{"id":"c675086d-7cfe-431b-b590-b3367c441e1a","arxiv_id":"2509.02681","paper_version":1,"verdict":"UNVERDICTED","confidence":"HIGH","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"A proposal to observe AB Aurigae with the Roman Coronagraph that could settle whether AB Aur b is a real protoplanet and test CGI's ability to image disks and planets.","lead":"This white paper proposes using the Roman Space Telescope coronagraph during commissioning to image the AB Aurigae protoplanetary disk and its candidate planet AB Aur b in visible light. Why it matters: the observations could clarify whether AB Aur b is a genuine protoplanet and demonstrate the coronagraph's power for finding fainter planets.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 0.4–0.7 h exposure estimates rely on an unquantified claim that the bright disk does not dominate noise at AB Aur b's separation; archival STIS data can test this directly.","rationale":"The reader's weakest assumption identifies the same condition: that the noise is dominated by the speckle floor rather than the bright disk. My reading sharpens this into a concrete, testable quantity: the local disk surface brightness at the planet's separation relative to the stellar PSF. Because the manuscript is a commissioning proposal rather than a research preprint presenting measurements, the UNVERDICTED classification remains appropriate regardless of whether this feasibility check passes. The concern does not invalidate the science case, but it would strengthen the proposal if addressed with existing archival data before the observation is scheduled.","tokens_in":4210,"tokens_out":8600,"duration_ms":89213,"concrete_test":"Use the archival HST/STIS coronagraph data already shown in Figure 1 to measure the disk surface brightness at 575 nm in a 0.1 arcsec aperture centered 0.6 arcsec from AB Aur at the reported position angle of AB Aur b, divided by the stellar peak flux to obtain a disk/star contrast per resolution element. Repeat at several off-arm azimuths to estimate the local disk background. If the disk contrast per element exceeds roughly 3e-5, add that background in quadrature to the point-source noise model behind Figure 2 and recompute the 10-sigma exposure times quoted in the 'Estimate of Time Needed' section to see whether the 0.4-0.7 hour estimates still hold.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim's quantitative support is the sentence in 'Estimate of Time Needed': 'For a goal of a 10-σ contrast of 10^-6 (set by the speckle floor, not the bright disk), we obtain exposure times of 0.4–0.5 hours in Band 1 and 0.6–0.7 hours in Band 4.' This assumes that once the stellar PSF is suppressed, the noise at the planet's position is the residual speckle floor rather than the photon noise or confusion from AB Aur's own disk. The paper gives no calculation for this, and the Figure 2 exposure-time curves are point-source SNR estimates with no disk surface-brightness term. AB Aur is proposed precisely because its disk is bright and structured out to 1.4 arcsec, with spiral arms at the planet's 0.6 arcsec separation. If the local disk surface brightness per resolution element is comparable to or brighter than the assumed 10^-4 planet contrast, then the stated exposure times undercount the noise and the claimed astrometry and photometry 'unbiased by processing artifacts' become questionable, because separating a point source from a bright spiral arm requires model-dependent subtraction rather than a pure dark-hole measurement. This is the most load-bearing unchecked assumption in the feasibility argument.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper is a Roman Coronagraph white paper proposing commissioning-epoch observations of the AB Aurigae protoplanetary system. It requests Band 1 (575 nm) and Band 4 (825 nm) imaging and polarimetry with the Hybrid Lyot and Shaped Pupil coronagraphs, covering radii from 0.15 to 1.4 arcseconds. The scientific rationale is that AB Aur b, an embedded protoplanet at 0.6 arcseconds, and the structured disk, including its spiral arms, are bright enough that even a modest dark-hole contrast of 1e-5 to 1e-6 would yield substantially better optical data than existing HST and ground-based observations. The paper's central quantitative claim is that a 10-sigma detection at 1e-6 contrast requires only 0.4-0.5 hours in Band 1 and 0.6-0.7 hours in Band 4, assuming noise is set by the speckle floor rather than the bright disk.","tokens_in":4602,"tokens_out":3100,"duration_ms":30475,"significance":"If the feasibility claim holds, the proposed program has high scientific value: AB Aurigae is a rare system with a directly imaged embedded protoplanet, CO spiral arms, and evidence of infalling material, and optical imaging and polarimetry could break degeneracies in dust properties and emission mechanisms. The paper also correctly identifies that current optical data are limited by PSF sampling and calibration biases, and it leverages the full field-of-view complement of CGI modes. The explicit use of previously detected source properties, the complete 360-degree coverage plan, and the presentation of pitch-angle windows for reference stars are useful strengths. However, the central exposure-time estimate is not derived from a stated noise model, and the assumption that the disk does