{"id":"7fc39e75-32f5-4ec3-9200-d4df5e086a60","arxiv_id":"1909.00785","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"In the RadioAstron AGN survey, about two-thirds of observed bright AGN are detected on space-ground baselines, indicating ubiquitous ultra-compact, tens to hundreds of microarcsecond core structures.","lead":"The RadioAstron space telescope detected extremely compact, bright structures in about two-thirds of a complete sample of active galactic nuclei. This is the first statistical demonstration that such ultra-compact cores are common, not rare, and it provides new constraints on extreme AGN jet physics.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The two-thirds detection fraction is reported for the observed C-band subsample, not the full 163-source sample; the abstract drops the 'observed' qualifier and no confidence interval or missing-source bound is provided.","rationale":"The reader and I converge on the same soft spot: the headline fraction may not represent the full 163-source sample. I would sharpen it slightly: the abstract omits even the observed qualifier, and the 95/147 C-band count is the only number in the paper that produces two-thirds, so the central claim's precision depends on a source-level union calculation that is not reported. This is a load-bearing concern because the paper's novelty is the prevalence of ultra-compact cores, not the existence of a handful of detections; the latter is already established by the listed extreme-baseline sources. I do not think the concern rises to rejection: the missing 10% is modest, the physical detections are real, and the sample is already selected for compact radio-strong AGN. But it justifies a conditional verdict: the two-thirds number should be qualified, and the missing-source and scheduling biases should be bounded. I therefore leave the reader's CONDITIONAL verdict unchanged.","tokens_in":10575,"tokens_out":7759,"duration_ms":91633,"concrete_test":"Build a per-source table from the survey pipeline listing, for every one of the 163 sample sources, whether it was observed in C, L, and/or K and whether a space-baseline fringe was found in any band. From this table compute the union detection fraction with the ~16 never-observed sources first coded as nondetections and then coded as detections (upper bound), with a binomial confidence interval for each. Also compare the 8 GHz correlated flux and RFC compactness distributions of the observed and unobserved subsets with a two-sample test. If the union fraction remains within 0.60–0.70 and the subsets are indistinguishable, the abstract's wording is acceptable; otherwise the paper should report the C-band detected fraction with the 'observed' qualifier and quantify the missing-source uncertainty.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central, load-bearing statistic is the two-thirds detection fraction. In Section 3 it is explicitly two thirds of the observed complete sample, and the only source-level count in Fig. 2 that matches that value is the C-band result: 95 detections out of 147 sources (64.6%). Section 2 reports that about 10% of the complete sample, mostly low-declination targets, had not been observed by June 2016, and the K-band sample is much smaller (108) while the L-band sample is slightly smaller (145). The abstract, however, says 'Two-thirds of them' are detected, omitting the 'observed' and C-band qualifications. Scheduling was driven by declination, tracking-station visibility, and ground-telescope uv coverage, so the unobserved 10% cannot be assumed exchangeable with the observed subset without a check. No confidence interval, no best/worst-case treatment of the missing sources, and no correction for the acknowledged under-scheduling of short baselines is provided. The paper therefore robustly establishes that many radio-strong AGN contain ultra-compact cores, but the population prevalence implied by 'two-thirds of the complete sample' is not yet supported at that precision. The physical conclusions about the detected sources themselves do not depend on this concern.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper presents detection statistics for the RadioAstron AGN survey, a space VLBI Key Science Program targeting a complete VLBI-flux-density-limited sample of 163 radio-strong AGN at 18, 6, and 1.3 cm. Using a PFD threshold of less than 0.01% and an SNR-to-PFD calibration derived from a fit to the low-SNR part of the fringe-SNR distribution, the authors report detection fractions as a function of projected ground-space baseline. They find that about two-thirds of the observed complete sample are detected on space VLBI baselines, with many detected sources showing brightness temperatures above the Compton catastrophe limit, and they note a possible excess of 18 cm detections at the longest baselines, attributed to interstellar scattering substructure.","tokens_in":10815,"tokens_out":4994,"duration_ms":59497,"significance":"If the headline result is stated accurately, this is an important empirical contribution: it demonstrates on a well-defined flux-limited sample that a