epochs per target. The central claim, stated in the abstract and repeated in Section 4, is that the continuum polarization of most symbiotics and red giants shows 'significant variation between observation epochs', while Hα line polarization is generally absent. The paper interprets these variations as evidence for changing, asymmetric circumstellar scattering geometries.","tokens_in":26261,"tokens_out":5413,"duration_ms":52800,"significance":"If the variability claim is quantitatively supported, this dataset is genuinely valuable: it is one of the few multi-epoch spectro-polarimetric surveys spanning more than two years for evolved cool stars, and it could constrain the geometry and time evolution of unresolved circumstellar scattering regions. The paper also connects measurements to published values for several objects (e.g., Serkowski 1974; Nikolov 2022), and the line-polarization figures include uncertainty shading. However, the headline variability result currently lacks the statistical and systematic-error support needed to establish it, so the dataset's legacy value depends on fixing that gap.","major_comments":[{"comment":"The abstract's claim that continuum polarization 'showed significant variation between observation epochs' is not quantified anywhere. Section 2.3 gives the formula σ_P = 1/(sqrt(N)/2 (S/N)) and states that a precision of 0.1–0.3% is desired, but the continuum polarization plots (Figures 5–7) show neither error bars nor the achieved SNR per echelle order, and no per-target ΔP or significance test is reported. With only two epochs for most targets, 'significant' can only mean that the epoch difference exceeds the measurement uncertainty, and that comparison is never made. Please provide per-epoch continuum P and θ with uncertainties for every target, state the SNR achieved in the continuum bins, and perform a statistical test (e.g., ΔP/σ or a chi-square or t-test against the null hypothesis of constant polarization) for each source. Without this, the central variability claim is an assertion rather than a demonstrated result.","section":"§2.3, Figs. 5–7; abstract"},{"comment":"The decision to omit instrumental polarization and interstellar polarization (ISP) corrections assumes that both are constant offsets across the 26-month campaign and across the relevant wavelength ranges, but no stability monitoring is presented. The text acknowledges the assumption ('This assumption is justified provided that the instrumental contribution remains stable over the relevant wavelength range and observing epochs'), yet there is no demonstration that ProtoPol's sub-0.1% instrumental polarization stays stable under different temperatures, flexures, hour angles, or airmasses. Please add repeated observations of unpolarized and polarized standard stars across the campaign, or otherwise bound the systematic drift, and propagate that systematic uncertainty into the epoch-difference significance estimates.","section":"§2.3"},{"comment":"The statement that T CrB 'has developed significant (~1.0%) intrinsic polarization' in the second epoch is not supported by the data as presented, because no instrumental-polarization or ISP correction has been applied. The comparison with the absolute polarization value from Nikolov (2022) assumes that the ProtoPol instrumental zero-point equals that of the earlier instrument; without validating this, the second-epoch excess cannot be attributed to intrinsic polarization. Please either apply a justified zero-point calibration or rephrase this as a relative change that needs confirmation.","section":"§3.1, T CrB paragraph"}],"minor_comments":[{"comment":"The AG Peg first-epoch label in Figure 1 reads 'May 2024', but Table 1 lists 27-04-2024; please make captions and tables consistent.","section":"Figure 1 and Table 1"},{"comment":"The T CrB epochs are listed as 25-01-2025, 01-03-2025, 16-01-2026, 18-02-2025, and the text says the monitoring covered 'over 16 months (2 orbital cycles)'; the dates span at most ~13 months (or ~12 months if the last date is a typo). Please correct the chronology and the period statement.","section":"Table 1 and §3.1 T CrB text"},{"comment":"The sentence 'AG Per shows a faint Raman-scattered λ6830 Å emission feature' refers to AG Peg; please fix the name.","section":"§3.1, AG Peg paragraph"},{"comment":"'Thompson scattering' should be 'Thomson scattering' in the UV Aur paragraph and in the earlier discussion of Hα polarization mechanisms.","section":"§3.1, UV Aur paragraph and earlier discussion"},{"comment":"Virtanen et al. (2020) is listed with journal 'Nature Medicine'; the correct journal is Nature Methods.","section":"References"},{"comment":"Figures 6 and 7 do not state the manual flux shift applied to the second-epoch spectra, although Figure 5 does; please add the same note for clarity.","section":"Figures 6 and 7 captions"}],"recommendation":"major_revision","confidential_remarks":"This is a dataset paper whose value rests on the variability claims. The main technical gap is the absence of error bars and significance tests for the continuum polarization measurements, which is straightforward to address with the existing data by propagating the Section 2.3 noise formula and adding per-target statistics. The instrumental/ISP stability assumption also needs explicit validation; without it, epoch differences could be partly systematic. The paper is not internally circular and the literature comparisons are appropriate, but the central claim is currently under-supported. Major revision is therefore appropriate rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things you should know. First, this is a real dataset contribution: multi-epoch (up to 26 months) optical spectropolarimetry of 6 symbiotic stars and 18 red giants, with first-ever spectropolarimetric coverage for AG Peg, RW Hya, and most of the red giants. The comparisons to prior photopolarimetric results (Serkowski, Brandi, Nikolov) are sensible, and the individual source notes, like the UV Aur Hα polarization enhancement and the T CrB continuum rise, are genuinely interesting leads. This is not a reanalysis of published data; it is new observations.