not dominate the noise is unquantified, so the feasibility claim is currently unverifiable from the text.","major_comments":[{"comment":"The exposure-time estimates of 0.4-0.5 h in Band 1 and 0.6-0.7 h in Band 4 are presented without a noise model or an SNR equation. The sentence 'set by the speckle floor, not the bright disk' asserts the key assumption, but no calculation is shown that the residual speckle floor dominates over photon noise and, critically, over the local surface brightness of AB Aur's disk at the planet's 0.6 arcsecond separation. Given that the disk is bright and structured out to 1.4 arcseconds, with spiral arms at the planet location, the disk surface brightness per resolution element could exceed the assumed 1e-4 planet contrast and invalidate the stated times. Please provide a quantitative estimate of the disk contribution (e.g., from archival HST/STIS or SCExAO/CHARIS data) and include it in the integration-time calculation, or clearly state and justify the conditions under which the disk is negligible.","section":"Estimate of Time Needed"},{"comment":"The paper treats 1e-6 contrast as the goal for the exposure time, while the abstract and the opening of the observing description describe 'marginally-successful' dark-hole generation as 1e-5 to 1e-6. At 1e-5 rather than 1e-6, the required exposure times scale by roughly a factor of 100, turning a few hours into many hours to a day. Because the central feasibility claim depends on reaching the more optimistic contrast, the paper should either adopt a conservative contrast for the nominal time estimate, present a sensitivity analysis over 1e-5 to 1e-6, or justify why 1e-6 is a realistic expectation for first-look commissioning observations at these wavelengths.","section":"Abstract and Estimate of Time Needed"},{"comment":"The middle and right panels of Figure 2 show integration time to reach SNR = 10 versus target magnitude for the HLC (at 0.2 arcseconds) and SPC (at 0.5 arcseconds), but neither the assumed instrument throughput, the assumed dark-hole contrast, nor the noise floor is specified. The curves appear to be point-source sensitivity curves, and it is unclear how they correspond to the stated goal of 10-sigma at 1e-6 contrast, especially because AB Aur b is itself a 1e-4 contrast source. Please state the model assumptions behind these curves and explain the mapping between the plotted integration times and the 1e-6, 10-sigma claim.","section":"Figure 2"}],"minor_comments":[{"comment":"The phrase 'striking, inspiring demonstrations' is grammatically awkward; consider 'striking and inspiring demonstrations' or 'striking demonstrations.'","section":"Abstract"},{"comment":"The sentence describing the visit sequence ends with 'followed by a second set of PSF reference observation'; the parallel structure would be clearer as 'a second set of PSF reference observations.'","section":"Observing Description"},{"comment":"Reference [5] contains the placeholder 'HIP xxxxxx' in the title and appears to be a submitted manuscript; please resolve the placeholder or mark the reference as forthcoming.","section":"References"},{"comment":"The gallery images would benefit from scale bars and a north-east arrow in each panel; as presented, the angular scale of the features is not immediately apparent.","section":"Figure 1"},{"comment":"The paper does not explicitly state the astrometric precision required to distinguish between the proposed planet-driven spiral models and alternative explanations; adding a brief quantitative statement would strengthen the case for the proposal.","section":"Anticipated Technology / Science Objectives"}],"recommendation":"major_revision","confidential_remarks":"The paper is a white paper rather than a full technical study, so the absence of a complete noise model is not unusual for the genre, but the exposure-time claim is the explicit quantitative basis for the proposal and is not verifiable from the text. The authors should be asked to supply a short derivation or to cite a validated instrument-simulation result, and to assess the disk surface brightness contribution using the archival data they already reference. The heavy self-citation reflects the authors' prior work on AB Aur b, which is legitimate here because that prior work is the source of the assumed planet contrast and astrometry. The topic is well matched to the journal's astrophysical instrumentation scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Hi [Colleague],\n\nQuick read of arXiv:2509.02681. It's a white paper, not a research preprint: it asks for Roman Coronagraph first-look time to image AB Aurigae. No new data, no derivation, no falsifiable prediction. That's fine for what it is, but it shouldn't be reviewed as a discovery paper.\n\nWhat it does well: the target choice is sensible. AB Aur is bright, has a known protoplanet candidate at 0.6 arcsec, and a disk that spans the CGI field. The proposal explains how HLC and SPC modes can cover the full disk, and why polarimetry would add constraints on dust properties. The pitch-angle windows for reference stars are a nice concrete touch. Exposure times are internally consistent, and the authors are appropriately modest, saying that even 1e-6 contrast would improve on HST data.