{"id":"398e7f0b-c6da-4bdc-8bdd-77afe0d66692","arxiv_id":"2506.15027","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":11,"one_line_summary":"Using 416 nm stellar intensity interferometry with VERITAS, the photosphere of gamma Cassiopeiae is measured as an ellipse with axis ratio 1.28 and near-critical rotation.","lead":"Astronomers measured the flattened, oblate shape of the rapidly spinning star gamma Cassiopeiae using the VERITAS gamma-ray telescopes as an optical intensity interferometer. It is the first stellar oblateness measurement made with this technique, and it opens a new way to study stars that rotate near their breaking point.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 416 nm SII band admits Hδ line emission from γ Cas's decretion disk, and the visibility model excludes the disk; the fitted 1.28 axis ratio may trace disk elongation rather than photospheric oblateness.","rationale":"The reader's weakest assumption is the same as mine: disk emission in the 416 nm band is unmodeled. I considered the §3.6 SNR censoring as an alternative. It is a real statistical issue—a hard threshold applied to the noisy area can bias surviving points, and if noise is orientation-dependent it could in principle synthesize oblateness. But the selection operates symmetrically in baseline length for a round source and, if anything, biases fitted diameters toward smaller values at low SNR, which would tend to reduce a true elongation; the authors also vary the threshold and handle the suspicious T1T4 point transparently. The disk contamination, by contrast, is physically expected to be elongated along the same axis as the fitted major axis, is explicitly acknowledged in Figure 15 ('The disk is not included in our model'), and is not estimated either spectrally or spatially. Therefore it is the most load-bearing threat to the photospheric interpretation. The uniform-ellipse result itself is supported by the tabulated visibilities, the slice analysis, and the RvZ fit; the claim that SII can detect shape is plausible. However, a benchmark 'first oblateness measurement' should quantify the disk contribution in the exact passband before the photosphere-only conclusion is accepted. The recommended disposition is unchanged: conditional acceptance pending the disk check.","tokens_in":21704,"tokens_out":16119,"duration_ms":197283,"concrete_test":"Use a Be-disk radiative transfer model (e.g., Sigut et al. 2020/BEDISK) with the actual 416 nm filter transmission to compute the disk flux fraction and the disk complex visibility at every (u,v) point in Table B1. Then refit the Table B1 visibilities with a photosphere+disk model, combining the RvZ photosphere image with the disk image and marginalizing over disk scale and density. If the recovered uniform-ellipse and RvZ parameters shift by less than the quoted systematic errors (Δr ≤ 0.02, ΔPA ≤ 7°), the disk concern is retired; if they shift by more, the headline oblateness is not established as photospheric.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim—first intensity-interferometric measurement of photospheric oblateness—requires that the measured anisotropy comes from the photosphere alone. Section 2 states that the observations use a λ=416 nm, 10 nm bandpass filter. Section 5 and Figure 15 state that in this same bandpass the Hδ line at 4101 Å is filled by emission from γ Cas's decretion disk, and the caption adds 'The disk is not included in our model.' A Be decretion disk is geometrically thin and lies in the equatorial plane, so its projected image is elongated perpendicular to the rotation axis (PA ≈116°), exactly the direction in which the fitted uniform ellipse has its major axis. If any non-negligible fraction of the light in the 416 nm band comes from the disk—line emission in the filter wings or free-free/continuum excess—then the measured squared visibilities are a flux-weighted average of photospheric and disk visibilities. Because an Hδ-emitting region of a few stellar radii (≈0.5–1 mas) is not fully resolved at baselines of 50–150 m, its visibility at the VERITAS baselines is not negligible and is orientation-dependent. An equatorially concentrated disk component therefore lowers |V|² preferentially at baselines oriented along the major axis, mimicking or inflating the 1.28±0.04 axis ratio and the near-critical rotation solution. The paper does not estimate the disk flux fraction in the 416 nm band, the disk's spatial scale, or its complex visibility at the VERITAS baselines; the spectral comparison in Figure 15 is not a spatial model. A secondary issue is the §3.6 hard SNR cut, which can also imprint anisotropy if noise correlates with baseline orientation, but the disk omission directly targets the photospheric interpretation of the shape signal.