large fraction of AGN cores contain ultra-compact, very bright structures on space-VLBI scales, directly constraining the prevalence of brightness temperatures in excess of the Compton catastrophe and equipartition limits. The paper's strengths include the sample definition, the multi-band comparison, the explicit false-detection threshold, and the transparent description of the SNR-PFD calibration procedure. The detection counts themselves are straightforward and the qualitative conclusion that many AGN cores are detected on space baselines is robust. However, the precise population-level claim of 'two-thirds' needs to be qualified and quantified before the result can be taken at face value, and the absence of confidence intervals weakens the baseline-dependent comparisons.","major_comments":[{"comment":"The headline 'two-thirds' claim is not stated precisely enough. The abstract says 'Two-thirds of them are indeed detected' after referring to the complete sample of 163 sources, and the Summary repeats 'Two thirds of the targets.' But the body of Section 3 reports 'about two thirds of the observed complete sample,' and the only number matching this in Figure 2 is the C-band result: 95 detections out of 147 sources (64.6%). The K-band and L-band fractions are 30/108 (27.8%) and 84/145 (57.9%), respectively. This is a load-bearing distinction because the paper's central claim is about the prevalence of ultra-compact cores in the complete sample. The abstract and Summary should explicitly state that the two-thirds figure refers to the C-band observed subset, and the corresponding K- and L-band fractions should be reported in the same places.","section":"Abstract and Section 3"},{"comment":"The manuscript acknowledges that about 10% of the complete sample had not been observed by June 2016 and that these are predominantly low-declination sources, but it does not account for this in the detection statistics. Because scheduling was driven by declination, tracking-station visibility, and ground-telescope constraints, the unobserved sources cannot be assumed exchangeable with the observed subset without justification. The paper should provide a sensitivity analysis or at least a worst-case bound: for the C-band subset, if all 16 unobserved sources are assumed to be non-detections, the fraction drops from 95/147 (64.6%) to 95/163 (58.3%). This bound does not overturn the qualitative conclusion that a large fraction of cores are detected, but it means the precise 'two-thirds' wording is not supported for the full sample, and the manuscript should either add such a bound or restrict all prevalence statements to the observed sample.","section":"Section 2 and Section 3"},{"comment":"Figure 2 displays detection fractions versus projected baseline without confidence intervals, per-bin source counts, or a significance test for the differences between bands. This is particularly important for the two secondary claims in this section: the apparent drop at short baselines (which the text attributes to an observational bias) and the 'possible excess of fractional detections at the longest RadioAstron projected baselines at 18 cm' relative to 6 cm, which is then interpreted as evidence for scattering substructure. Without binomial confidence intervals or a statistical test, the reader cannot judge whether the reported excess is significant or merely noise. At minimum, the figure should include error bars or shaded confidence bands, and the claim about the 18 cm excess should be accompanied by a significance estimate.","section":"Section 3, Figure 2"}],"minor_comments":[{"comment":"The text says 'About 10% of the complete sample had not been observed by June 2016,' but the reader is not told the exact per-band numbers of observed sources or whether the same 163 sources are included in each band's denominator. Please clarify that the denominators in Figure 2 (108, 147, 145) are the observed subsets in each band and give the overlap between bands.","section":"Section 2"},{"comment":"The sentence 'RadioAstron has delivered detections in just over one third of the observing segments' is ambiguous because the paper elsewhere discusses detected sources, not segments. Please define what an 'observing segment' is and state how this segment-level detection rate relates to the source-level detection fractions in Figure 2.","section":"Section 3"},{"comment":"The description of the SNR-PFD calibration says the theoretical distribution is fitted to the low-SNR peak and that the parameters depend on the number of spectral channels, correlator integration time, and scan length. It would be helpful to state how many distinct parameter sets were used and how the resulting PFD-SNR mapping varies across them, since only one example is shown.","section":"Section 3, Figure 4"},{"comment":"The phrase 'the apparent drop in the detection fraction at 0 to 2 Earth Diameters ... can be observed in the K- and L-band histograms' is presented as an observational bias, but no quantitative comparison is made with the C-band first bin, which is called 'unbiased.' A short quantitative justification, even a