{"id":"b9922cc6-1151-40cd-85fc-98660dc37d4a","arxiv_id":"1908.00973","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Single-session geodetic VLBI observations of sources near the Sun constrain the PPN parameter γ with a formal error of about one part in ten thousand, matching or beating global solutions.","lead":"Two dedicated VLBI sessions tracked radio sources just 1 to 3 degrees from the Sun and fitted the relativistic deflection parameter γ with a formal precision near 1 part in 10,000. This is the first time a single geodetic VLBI session has matched the formal accuracy of global multi-decade solutions, potentially opening a path toward competitive ground-based tests of general relativity.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Coronal plasma cancellation is assumed, not tested; the quoted 1e-4 formal error is not established as a realistic uncertainty.","rationale":"The paper is an honest report of a small, purpose-built experiment. It provides independent analyses with two software packages, gives a complete Table 1, and explicitly acknowledges limitations, including the 2.7-sigma outlier and the quiet-region geometry. I agree with the reader that the central issue is not the arithmetic of the formal error but the interpretation of that error as a realistic uncertainty. The abstract is careful to say 'formal accuracy,' which limits the claim; nevertheless, the conclusion that a factor-of-ten improvement in the uncertainty on gamma is expected, and the comparison with global solutions, implicitly rely on the cancellation assumption. A split-session stability test is a direct and inexpensive check. The 2.7-sigma Calc/Solve result for 0229+131 is concerning but not by itself decisive, since many estimates are reported and multiple comparisons inflate the chance of a 2-3-sigma outlier. The source-structure degradation in AOV022 further shows that the method is sensitive to source properties. Overall, the reader's conditional verdict is appropriate; my stress test does not move it.","tokens_in":5565,"tokens_out":5594,"duration_ms":61463,"concrete_test":"Split the AUA020 data into four non-overlapping sub-sessions of about four hours each and estimate gamma independently with the same OCCAM setup used for the 0.94e-4 solution, including Sejong downweighted. If the weighted scatter of the four sub-session estimates is consistent with the full-session formal error, the cancellation assumption survives; if the scatter exceeds the formal error by a factor of two or more, the quoted precision is optimistic and the conditional verdict should stand.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The abstract's central quantitative claim is that a single 17-hour VLBI session reaches a formal gamma precision below 1e-4, better than global solutions. Table 1 supports that as a formal covariance statement. The load-bearing step is in Section 4, where small-scale coronal density fluctuations are assumed to 'cancel out over the period of observations' and the two ray paths 'happened to be in quiet regions.' This assumption converts a formal error into a scientific uncertainty and is not tested. The same section reports an unexplained 2.7-sigma Calc/Solve value for 0229+131, showing that at least one source has a significant deviation not captured by the quoted sigma. With only two sources and one session, a session-length correlated plasma contribution common to both sources would not average down in the assumed way, and the quiet-region statement is post hoc. The comparison with global solutions is also indirect, since no same-software global solution is run here; the 'better than global' wording relies on older, differently modeled published estimates. None of this contradicts the formal-error bookkeeping, but it means the headline precision is not yet demonstrated to be a true measurement uncertainty.