\n\nSecond, the central claim — \"significant variation\" in continuum polarization between epochs — is not quantitatively supported. With only two epochs for most targets and no error bars on the continuum polarization values in Figures 5–7, \"significant\" is an assertion. The paper gives the uncertainty formula (sigma_P = 1/(sqrt(N)/2 (S/N))) but never reports the achieved SNR or per-point uncertainties for the continuum. The instrument-stability and ISP-constancy assumptions are acknowledged in Section 2.3 and are reasonable as stated, but even if both are perfectly stable, the random uncertainties still need to be propagated. For a few targets, like X Her (3% to <0.5%) or LQ Her, the changes are large enough that they will survive error bars; for the bulk of the sample, we simply do not know. The wording \"significant\" should be softened until the errors are shown.\n\nOther soft spots: no machine-readable data release. For a paper whose main product is the dataset, that is a problem; the community cannot re-derive the variability or check the claims. The line-spectra plots do include gray error strips, so the issue is specific to the continuum measurements, which happen to be the headline result.\n\nThe paper is honest about its limitations, the references are appropriate, and there is no circularity. The weakness is in the analysis and reporting, not in the observations. I would send it to a referee, but with a clear request: add error bars to the continuum polarization, run a simple chi-square or bootstrap significance test on the two-epoch differences, and release the polarization tables as data behind the paper. If those are done, this is a solid contribution that people will cite.","headline":"Useful new dataset, but the 'significant variation' headline is not backed by error bars or significance tests — publish after tightening the analysis.","tokens_in":26869,"tokens_out":2237,"would_cite":true,"duration_ms":23717,"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":"Multi-epoch spectropolarimetry of 24 evolved cool stars shows continuum polarization varying between epochs over 26 months, while Hα emission in most symbiotics stays unpolarized.","keywords":["Spectro-polarimetry","Symbiotic stars","Red giant stars","stellar polarization","circumstellar matter","multi-epoch observations","Raman scattering","ProtoPol"],"falsifier":"Point the same instrument at a stable, non-variable polarized standard star across the same 26-month window; if its measured polarization degree or angle drifts by the same amplitude as the claimed source variability, the epoch-difference interpretation collapses. Alternatively, a single epoch of the same targets with an independent polarimeter should reproduce the observed polarization levels and angles if the variability is intrinsic.","tokens_in":25937,"feed_emoji":"🔭","tokens_out":6377,"duration_ms":53610,"temperature":0.7,"pith_summary":"This paper reports a 26-month spectropolarimetric campaign of 6 symbiotic stars and 18 red giants, all observed with the same medium-resolution echelle spectropolarimeter. Its central finding is that continuum polarization changes between epochs for most targets, while the Hα emission line stays essentially unpolarized in most symbiotics—a contrast with the strongly polarized Raman-scattered O VI features at 6830 and 7088 Å. The authors argue that epoch-to-epoch changes in polarization degree and position angle trace rearrangements of asymmetric circumstellar scattering material on timescales of months to years. They also note that this is among the rare multi-epoch spectropolarimetric datasets spanning more than two years for such evolved stars, providing a resource for studying unresolved circumstellar geometry.","feed_headline":"24 evolved stars show shifting polarization over 26 months","feed_subtitle":"Multi-epoch spectropolarimetry reveals changing scattering geometry in symbiotics and red giants.","key_machinery":"The central machinery is multi-epoch optical spectropolarimetry with the ProtoPol instrument, a medium-resolution echelle spectropolarimeter covering 4000–9600 Å at 0.4–0.75 Å resolution, from which Stokes parameters are derived from ordinary and extraordinary ray intensities at four half-wave-plate positions. The degree and angle of polarization are computed per spectral bin with an adaptive binning scheme that holds polarization uncertainty roughly constant across the profile. Comparison of the same targets across epochs from March 2024 to May 2026 is the variability probe. The physical mechanisms being tested are scattering processes: Thomson scattering and Raman scattering of Lyman-β photons can polarize Hα, while scattering by dust or molecules in a non-spherical envelope polarizes the continuum; the epoch comparison isolates changes in scattering geometry from fixed offsets like instrumental polarization and interstellar polarization.","core_discovery":"The authors establish that continuum polarization is common—and often time-variable—among evolved cool stars. Across the symbiotic sample, Hα emission shows no notable polarization signature in most systems (AG Peg, RW Hya, T CrB, AG Dra, Z And), whereas the Raman-scattered features λλ6830, 7088 Å, where present, do show polarization enhancement or angle rotation (AG Dra, Z And). In the red-giant sample, every one of 18 targets shows measurable continuum polarization, with several (LQ Her, ω Vir, ST UMa, SW Vir, U Her, X Her) exhibiting clear