{"id":"9dc95748-bfc0-493f-a119-0b63cd7fa025","arxiv_id":"2604.24991","paper_version":1,"verdict":"CONDITIONAL","confidence":"LOW","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"PDS70c exhibited a 170-228% radio flux rise over ~1 hour in 2017 ALMA data at 3.3-3.7 sigma significance, with lower but detectable scatter in 2023 data, consistent with accretion shock emission.","lead":"The paper detects hourly radio flux variability in the protoplanet candidate PDS70c using ALMA Band 7 observations and a time-differential photometry technique in the visibility domain. If real, the signal supports free-free emission from an accretion shock and requires a very low planet-to-environment mass ratio to avoid smoothing.","discovery_kind":"new_application","skeptic_critique":{"model":"grok-4.3","headline":"Synthetic controls may not capture all reduction artifacts from visibility alignment and time-differential subtraction on an embedded source.","rationale":"The reader's weakest assumption matches the load-bearing step: whether the differential photometry isolates true variability. The proposed injection test directly quantifies pipeline fidelity for an embedded source and would either confirm or refute the 3.7σ claim without requiring new data.","tokens_in":1851,"tokens_out":333,"duration_ms":54020,"concrete_test":"Reprocess the 2017 visibilities after injecting a constant-flux point source at the exact PDS70c location with the same uv-coverage and noise; run the identical alignment/self-cal/time-differential pipeline; if the recovered differential flux shows >50% of the reported 170% excursion or exceeds 2σ in any EB, the detection is not robust to the reduction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim requires that the reported 3.7σ rise (2017) and 2.6σ scatter (2023) are intrinsic point-source variability after subtracting the time-averaged extended disk. The pipeline applies visibility alignment plus self-calibration per EB, then differential photometry. While residuals are stated to be near-thermal and synthetic controls are used, an embedded source experiences different uv-sampling and subtraction residuals than an injected control at a clean location. Any epoch-dependent phase or amplitude errors that correlate with the planet's position would survive the differential step and appear as variability. The 170% amplitude change is large enough that even a 10-20% residual systematic could dominate the detection.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The manuscript analyzes ALMA Band 7 (~343 GHz) observations of PDS70c from 2017 (three ~1h execution blocks) and 2023 (~2h blocks), claiming detection of hourly-scale radio flux variability via time-differential photometry performed in the visibility domain after per-EB visibility alignment and self-calibration. It reports a 228%±69% (3.3σ) rise in imaged flux on 6 Dec 2017 and a 170%±46% (3.7σ) rise via differential photometry, with 2023 data showing constant flux within ~15% scatter but 2.6σ-significant intrinsic dispersion (49%±21%) when split into 20-min intervals; variability is validated against synthetic control sources and interpreted as consistent with HI free-free emission from an accretion shock on a circumplanetary disk surface, requiring a planet-to-environment mass ratio <10^{-4} to avoid radiative-diffusion smoothing.","tokens_in":2011,"tokens_out":778,"duration_ms":53748,"significance":"If the reported variability is confirmed as intrinsic point-source emission rather than residual artifact, the result would provide a rare observational link between radio flux changes on hourly timescales and accretion physics around embedded protoplanets, supporting free-free shock models and placing quantitative limits on mass ratios that could be tested with future multi-epoch monitoring. The pipeline's use of visibility-domain differential photometry and synthetic controls represents a reproducible approach to isolating variable embedded sources amid extended disk emission.","major_comments":[{"comment":"The validation with synthetic control point sources (described in the section on time-differential photometry and results for the 2017/2023 datasets) places controls at clean locations without extended disk emission. However, PDS70c is embedded within the disk, so its uv-sampling, self-calibration residuals, and time-differential subtraction behavior differ from injected controls; any epoch-dependent phase/amplitude errors correlated with the planet's position would survive the differential step and could produce apparent variability at the level of the claimed 170%±46% change. This directly undermines the central claim that the 3.7σ (2017) and 2.6σ (2023) signals are intrinsic after subtracting the time-averaged extended disk.","section":"section on synthetic control sources and time-differential photometry"},{"comment":"In the results for the 2023 dataset (20-min binning and chi-squared test), the reported 2.6σ significance and 49%±21% intrinsic dispersion rely on binned data and chi-squared statistics, but the manuscript provides limited explicit detail on full error propagation, covariance between bins, or data-exclusion criteria. Given that the 2017 detection rests on only three EBs with one showing the rise, small unaccounted systematics in the differential photometry could shift the significance below the threshold for a robust variability claim.","section":"results section on 2023 scatter and chi-squared test"}],"minor_comments":[{"comment":"The abstract states the 2017 rise as both 228%±69% (3.3σ) in images and 170%±46% (3.7σ) in differential photometry without clarifying whether these are independent measurements or derived from the same underlying data; a brief reconciliation in the results section would improve clarity.","section":"abstract and results"},{"comment":"Notation for flux