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents stellar intensity interferometry (SII) observations of the rapid rotator γ Cassiopeiae using the VERITAS telescopes, with a 416 nm narrow-band filter. From more than 160 pair-hours of data, the authors extract squared visibilities as a function of baseline length and orientation, then fit them with (i) a uniform ellipse model and (ii) a Roche–von Zeipel rapid-rotator model with PHOENIX atmospheres. The uniform ellipse fit yields a minor-axis angular diameter of 0.43±0.02 mas, an axis ratio of 1.28±0.04, and a rotation-axis position angle of 116°±5°. The Roche–von Zeipel model gives an equatorial angular diameter of 0.604+0.041−0.034 mas and a 1σ lower limit on the rotation rate of 97.7% of breakup. The authors conclude that this is the first measurement of a stellar photosphere's oblateness using intensity interferometry, and that the measured orientation is consistent with infrared interferometric measurements of the decretion disk.","tokens_in":21967,"tokens_out":7018,"duration_ms":69654,"significance":"If the central claim is secure, this is a milestone for intensity interferometry: it demonstrates that SII with IACT-class collectors can measure not only stellar angular sizes but also photospheric shapes and orientations at sub-milliarcsecond scales. The paper's strengths include the publication of the full visibility data table (Table B1), a transparent bootstrap-based fitting procedure, and the internal consistency between the simple geometric model and the more physical atmosphere model. The inferred rotation rate is consistent with independent spectroscopic and interferometric constraints on γ Cas. The result would open a new niche for SII in stellar astrophysics, complementing Michelson interferometry at longer wavelengths.","major_comments":[{"comment":"The 416 nm bandpass (10 nm wide) includes the Hδ line at 4101 Å. The paper itself notes in Section 5 and in the caption of Figure 15 that the Hδ core is filled in by emission from γ Cas's decretion disk, and that 'The disk is not included in our model.' Because the disk is geometrically thin and lies in the equatorial plane (position angle ≈116°, consistent with the fitted rotation axis), any non-negligible disk contribution in this band will add an orientation-dependent visibility term that can mimic or bias the measured oblateness. The paper does not estimate the disk flux fraction in the 416 nm band, the angular scale of the emitting region, or its complex visibility at the 50–150 m baselines. I request a quantitative assessment—either from the ELODIE spectrum convolved with the filter transmission or from a simple disk model—showing that the fitted ellipse parameters (θ_min, r, φ*) and the near-critical rotation solution are robust to the inclusion of a disk component. Without this, the central claim of a photospheric oblateness measurement is not fully supported.","section":"§5, Figure 15"},{"comment":"All squared-visibility measurements below a fixed signal-to-noise threshold are discarded and excluded from the fits. The excluded points are predominantly the longest-baseline, lowest-visibility measurements, which are precisely the data that most strongly constrain the angular diameter and shape. Discarding them rather than modeling them as censored data can bias the fitted parameters, especially if the noise distribution is asymmetric or if the exclusion correlates with baseline orientation. I recommend either performing a censored-likelihood fit that includes all measurements (treating sub-threshold points as upper limits) or demonstrating with simulations that the threshold cut—including the 3.0–4.1 fs variation—does not bias θ_min, r, and φ*. The current systematic uncertainty estimate varies the threshold value but does not test the effect of the selection itself.","section":"§3.6, Figure 7"}],"minor_comments":[{"comment":"In the row for 2024-02-21, the telescope pairs are listed as '(1,3), 2,3)'; the second pair is missing its opening parenthesis and should read '(2,3)'.","section":"Table 1"},{"comment":"The text states that T_pole = 26500 K provides a reasonable match to the spectrophotometry, but Table 3 reports '≃26500−28000'; the value and range should be made consistent.","section":"§5, Table 3"},{"comment":"The reduced χ² values for the uniform disk (213/114 ≈ 1.87) and uniform ellipse (170/112 ≈ 1.52) fits are notably larger than unity; the paper should discuss the likely sources of the excess scatter (e.g., residual correlated noise, unmodeled surface