epoch-to-epoch changes, sometimes dropping from roughly 2–3% to near zero. Because a spherically symmetric unresolved envelope would produce no net linear polarization, the detection of continuum polarization across the full red-giant sample implies that weak asymmetries are common in their extended atmospheres and circumstellar envelopes, and the temporal variability indicates that those scattering geometries evolve between observation epochs.","pith_inferences":["A natural extension is to correlate the continuum polarization changes with optical light-curve phase for the Mira and semiregular variables; a pulsation-locked polarization signal would strengthen the intrinsic interpretation and link the scattering geometry to atmospheric shocks.","The contrast between polarized Raman O VI features and unpolarized Hα in the same symbiotics suggests that Hα photons are not formed in the same neutral-wind scattering region as the O VI photons, or that the Hα scattering region is viewed at an angle where net polarization cancels; this can be tested by modeling the two scattering geometries jointly.","The assumption of constant interstellar polarization could be checked by measuring polarization of nearby field stars; if interstellar polarization varies along the line of sight on arcminute scales, part of the claimed variability could be foreground rather than intrinsic to the sources."],"forward_implications":["The non-detection of Hα polarization in most symbiotics, despite strong Raman-feature polarization, indicates that the Hα line-forming region is either more symmetric or viewed at a geometry that suppresses net polarization in these systems.","Epoch-to-epoch continuum polarization changes in red giants like X Her, U Her, and SW Vir provide direct evidence that their circumstellar scattering geometry changes on timescales of months to years, consistent with clumpy or episodic mass loss.","The dataset offers 24 sources with multi-epoch Stokes spectra that can be used to test theoretical predictions of phase-dependent polarization in symbiotic binaries and of pulsation-linked polarization in Miras and semiregular variables.","If the T CrB continuum polarization increase from roughly 0.5% to 1.1% is intrinsic, it may signal a developing asymmetry in the accretor's environment ahead of the predicted outburst."],"supporting_citations":[{"why":"Supplied the original spectropolarimetric detection of polarization in Raman-scattered 6830/7088 features in symbiotics, which this paper extends to multi-epoch behavior.","marker":"H. Schmid & H. Schild 1994"},{"why":"Phase-resolved multi-epoch spectropolarimetry of AG Dra establishing phase-locked polarization-angle variations; the comparison baseline for the AG Dra results.","marker":"H. Schmid & H. Schild 1997a"},{"why":"Multi-epoch survey of Z And over one orbital cycle showing phase-locked variations; used as the baseline for the Z And results.","marker":"H. Schmid & H. Schild 1997b"},{"why":"Spectropolarimetric survey of Raman features in symbiotics; source of the expected polarization amplitudes for these features.","marker":"T. Harries & I. Howarth 1996a"},{"why":"Theoretical prediction that Raman scattering of Lyβ by neutral hydrogen produces broad, polarized Hα wings; the mechanism used to interpret UV Aur's Hα polarization.","marker":"H.-W. Lee 2000"},{"why":"Multi-epoch imaging polarimetry of symbiotics showing dynamic continuum polarization; the prior work this campaign extends to spectropolarimetry.","marker":"E. Brandi et al. 2000"},{"why":"The ProtoPol instrument characterization paper, establishing sub-0.1% instrumental polarization and the data-reduction pipeline used here.","marker":"A. Maiti et al. 2026b"},{"why":"Supplies the uncertainty formula relating polarization error to signal-to-noise ratio, which underlies the required SNR and adaptive binning.","marker":"F. Patat & M. Romaniello 2006"}],"fun_headline_variants":["26 months of polarization shifts in 24 evolved stars","Variable polarization across 24 cool evolved stars","Shifting scattering geometry in symbiotics and red giants","Multi-epoch spectropolarimetry reveals time-varying polarization","24 stars, 26 months, changing polarization"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the instrumental polarization of the spectropolarimeter and the interstellar polarization along each line of sight both remain constant across the 26-month campaign, so any epoch-to-epoch difference reflects a real change in the source rather than a change in the instrument or foreground.","fun_headline_variants_meta":{"raw":{"variants":["26 months of polarization shifts in 24 evolved stars","Variable polarization across 24 cool evolved stars","Shifting scattering geometry in symbiotics and red giants","Multi-epoch spectropolarimetry reveals time-varying polarization","24 stars, 26 months, changing polarization"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00105,"raw_usage":{"total_tokens":4470,"prompt_tokens":1067,"completion_tokens":3403,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":683,"completion_tokens_details":{"reasoning_tokens":3329}},"tokens_in":683,"tokens_out":3403,"duration_ms":22190,"temperature":1.0,"reasoning_tokens":3329,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T18:02:45.133847+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Point the same instrument at a stable, non-variable polarized standard star across the same 26-month window; if its measured polarization degree or angle drifts by the same amplitude as the claimed source variability, the epoch-difference interpretation collapses. Alternatively, a single epoch of the same targets with an independent polarimeter should reproduce the observed polarization levels and angles if the variability is intrinsic.","supporting_citations":[],"review_version":1}