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports two dedicated geodetic VLBI sessions, AUA020 (May 2017) and AOV022 (May 2018), that observed two ICRF radio sources at solar elongations between 1 and 3 degrees. Using two independent analysis packages (OCCAM and Calc/Solve), the authors estimate the PPN parameter gamma from single-session group delays. For AUA020, formal errors reach 0.94e-4 (OCCAM with Sejong downweighted) and 1.10e-4 (Calc/Solve), which the paper presents as comparable to or better than formal errors from global VLBI solutions. AOV022 produced degraded statistics attributed to source structure. The authors conclude that small-scale coronal and ionospheric perturbations cancel over the 17-hour session and that future experiments near solar minimum could challenge the Cassini bound on gamma.","tokens_in":5775,"tokens_out":4462,"duration_ms":44666,"significance":"If the formal errors are accepted as realistic uncertainties, this is an important proof-of-concept: a single 17-hour VLBI session near the Sun constrains gamma at the 1e-4 level, roughly two orders of magnitude better than typical per-session estimates and competitive with global VLBI solutions. The paper's strengths include the dual-software analysis with two independent packages, explicit reporting of formal errors and postfit rms in Table 1, an honest acknowledgment of the unexplained 2.7-sigma Calc/Solve result for 0229+131, and a clear statement of the degraded AOV022 statistics. However, the headline precision is a formal covariance statement, and the manuscript does not yet establish it as a realistic total uncertainty because the cancellation of coronal plasma perturbations is assumed rather than tested. The central values are consistent with general relativity, but the paper is best read as a demonstration of attainable formal precision rather than as a completed measurement of gamma.","major_comments":[{"comment":"The paper's central claim of ~1e-4 precision depends on the assumption that small-scale coronal perturbations 'cancel out over the period of observations (17 hours with observations angularly close to the Sun).' This is the step that converts a formal covariance error into a scientific uncertainty, and it is not tested. The very next sentence concedes that the ray paths 'happened to be in quiet regions,' a post hoc justification. With only two sources and a single session, a slowly varying or common-mode coronal contribution would not cancel in the assumed way, and the two targets are close enough in the sky that their ray paths may share structure. Please provide a quantitative test, such as estimating an additional coronal electron-density parameter, correlating results with a solar-activity index, or adding a conservative systematic term to the error budget, before the headline precision is presented as a measurement uncertainty.","section":"Section 4"},{"comment":"The Calc/Solve solution using only 0229+131 gives gamma-1 = (-6.84 +/- 2.53) x 10^-4, a 2.7-sigma deviation, while the OCCAM solution for the same source with Sejong downweighted gives (0.32 +/- 2.83) x 10^-4. The text states the origin is unclear and suggests differences in estimation method or troposphere/clock handling. This unresolved discrepancy is evidence that the quoted formal errors may not capture some systematic effect, and it sits in the same paragraph as the statement 'No large systematics are detected.' The anomaly should be modeled or propagated into the uncertainty, or at minimum explicitly listed as a limitation on the sub-10^-4 precision claim.","section":"Table 1 and Section 3"}],"minor_comments":[{"comment":"There are several typographical errors: 'publsihed' in Section 2, 'responce' in Section 4, 'Auckand' in the affiliation block, and 'abd' in the author list. These should be corrected.","section":"General"},{"comment":"The AOV022 results are described only qualitatively as '2-3 times worse' than AUA020. A table or at least quoted values for the formal errors and postfit rms would allow the reader to assess the degradation quantitatively.","section":"Section 3"},{"comment":"In the OCCAM rows, the entries labeled 'Both' are identical to 'All scans' for both station-weighting cases. Please clarify whether 'Both' includes only the two target sources and why it coincides exactly with the all-scan solution.","section":"Table 1"},{"comment":"The comparison with 'global solutions' should state the exact published value of the global-solution formal error being cited; the current text references Lambert and Le Poncin-Lafitte (2009, 2011) but does not reproduce their numerical uncertainties, making the abstract's 'better than' claim hard to verify.","section":"Section 3"},{"comment":"The sentence 'It was a general misconception in the past that the effect of the plasma of the solar corona completely disturbs the interferometric responce for light rays passing within several degrees from the Sun' would benefit from a citation or a more hedged phrasing, as it appears to summarize a broad historical claim without a specific reference.","section":"Section 4"}],"recommendation":"major_revision","confidential_remarks":"The paper is a measurement report appropriate for a specialized VLBI or astrometry journal. The main gap is the formal-to-total uncertainty conversion in Section 4; if the authors can supply a systematic error budget or reframe the abstract strictly as a formal-precision result, the paper would be acceptable. I see no circularity in the gamma estimation: the target value is not defined by the scheduling paper, and the analysis uses standard VLBI delay models."