density (F_B7) and error bars is used consistently but could be defined once in the methods with explicit reference to the visibility-domain measurement equation to aid readers unfamiliar with ALMA differential photometry.","section":"methods"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for their constructive and detailed review of our manuscript on the hourly radio variability of PDS70c. The comments raise important points about the validation of our time-differential photometry approach and the statistical details of our analysis. We address each major comment below and describe the revisions we will make to improve the robustness and clarity of the paper.","responses":[{"response":"We appreciate the referee's emphasis on the limitations of our control source placement. The synthetic controls were positioned in clean regions to establish a baseline for the differential photometry pipeline's performance in the absence of extended emission, confirming that the method itself does not generate artificial variability. We acknowledge that the embedded location of PDS70c within the disk means its uv-coverage and potential residual errors differ, and position-correlated systematics could in principle affect the differential measurements. To directly address this, the revised manuscript will include new simulations injecting synthetic point sources at the precise location of PDS70c, using a model of the surrounding disk emission. These tests will quantify any position-dependent effects on the recovered variability and allow us to re-evaluate the significance levels. We maintain that the per-EB self-calibration and time-differential subtraction mitigate much of the extended emission, but agree that these additional position-matched simulations are necessary to strengthen the validation.","revision_made":"yes","referee_comment":"The validation with synthetic control point sources (described in the section on time-differential photometry and results for the 2017/2023 datasets) places controls at clean locations without extended disk emission. However, PDS70c is embedded within the disk, so its uv-sampling, self-calibration residuals, and time-differential subtraction behavior differ from injected controls; any epoch-dependent phase/amplitude errors correlated with the planet's position would survive the differential step and could produce apparent variability at the level of the claimed 170%±46% change. This directly undermines the central claim that the 3.7σ (2017) and 2.6σ (2023) signals are intrinsic after subtracting the time-averaged extended disk."},{"response":"We agree that additional details on the statistical procedures are required for full transparency. In the revised manuscript, we will expand the methods and results sections to provide explicit descriptions of the error propagation for the binned time-differential photometry, including how per-bin uncertainties are derived from the visibility data and any treatment of covariance between adjacent time bins. We will also document the data-exclusion criteria applied and report the complete chi-squared statistics with degrees of freedom. For the 2017 dataset, we recognize the constraint of only three execution blocks and the fact that the rise appears in one; we will add a dedicated discussion of robustness, incorporating Monte Carlo simulations to assess how plausible systematics could affect the 3.7σ significance. These revisions will clarify the analysis without altering the core findings.","revision_made":"yes","referee_comment":"In the results for the 2023 dataset (20-min binning and chi-squared test), the reported 2.6σ significance and 49%±21% intrinsic dispersion rely on binned data and chi-squared statistics, but the manuscript provides limited explicit detail on full error propagation, covariance between bins, or data-exclusion criteria. Given that the 2017 detection rests on only three EBs with one showing the rise, small unaccounted systematics in the differential photometry could shift the significance below the threshold for a robust variability claim."}],"tokens_in":1744,"tokens_out":732,"duration_ms":47638,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The one thing to know is that this paper claims a 3.7-sigma detection of a 170% rise in PDS70c's 343 GHz flux over roughly an hour in the 2017 ALMA execution blocks, confirmed by time-differential photometry after visibility alignment and self-calibration, while the 2023 data stays flat on hour scales but shows 2.6-sigma scatter on 20-minute bins. If the signal is real, it links short-term radio changes to accretion shocks on a circumplanetary disk, with a planet-to-environment mass ratio below 10^-4 needed to keep thermal emission unsmoothed. This specific hourly variability measurement and the mass-ratio bound are new relative to earlier PDS70 ALMA papers. They do a solid job checking the result against synthetic control sources and running chi-squared tests on the binned fluxes, and the free-free interpretation follows standard physics without forcing new parameters. The soft spot is exactly the one in the stress-test note. An embedded source sits in different uv-sampling and experiences different subtraction residuals than a control placed in a clean field, so epoch-dependent phase or amplitude errors that correlate with the planet's location can survive the differential step and appear as variability. A 170% swing means even 10-20% systematics could dominate, and the abstract gives little on full covariance or data exclusion rules. This is for observers tracking protoplanet radio emission and modelers of late-stage accretion. A reader who cares about ALMA techniques for point sources inside disks would get practical value from the method. The claim is concrete and falsifiable enough to deserve referee time, even if the authors will need to add