structure) and consider whether the quoted statistical uncertainties need to be rescaled.","section":"§4, Table 2"},{"comment":"The caption states that the zero-baseline squared visibility is fixed to the value from the uniform ellipse fit when fitting the individual uv slices; this constraint can bias the recovered modulation in θ_UD(φ_b). I suggest checking whether freeing this parameter per slice changes the derived δ and φ*.","section":"§4, Figure 8"},{"comment":"The wording 'a lower limit on the angular velocity very near the critical value, Ω/Ωc = 0.977' is ambiguous: the best-fit value is 0.990 and 0.977 is the 1σ lower limit. Please clarify this phrasing in the text and abstract.","section":"§6"}],"recommendation":"major_revision","confidential_remarks":"The paper is well within the scope of the journal and the data are of high quality, but the disk-contamination issue is exactly the kind of hidden systematic that can invalidate a 'first measurement' claim. Given the collaboration's modeling resources, the requested quantitative test should be feasible. The censored-data concern is also important for the statistical rigor. I therefore recommend major revision rather than rejection or acceptance. If the authors can convincingly show that the disk contribution is negligible or properly model it, the paper would be a strong addition to the literature."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The claim in the abstract holds up as a first: nobody has used intensity interferometry to measure a stellar photosphere's shape and orientation. The VERITAS data, the uniform-ellipse fit, and the Roche–von Zeipel model all point the same way, and the authors are appropriately careful with the one suspicious T1T4 point, reporting fits with and without it and propagating threshold/cut variations into systematics. The full visibility table is in the paper, which is genuinely useful for reanalysis.\n\nThat said, the central weakness is exactly the one the paper itself flags in the Figure 15 caption: the 416 nm bandpass includes Hδ, which is filled by emission from the decretion disk, and the disk is not in the model. The disk is equatorial and therefore elongated along the same sky direction as the photosphere's major axis. If even a modest fraction of the bandpass flux comes from the disk, the fitted 1.28±0.04 axis ratio and the near-critical rotation rate will be inflated. The paper gives no estimate of the disk flux fraction, its angular scale, or its complex visibility at VERITAS baselines. The spectral comparison in Figure 15 is not a spatial model. This is a load-bearing assumption for the central claim, not a cosmetic detail.\n\nThe sub-threshold SNR cut is a softer issue. Discarding low-SNR points rather than modeling censored data can in principle bias the anisotropy if noise correlates with baseline orientation, but the authors vary the threshold and quote systematics, so I would treat this as a moderate concern, not a fatal one.\n\nOn citation pattern and internal consistency, the paper is honest: the fitted parameters are compared with independent Hα and IR interferometric constraints, and the T1T4 outlier is shown in red and discussed. No circularity.\n\nWho gets value from this: SII practitioners deciding what the technique can do, and stellar astronomers modeling Be stars. It deserves a serious referee, but I would want a quantitative disk-contamination estimate (or a narrower filter) before endorsing the numerical result as a benchmark. This is a conditional accept for me, not a desk reject.","headline":"A genuine first SII oblateness measurement, but the unquantified Hδ disk contamination in the 416 nm band is a real soft spot that should be addressed before the axis ratio becomes a benchmark.","tokens_in":23013,"tokens_out":3145,"would_cite":true,"duration_ms":35427,"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":"Intensity interferometry measures an oblate stellar photosphere for the first time, on gamma Cassiopeiae.","keywords":["stellar intensity interferometry","gamma Cassiopeiae","rapid rotators","photosphere oblateness","Be stars","squared visibility","Roche-von Zeipel model","gravity darkening"],"falsifier":"Measure gamma Cas again through a narrow band that excludes H-delta, for example a 10 nm band centered near 420 nm, and refit the uniform ellipse; if the fitted axis ratio or position angle moves by more than the quoted uncertainties, the oblateness signal is contaminated by decretion-disk emission rather than indicating the photosphere.","tokens_in":21400,"feed_emoji":"🌟","tokens_out":12747,"duration_ms":117334,"temperature":0.7,"pith_summary":"The paper reports