\n\nThe soft spot is the exposure-time estimate. The sentence in the Estimate of Time Needed section says the 10-sigma contrast is 'set by the speckle floor, not the bright disk,' but no calculation is shown. That is exactly the load-bearing assumption: if the disk surface brightness at 0.6 arcsec is near the planet's contrast, the stated times are optimistic. Given that the authors have STIS and MUSE data, they could have plotted a disk surface-brightness profile or at least estimated the local background. Not doing so makes the feasibility claim unverifiable. Also, the reference stars eps Ori and gamma Ori are several magnitudes brighter than AB Aur; the paper doesn't discuss whether that affects the PSF reference calibration. That's a minor omission.\n\nOverall, as a proposal for a time allocation committee, it's good enough to request a few hours of CGI time. As a research paper, there is nothing here to referee. I wouldn't send it to a standard journal; it belongs in the mission's white-paper review. If it must go through peer review, a technical note format might work, but I wouldn't spend a referee's time on it in its current form.\n\nBottom line: useful for the Roman project, not a contribution to the literature on its own.","headline":"A sensible commission-time proposal, not a research result; the exposure estimates rest on an untested assumption about disk noise.","tokens_in":4945,"tokens_out":4028,"would_cite":false,"duration_ms":39745,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Even a modest coronagraph dark hole would give a clean optical view of the protoplanet AB Aur b.","keywords":["high-contrast imaging","coronagraphy","protoplanetary disk","directly imaged protoplanet","AB Aurigae","optical polarimetry","Roman Coronagraph","commissioning observations"],"falsifier":"Measure the on-sky dark-hole contrast at 575 nm and 825 nm on a star similar to AB Aurigae. If the achieved contrast at separations of roughly $0\\farcs2$–$0\\farcs5$ is worse than about $10^{-5}$, or if the disk light rather than the speckle floor dominates the noise, then the listed 0.4–0.7 hour integration times will not reach a $10\\sigma$ AB Aur b detection and the claimed optical improvement does not follow.","tokens_in":4037,"feed_emoji":"🔭","tokens_out":11102,"duration_ms":96513,"temperature":0.7,"pith_summary":"This white paper argues that AB Aurigae should be one of the first targets of the Roman Coronagraph instrument. The central claim is that a merely average dark hole—where the coronagraph suppresses starlight to about one part in 100,000 or one part in a million—would already provide a far cleaner optical view of the embedded protoplanet AB Aur b than current HST or ground-based observations. That view matters because AB Aur b's optical emission is unconstrained: it could be scattered starlight, accretion luminosity, or a young planet's own glow. With a few hours of total-intensity imaging and polarimetry in two bands, the authors say, the disk's dust properties and the planet's emission source could both be pinned down.","feed_headline":"One-in-a-million dark hole would reveal AB Aur b in visible light","feed_subtitle":"A few hours of optical imaging and polarimetry could pin down the protoplanet's light and the disk's dust.","key_machinery":"The central mechanism is the coronagraph dark hole: a region of the image where wavefront control suppresses the stellar point-spread function to $10^{-5}$–$10^{-6}$ of the star. The proposal uses two mask technologies—the hybrid Lyot coronagraph in narrow- and wide-field Band 1 modes at 575 nm, and the shaped-pupil coronagraph in wide-field Band 4 at 825 nm—together with angular differential imaging and reference-star differential imaging. The dark hole does the load-bearing work by making the planet and spiral arms visible in raw or lightly processed images, so that photometry and astrometry are not distorted by aggressive post-processing. Polarimetry adds the polarized fraction of that light as a second observable, which encodes grain properties and the balance between scattered and direct emission.","core_discovery":"The strongest claim is that AB Aur b and its disk are bright enough that the Roman Coronagraph need not reach its deepest designed contrast to make a decisive measurement. A dark hole contrast of $10^{-5}$ to $10^{-6}$ at 575 nm and 825 nm would yield a $10\\sigma$ detection in roughly 0.4–0.5 hours in Band 1 and 0.6–0.7 hours in Band 4, with the noise set by the speckle floor rather than by the bright disk. Combining narrow-field hybrid-Lyot and wide-field shaped-pupil modes would cover the full $360^\\circ$ disk from $0\\farcs15$ to $1\\farcs4$. The paper's conclusion is that such data would give AB Aur b astrometry and photometry free of the post-processing biases that plague existing HST data, and would, together with polarimetry, constrain the disk's dust and the nature of AB Aur b's optical emission.","pith_inferences":["Beyond the paper: the same two-band sequence would serve as an on-sky calibration of the dark hole's achieved contrast around a bright star, giving a reference for how deep later coronagraph surveys can realistically push.","Beyond the paper: other bright A-type stars with structured disks could be added to the same commissioning program, effectively turning a single demonstration into a small survey of embedded protoplanets.","Beyond the paper: if AB Aur