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two new VLBI sessions track sources at 1.1–2.6 degrees from the Sun, and the single-session fits to PPN gamma come out with formal errors around 1e-4, matching or beating older global solutions. The cross-check between OCCAM and Calc/Solve is a genuine robustness check. The paper is honest about its own limitations: a 2.7-sigma anomaly in one Calc/Solve solution, degraded statistics in AOV022, and no quantified systematic error budget.\n\nWhat is actually new here is the data: AUA020 and AOV022, with the AUA020 results in Table 1. The design, scheduling, and analysis strategy are carried over from the group's own Titov et al. (2018), so the novelty is incremental but real. The formal claim in the abstract is supported—the best single-session sigma is 0.94e-4.\n\nThe load-bearing soft spot is Section 4. The paper assumes that small-scale coronal plasma perturbations cancel over the 17-hour session because the ray paths 'happened to be in quiet regions.' That is exactly the assumption that converts a formal error into a scientific uncertainty, and it is not tested. With only two sources, a session-length, source-common plasma contribution would not average down. The 2.7-sigma deviation in 0229+131 shows at least one effect not captured by the quoted sigma. Also, the comparison with global solutions is indirect: no same-software global solution is run here; 'better than global' rests on older, differently modeled published estimates. None of this contradicts the formal-error bookkeeping, but it means the headline precision is not yet demonstrated to be a true measurement uncertainty.\n\nWho is this for? Anyone working on GR tests with VLBI, and to a lesser extent the astrometry community. It deserves a serious referee, because if the formal precision holds under systematics, this is a path toward challenging Cassini. The referee should ask for a systematics discussion and a softer abstract (say 'formal' more prominently), and ideally a check of the cancellation assumption, e.g., a time-split of the session or a plasma model. I'd bring it to a reading group, but I wouldn't yet cite the gamma value in my own work.","headline":"Plausible single-session VLBI gamma measurement with formal errors near 1e-4, but the coronal cancellation assumption is the untested load-bearing part.","tokens_in":6382,"tokens_out":2840,"would_cite":false,"duration_ms":27953,"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":"A dedicated VLBI session observing radio sources 1.15°–2.6° from the Sun measures the PPN parameter $\\gamma$ with a formal uncertainty below $1\\times10^{-4}$, about 0.01 percent, matching or beating the formal accuracy of global…","keywords":["VLBI","general relativity","PPN parameter gamma","solar gravitational deflection","Shapiro delay","solar corona","geodetic VLBI","radio astrometry"],"falsifier":"Split the AUA020 session into independent time blocks, or rerun the same two-source schedule when the solar corona is active rather than quiet, and compare the fitted $\\gamma$ values; if the estimates differ by substantially more than their formal errors, the cancellation assumption fails and the stated $\\sim 10^{-4}$ accuracy is not the real uncertainty.","tokens_in":5392,"feed_emoji":"☀️","tokens_out":13812,"duration_ms":120953,"temperature":0.7,"pith_summary":"Geodetic very long baseline interferometry (VLBI) can test general relativity by timing how much radio waves slow down as they pass the Sun's gravitational field. This paper reports that two specially scheduled 17-hour sessions, tracking reference radio sources at angular distances of 1.15° to 2.6° from the Sun, determine the PPN parameter $\\gamma$ — which equals 1 in general relativity — with a single-session formal uncertainty near $1\\times10^{-4}$, below 0.01 percent. That precision matches or beats the formal error of global VLBI solutions built from all