the first measurement of an oblate stellar photosphere using intensity interferometry. Observing gamma Cassiopeiae at 416 nm with an array of four 12-meter telescopes, it finds that the star's disk is measurably elongated: a uniform-ellipse fit gives a minor-axis angular diameter of $0.43\\pm0.02$ mas and a major-to-minor axis ratio of $1.28\\pm0.04$, with the rotation axis at position angle $116^\\circ\\pm5^\\circ$. A rapidly rotating stellar atmosphere model with limb and gravity darkening matches the same visibility data with an equatorial angular diameter of $0.604^{+0.041}_{-0.034}$ mas and a rotation rate with a 1 $\\sigma$ lower limit of 97.7 percent of breakup. These parameters agree with earlier H-$\\alpha$ spectroscopy and infrared interferometry of the star's decretion disk. If correct, the measurement establishes that intensity interferometry can recover stellar shape and orientation, not just size.","feed_headline":"First oblate stellar photosphere measured by intensity interferometry","feed_subtitle":"Gamma Cassiopeiae's photosphere is 28 percent wider at the equator and spins at 97.7 percent of breakup speed.","key_machinery":"The central mechanism is stellar intensity interferometry: the normalized correlation of photomultiplier currents at two telescopes, whose fitted peak area is proportional to the squared visibility $|V|^2$ for a given projected baseline. The argument is carried by (1) coverage of the $u$--$v$ plane from six telescope pairs over many hour angles, (2) a uniform-ellipse visibility model, $A(H)=C_{\\rm norm}\\,[2J_1(\\pi\\theta_{\\min}s)/(\\pi\\theta_{\\min}s)]^2$, with $s$ encoding the ellipse axis ratio and orientation, and (3) a rapidly rotating Roche--von Zeipel model with limb darkening, gravity darkening, a grid of stellar-atmosphere intensity fields, and fixed inclination and parallax. These objects convert baseline-dependent visibility amplitudes first into a geometric description of the photosphere (minor-axis diameter, oblateness, position angle) and then into physical quantities (equatorial radius, near-breakup rotation rate).","core_discovery":"The central discovery, stated on the paper's own terms, is that intensity-interferometry visibilities taken over baselines of varied length and orientation resolve the equatorial bulge of a rapidly rotating star. Fitting the squared visibilities with a uniformly illuminated ellipse yields a minor-axis angular diameter of $0.43\\pm0.02$ mas, a major-to-minor axis ratio of $1.28\\pm0.04$, and a rotation-axis position angle of $116^\\circ\\pm5^\\circ$ (statistical; comparable systematic uncertainties are reported). A Roche--von Zeipel rapid-rotator model with limb and gravity darkening describes the same data with an equatorial angular diameter of $0.604^{+0.041}_{-0.034}$ mas, an equatorial radius of $10.9^{+0.8}_{-0.6}\\,R_\\odot$, a 1 $\\sigma$ lower limit of 97.7 percent of the breakup rotation rate, and a position angle of $114.7^{+6.4}_{-5.7}$ degrees. The paper states that this is the first measurement of an oblate photosphere using intensity interferometry.","pith_inferences":["A direct test of the disk-contamination assumption would be multi-band intensity interferometry across the Balmer jump; a stable fitted axis ratio would confirm the photospheric origin, while a wavelength-dependent ratio would map the disk contribution.","The same u--v coverage could be turned on other bright Be stars, and because the visibility zeros depend on baseline orientation, shape extraction does not require full model-independent imaging.","If disk emission does contaminate the 416 nm band, the method would still be useful, since the line-emitting disk itself could be mapped in H-alpha or H-delta light with the same correlator.","With longer optical baselines, the visibility function's dependence on the intensity distribution could reveal latitude-dependent gravity darkening rather than only the outer ellipse."],"forward_implications":["Rapid rotators among bright stars can now have their photospheric oblateness and spin-axis orientation measured at optical wavelengths, not just their disk geometry at infrared wavelengths.","Agreement between photosphere and disk position angles becomes a direct test of whether decretion disks form in the equatorial plane.","For rapid rotators, circular-disk fits are orientation-dependent, so previous single-orientation intensity-interferometry size measurements may need elliptical reanalysis.","Near-critical stellar rotation models must now reproduce geometric constraints from intensity interferometry, not just spectra and spectral energy distributions.","More hour-angle coverage and additional