b's optical flux is mostly scattered light, its measured polarization fraction at 575 nm would provide a short-wavelength handle on the dust grain population near the planet, complementing the near-infrared spectropolarimetry already obtained."],"forward_implications":["At $10^{-5}$–$10^{-6}$ contrast the planet should be detected at $10\\sigma$ in under an hour in each band, making the observation compatible with a commissioning or early-science slot.","Optical astrometry and photometry of AB Aur b would no longer depend on badly sampled HST PSFs or self-subtracting WFC3 processing, so the measured position and brightness could be compared directly with near-infrared and radio measurements.","The combined HLC and SPC coverage would image the full $0\\farcs15$–$1\\farcs4$ environment, capturing the spiral arms and the planet's location in the same data set.","Combined total-intensity and polarimetric data would place new constraints on the disk's dust properties and narrow the range of possible emission sources for AB Aur b's optical light.","A successful early image would publicly demonstrate the instrument's ability to image a bright, complex disk and a faint embedded companion."],"supporting_citations":[{"why":"Establishes AB Aur b as an embedded Jovian protoplanet at wide separation and provides the direct-imaging detection the proposed optical observations would improve upon.","marker":"[8]"},{"why":"Reports the H-alpha detection of AB Aur b with an inverse P Cygni profile, defining the emission-source puzzle that optical total-intensity and polarimetric data would address.","marker":"[14]"},{"why":"Supplies near-infrared scattered-light and spectropolarimetric measurements of the AB Aur disk that the proposed Band 1 and Band 4 optical data would extend.","marker":"[10]"},{"why":"Develops constrained reference-star differential imaging, the data-reduction approach the proposal relies on for high-fidelity disk imaging.","marker":"[9]"},{"why":"Presents SPHERE and ALMA evidence of ongoing planet formation in AB Aurigae, providing the spiral-arm and disk context for the target.","marker":"[7]"},{"why":"Imaged the inner gaseous spirals inside the dust cavity, linking the predicted planet position with the observed AB Aur b location.","marker":"[13]"}],"fun_headline_variants":["Roman's CGI: 10^-6 contrast reveals AB Aur b in under an hour","AB Aur b: Moderate dark hole yields bias-free astrometry","One hour of Roman imaging strips HST artifacts from AB Aur b","Roman Coronagraph: Mid-contrast mode sees AB Aur b and disk"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The plan assumes the Roman Coronagraph will actually reach a contrast of about one part in a million at 575 nm and 825 nm for a star of brightness $V = 7.05$, and that the speckle floor—not the bright surrounding disk—sets the noise.","fun_headline_variants_meta":{"raw":{"variants":["Roman's CGI: 10^-6 contrast reveals AB Aur b in under an hour","AB Aur b: Moderate dark hole yields bias-free astrometry","One hour of Roman imaging strips HST artifacts from AB Aur b","Roman Coronagraph: Mid-contrast mode sees AB Aur b and disk"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000288,"raw_usage":{"total_tokens":1686,"prompt_tokens":941,"completion_tokens":745,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":557,"completion_tokens_details":{"reasoning_tokens":665}},"tokens_in":557,"tokens_out":745,"duration_ms":7142,"temperature":1.0,"reasoning_tokens":665,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T16:35:18.350055+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the on-sky dark-hole contrast at 575 nm and 825 nm on a star similar to AB Aurigae. If the achieved contrast at separations of roughly $0\\farcs2$–$0\\farcs5$ is worse than about $10^{-5}$, or if the disk light rather than the speckle floor dominates the noise, then the listed 0.4–0.7 hour integration times will not reach a $10\\sigma$ AB Aur b detection and the claimed optical improvement does not follow.","supporting_citations":[{"cited_title":"Images of Embedded Jovian Planet Formation At A Wide Separation Around AB Aurigae","cited_arxiv_id":"2204.00633","evidence_quote":"Establishes AB Aur b as an embedded Jovian protoplanet at wide separation and provides the direct-imaging detection the proposed optical observations would improve upon."},{"cited_title":"SCExAO/CHARIS Near-infrared Scattered-light Imaging and Integral Field Spectropolarimetry of the AB Aurigae Protoplanetary System","cited_arxiv_id":null,"evidence_quote":"Supplies near-infrared scattered-light and spectropolarimetric measurements of the AB Aur disk that the proposed Band 1 and Band 4 optical data would extend."},{"cited_title":"Constrained Reference Star Differential Imaging: Enabling High-Fidelity Imagery of Highly Structured Circumstellar Disks","cited_arxiv_id":"2208.01606","evidence_quote":"Develops constrained reference-star differential imaging, the data-reduction approach the proposal relies on for high-fidelity disk imaging."},{"cited_title":"Planet Formation in AB Aurigae: Imaging of the inner gaseous Spirals observed inside the Dust Cavity","cited_arxiv_id":"1704.02699","evidence_quote":"Imaged the inner gaseous spirals inside the dust cavity, linking the predicted planet position with the observed AB Aur b location."}],"review_version":2}