available observations at arbitrary elongation, and both independent analyses give values consistent with $\\gamma=1$. The result matters because it shows that solar-corona plasma does not necessarily destroy close-to-the-Sun astrometry: standard dual-frequency calibration removes the quiet corona's dispersive delay, so dedicated close-approach sessions are a viable route toward challenging the Cassini bound of about $2\\times10^{-5}$.","feed_headline":"One night of radio astronomy near the Sun tests relativity to 0.01%","feed_subtitle":"Tracking two sources just 1–3 degrees from the Sun matched decades of global data in a single 17-hour session.","key_machinery":"The load-bearing object is the VLBI group-delay observable with a PPN-delay model: the extra travel time of a radio signal passing near the Sun is proportional to $(1+\\gamma)/2$ times the Shapiro delay, so fitting $\\gamma$ to delays recorded while the source is close to the limb measures the deflection. The analysis uses dual-frequency S/X-band calibration to remove the dispersive delay of the solar corona and ionosphere, estimates troposphere and clock parameters as nuisances, and restricts attention to compact sources whose structure delay is below about 2 ps. The 1.15°–2.6° geometry makes the gravitational delay large, while the 17-hour session and quiet-corona ray paths keep random coronal scatter small enough for the formal error to reach $\\sim 10^{-4}$.","core_discovery":"On the paper's own terms, the central claim is that a single geodetic VLBI session can measure gravitational light deflection near the Sun with better formal accuracy than the global solution using all available VLBI data. Tracking the compact ICRF sources 0229+131 and 0235+164 at elongations of 1.15°–2.6° and fitting the PPN parameter $\\gamma$ to the 17 hours of group delays yields $\\sigma_\\gamma$ between $0.9\\times10^{-4}$ and $4\\times10^{-4}$ depending on the source subset; the best solution with one station downweighted gives $\\gamma-1 = (0.91 \\pm 0.94)\\times10^{-4}$, and the independent software solution gives $(-0.22 \\pm 1.10)\\times10^{-4}$. Both are consistent with $\\gamma=1$. The paper interprets these results as demonstrating that coronal plasma perturbations are random rather than systematic for ray paths in quiet coronal regions, so they cancel over the session after dual-frequency calibration. A second session, AOV022, had statistics two to three times worse, attributed to source structure rather than the corona.","pith_inferences":["A direct extension of the paper's cancellation assumption: split the AUA020 session into two or three independent time blocks and compare the fitted $\\gamma$ values; if they scatter by more than the formal errors, the reported $\\sim 10^{-4}$ accuracy is optimistic.","The gravitational deflection grows toward the solar limb, so scheduling even closer than the current 1.15° minimum (possible near solar minimum with strong, compact sources) could improve precision faster than adding more sessions, an inference the paper leaves implicit.","The same close-approach geometry could in principle be used to constrain other solar-system relativistic effects, such as the solar quadrupole moment or the PPN parameter $\\beta$, because their delay signatures have different dependence on impact parameter; the paper does not discuss this.","The 2.7$\\sigma$ deviation in one single-source solution hints at a source- or model-dependent systematic, so future sessions should include a third compact source at similar elongation to distinguish a modeling artifact from real physics."],"forward_implications":["Single dedicated sessions near the Sun can constrain $\\gamma$ as tightly as multi-decade global VLBI solutions, making general-relativity tests schedulable instead of requiring years of accumulated data.","The closer source (0235+164) gives roughly a factor of two smaller formal error than the farther source (0229+131), confirming that low-elongation observations carry most of the sensitivity.","Standard dual-frequency calibration can handle the quiet solar corona, so future sessions near solar minimum can push to even smaller elongations and higher signal-to-noise.","Source structure, not the corona, limited the second session, so compactness screening of target sources is a prerequisite for further improvement.","Accumulating several such sessions should move the VLBI $\\gamma$ uncertainty toward the Cassini limit of about $2\\times10^{-5}$."],"supporting_citations":[{"why":"introduces the gravitational delay of radio waves, the effect the paper uses to estimate gamma.","marker":"Shapiro, 1964"},{"why":"the Cassini radio-science result whose gamma limit is the benchmark this method aims to approach.","marker":"Bertotti et al (2003)"},{"why":"global VLBI gamma solution whose formal accuracy is compared with the single-session result.","marker":"Lambert and Le Poncin-Lafitte, 2009"},{"why":"updated global VLBI gamma estimate used as the precision benchmark in the comparison.","marker":"Lambert and Le Poncin-Lafitte, 2011"},{"why":"describes the scheduling and analysis design of the dedicated AUA020 experiment.","marker":"Titov et al, 2018"},{"why":"previous VLBI measurements of gravitational bending that establish the technique's baseline accuracy.","marker":"Fomalont et al, 2009"},{"why":"source structure indices used to ensure structure delay below 2 ps.","marker":"Fey and Charlot, 1997"},{"why":"documents one of the two independent geodetic VLBI analysis packages.","marker":"Ma et al., 1986"},{"why":"documents the other independent analysis package.","marker":"Titov et al., 2004"},{"why":"reports degradation of VLBI gamma estimates with elongation, the systematic this close-approach design avoids.","marker":"Shapiro et al, 2004"}],"fun_headline_variants":["VLBI night near Sun beats global data in relativity test","Single VLBI session measures light deflection near Sun to 0.01%","Near-sun VLBI test: one session beats all-old data for gamma","VLBI near Sun: relativity precision 0.01% in one night","Gravity's bend on light: single VLBI session hits 0.01%"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing assumption is that coronal and small-scale plasma perturbations produce only random scatter that cancels over the 17-hour session, so the formal $\\sim 1\\times10^{-4}$ uncertainty is the real accuracy; the paper states this explicitly and notes it held because the ray paths were in quiet coronal regions.","fun_headline_variants_meta":{"raw":{"variants":["VLBI night near Sun beats global data in relativity test","Single VLBI session measures light deflection near Sun to 0.01%","Near-sun VLBI test: one session beats all-old data for gamma","VLBI near Sun: relativity precision 0.01% in one night","Gravity's bend on light: single VLBI session hits 0.01%"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000544,"raw_usage":{"total_tokens":2595,"prompt_tokens":927,"completion_tokens":1668,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":543,"completion_tokens_details":{"reasoning_tokens":1569}},"tokens_in":543,"tokens_out":1668,"duration_ms":11432,"temperature":1.0,"reasoning_tokens":1569,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:38:23.520052+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Split the AUA020 session into independent time blocks, or rerun the same two-source schedule when the solar corona is active rather than quiet, and compare the fitted $\\gamma$ values; if the estimates differ by substantially more than their formal errors, the cancellation assumption fails and the stated $\\sim 10^{-4}$ accuracy is not the real uncertainty.","supporting_citations":[{"cited_title":"Phys Rev Lett 13(26):789--791","cited_arxiv_id":null,"evidence_quote":"introduces the gravitational delay of radio waves, the effect the paper uses to estimate gamma."},{"cited_title":"Nature 425:374--376","cited_arxiv_id":null,"evidence_quote":"the Cassini radio-science result whose gamma limit is the benchmark this method aims to approach."},{"cited_title":"Astronomy & Astrophysics 499:331--335","cited_arxiv_id":null,"evidence_quote":"global VLBI gamma solution whose formal accuracy is compared with the single-session result."},{"cited_title":"Astronomy & Astrophysics 529:A70","cited_arxiv_id":null,"evidence_quote":"updated global VLBI gamma estimate used as the precision benchmark in the comparison."},{"cited_title":"Astrophysical Journal 699:1395--1402","cited_arxiv_id":null,"evidence_quote":"previous VLBI measurements of gravitational bending that establish the technique's baseline accuracy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"source structure indices used to ensure structure delay below 2 ps."},{"cited_title":"Physical Review Letters 92(12):121101","cited_arxiv_id":null,"evidence_quote":"reports degradation of VLBI gamma estimates with elongation, the systematic this close-approach design avoids."}],"review_version":1}