telescopes would sharpen the derived equatorial radius and rotation rate because the remaining uncertainty is statistical plus comparable systematic."],"supporting_citations":[{"why":"Establishes that correlated intensity fluctuations at two telescopes measure source angular structure, the physical principle the whole measurement rests on.","marker":"Hanbury Brown & Twiss 1954"},{"why":"Demonstrates the use of an imaging atmospheric Cherenkov telescope array as an intensity-interferometry light collector, the instrumental method this paper adapts.","marker":"Abeysekara et al. 2020"},{"why":"Supplies the correlation-processing, stray-light correction, and noise-estimation procedures reused here, and its reanalysis validates the new uncertainty treatment.","marker":"Acharyya et al. 2024"},{"why":"Reports a prior intensity-interferometry angular diameter for gamma Cas with a circular model, providing the comparison that motivates elliptical fitting for rapid rotators.","marker":"Abe et al. 2024"},{"why":"Provides H-alpha spectroscopy and a disk model giving the disk position angle used to check that the fitted photosphere axis agrees with the disk orientation.","marker":"Sigut et al. 2020"},{"why":"Introduces the Roche--von Zeipel rapid-rotator model with limb and gravity darkening that the paper fits to the visibilities.","marker":"Aufdenberg et al. 2006"},{"why":"Supplies the current implementation of the rotating-star model grid used to compute synthetic visibilities for gamma Cas.","marker":"Sackrider & Aufdenberg 2023"},{"why":"Provides the model stellar atmosphere intensity fields from which the synthetic photosphere intensities are interpolated.","marker":"Hauschildt & Baron 1999"},{"why":"Gives the uniform-ellipse visibility formula used to extract size, oblateness, and position angle from the baseline-dependent visibilities.","marker":"Tycner et al. 2006"},{"why":"Constrains the stellar inclination to 60 degrees, a fixed parameter in the rotating-model fits.","marker":"Lailey & Sigut 2024"}],"fun_headline_variants":["Intensity interferometry measures Gamma Cas's oblate photosphere","First oblate stellar photosphere via intensity interferometry","VERITAS intensity interferometry resolves Gamma Cas's bulge","Gamma Cas's 28% wider equator measured with intensity interferometry","Gamma Cas near breakup: oblate photosphere via intensity interferometry"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that light from gamma Cas's decretion disk contributes negligibly to the correlated 416 nm signal, even though the 10 nm bandpass includes the H-delta line (which the disk fills with emission) and the model does not include the disk.","fun_headline_variants_meta":{"raw":{"variants":["Intensity interferometry measures Gamma Cas's oblate photosphere","First oblate stellar photosphere via intensity interferometry","VERITAS intensity interferometry resolves Gamma Cas's bulge","Gamma Cas's 28% wider equator measured with intensity interferometry","Gamma Cas near breakup: oblate photosphere via intensity interferometry"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000899,"raw_usage":{"total_tokens":3938,"prompt_tokens":1076,"completion_tokens":2862,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":692,"completion_tokens_details":{"reasoning_tokens":2777}},"tokens_in":692,"tokens_out":2862,"duration_ms":21346,"temperature":1.0,"reasoning_tokens":2777,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:46:27.730746+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure gamma Cas again through a narrow band that excludes H-delta, for example a 10 nm band centered near 420 nm, and refit the uniform ellipse; if the fitted axis ratio or position angle moves by more than the quoted uncertainties, the oblateness signal is contaminated by decretion-disk emission rather than indicating the photosphere.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides H-alpha spectroscopy and a disk model giving the disk position angle used to check that the fitted photosphere axis agrees with the disk orientation."},{"cited_title":"L., & Aufdenberg, J","cited_arxiv_id":null,"evidence_quote":"Supplies the current implementation of the rotating-star model grid used to compute synthetic visibilities for gamma Cas."},{"cited_title":"C., Zavala, R","cited_arxiv_id":null,"evidence_quote":"Gives the uniform-ellipse visibility formula used to extract size, oblateness, and position angle from the baseline-dependent visibilities."},{"cited_title":"D., & Sigut, T","cited_arxiv_id":null,"evidence_quote":"Constrains the stellar inclination to 60 degrees, a fixed parameter in the rotating-model fits."}],"review_version":1}