more residual tests and error details.","headline":"The paper reports a 3.7-sigma hourly radio flux rise for PDS70c in 2017 ALMA data via differential photometry, but the synthetic controls likely miss position-dependent reduction artifacts for this embedded source.","tokens_in":2522,"tokens_out":432,"would_cite":false,"duration_ms":51153,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"PDS70c exhibits hourly radio flux variability consistent with HI free-free emission from an accretion shock on a circumplanetary disk.","keywords":["PDS70c","protoplanet","radio variability","ALMA","accretion shock","circumplanetary disk","free-free emission","time-differential photometry"],"falsifier":"Additional ALMA observations of PDS70c at 343 GHz with comparable hourly cadence that show no significant flux changes above the thermal noise level would falsify the reported variability.","tokens_in":2775,"feed_emoji":"🪐","tokens_out":831,"duration_ms":62158,"temperature":0.7,"pith_summary":"The paper applies time-differential photometry to ALMA Band 7 data to isolate short-term changes in the point-source radio emission from PDS70c. It reports a 170 percent flux rise over roughly one hour in 2017 observations at 3.7 sigma significance, while 2023 data remains steady within 15 percent on two-hour scales but shows 49 percent dispersion on 20-minute intervals. This pattern of variability that appears on hourly timescales yet averages out over longer periods matches the expected behavior for free-free radiation from hydrogen at an accretion shock on the surface of a circumplanetary disk. The same observations impose a planet-to-environment mass ratio below 10 to the minus 4 if the emission were instead steady thermal radiation from the surrounding material. Such a link between radio variability and accretion physics provides a direct observational handle on the growth of embedded protoplanets.","feed_headline":"PDS70c radio flux rises 170% in one hour","feed_subtitle":"ALMA time-differential photometry detects hourly changes matching HI free-free emission at an accretion shock on a circumplanetary disk.","key_machinery":"Time-differential photometry performed in the visibility domain after visibility alignment and self-calibration to extract variable point-source flux from PDS70c while subtracting the time-averaged extended disk emission.","core_discovery":"Using visibility alignment, self-calibration, and time-differential photometry on ALMA 343 GHz observations, PDS70c was detected only in one 2017 execution block where its flux density rose by 170 percent with 3.7 sigma significance, while control sources remained stable. The 2023 dataset shows constant flux within a 15 percent scatter over two-hour blocks, yet splitting those blocks into 20-minute intervals yields an intrinsic dispersion of 49 percent significant at 2.6 sigma. The observed hourly variability, which is averaged out on daily timescales, is precisely the signature expected for HI free-free emission originating at an accretion shock on a circumplanetary disk surface; a planet- ","pith_inferences":["The same photometry technique could be applied to other ALMA datasets targeting embedded protoplanets to test whether hourly radio variability is a general signature of active accretion.","If the variability is confirmed, radio light curves may become a practical probe of the inner accretion flow around forming planets at spatial scales inaccessible to other wavelengths.","Independent mass estimates for the circumplanetary disk around PDS70c could directly test the reported mass-ratio upper limit."],"forward_implications":["The radio emission mechanism for PDS70c is HI free-free radiation from an accretion shock rather than steady thermal emission from the circumplanetary environment.","Accretion onto PDS70c proceeds with variations on hourly timescales that are smoothed when averaged over days.","A planet-to-environment mass ratio below 10 to the minus 4 is required if the emission were thermal, to prevent radiative diffusion from erasing the short-term signal.","Time-differential photometry in the visibility domain can separate variable embedded point sources from bright extended disk emission in ALMA data."],"fun_headline_variants":["ALMA finds 170% hourly radio increase from PDS70c","Time-differential photometry detects PDS70c hourly changes","343 GHz data show PDS70c radio flux rise of 170%","PDS70c flux varies 170% over one hour in 2017 ALMA"],"cache_read_input_tokens":64,"weakest_assumption_plain":"Residual calibration errors, extended disk emission, or self-calibration artifacts have been fully removed by visibility alignment and time-differential subtraction, leaving only genuine variability from the point source PDS70c.","fun_headline_variants_meta":{"raw":{"variants":["ALMA finds 170% hourly radio increase from PDS70c","Time-differential photometry detects PDS70c hourly changes","343 GHz data show PDS70c radio flux rise of 170%","PDS70c flux varies 170% over one hour in 2017 ALMA"]},"model":"grok-4.3","cost_usd":0.011748,"raw_usage":{"total_tokens":5143,"prompt_tokens":834,"num_sources_used":0,"completion_tokens":76,"cost_in_usd_ticks":117478000,"prompt_tokens_details":{"text_tokens":834,"audio_tokens":0,"image_tokens":0,"cached_tokens":64},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":4233,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":834,"tokens_out":76,"duration_ms":67083,"temperature":1.0,"reasoning_tokens":4233,"cache_read_input_tokens":64,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-07T17:46:27.654952+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Additional ALMA observations of PDS70c at 343 GHz with comparable hourly cadence that show no significant flux changes above the thermal noise level would falsify the reported variability.","supporting_citations":[],"review_version":1}