statement of the number of scheduled scans in that bin, would strengthen the claim.","section":"Section 3"},{"comment":"The Summary states 'Two thirds of the targets have delivered significant interferometric fringes at space VLBI baselines,' but the body text consistently says 'the observed complete sample.' Please align the Summary with the qualified wording used in Section 3.","section":"Summary"}],"recommendation":"major_revision","confidential_remarks":"To the editor: The manuscript reports a straightforward and valuable empirical data set, and the core detection counts appear sound. The main issue is that the abstract and summary state the 'two-thirds' result for the complete sample, while the body restricts it to the C-band observed subset, and the paper does not provide confidence intervals or a missing-source sensitivity bound. These issues are addressable by revision and do not require redoing the observations. I recommend major revision rather than rejection."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a real survey result and the two-thirds detection fraction is real for the observed C-band subsample. But the abstract overstates it, and the paper doesn't give error bars or bound the unobserved 10%. None of that kills the work; it needs tightening, not rethinking.\n\nWhat's new: first statistical detection-rate curves versus projected baseline for a complete AGN sample on space-ground baselines out to 28 Earth diameters. The sample is clearly defined (163 sources, 8 GHz correlated flux density > 600 mJy at >200 Mλ), and the detection criterion (PFD < 0.01%) is quantified via a fit to the low-SNR part of the fringe SNR distribution. That calibration is standard and doesn't force the detection fraction. The short-baseline drop is explicitly identified as a scheduling bias, which is honest.\n\nThe main result: 95 of 147 observed C-band sources are detected on space baselines (64.6%), 84 of 145 at L-band (57.9%), 30 of 108 at K-band (27.8%). So 'two-thirds' is really the C-band number. Section 3 says 'observed complete sample,' which is accurate; the abstract says 'Two-thirds of them,' which is not. That's a real imprecision, and the stress-test is right to flag it. The unobserved ~10% are low-declination sources, likely harder to schedule; no attempt is made to bound the detection fraction under worst-case assumptions. That's a minor omission for a survey paper, not a fatal one, because even the L-band fraction is high enough to support the qualitative conclusion about ultra-compact cores.\n\nOther soft spots: Figure 2 has no confidence intervals (easy to compute, but the paper should show them or at least state the counts in bins). The K-band sample is small and sensitivity-limited, so the low K-band detection fraction is not surprising. The claimed 18 cm excess at the longest baselines comes from a handful of sources and is speculative; it's flagged as such, fair enough.\n\nThe paper ships no data or code, but this is a survey statistics report, not an analysis paper; the counts are directly readable from the text. Citation pattern looks appropriate. It's a useful reference for anyone working on space VLBI, AGN brightness temperatures, or the Compton catastrophe.\n\nRecommendation: send it to peer review. A referee should ask for a corrected abstract, error bars on the detection fractions, and a sentence acknowledging the unobserved 10% limitation. After that, it's a solid contribution to Advances in Space Research or similar.","headline":"Solid survey detection statistics let down by an overbroad abstract 'two-thirds' claim and missing uncertainty bars, but the core result stands.","tokens_in":11744,"tokens_out":2852,"would_cite":true,"duration_ms":28049,"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":"A space-VLBI survey finds that two-thirds of bright AGN contain ultra-compact cores with brightness temperatures exceeding the inverse-Compton limit.","keywords":["active galactic nuclei","quasars","galaxies: jets","radio continuum: galaxies","space VLBI","brightness temperature","interstellar scattering"],"falsifier":"Re-observing the roughly 16 unobserved low-declination targets with a southern tracking station and uniform short-baseline scheduling would settle the matter: if the detection fraction on baselines beyond a few Earth diameters among these sources falls below about one-third, the claim that two-thirds of the complete sample contains ultra-compact cores would not survive.","tokens_in":10402,"feed_emoji":"📡","tokens_out":7035,"duration_ms":65901,"temperature":0.7,"pith_summary":"This paper reports detection statistics from the RadioAstron space-VLBI survey of a complete, flux-density-limited sample of 163 radio-strong AGN at 18, 6, and 1.3 cm. The central result is that about two-thirds of the observed targets produce significant interferometric fringes on ground-space baselines, implying that many AGN cores contain extremely compact, tens-to-hundreds-of-microarcsecond structures of very bright synchrotron emission. If this holds for the whole sample, the measured brightness temperatures exceed the inverse-Compton (Compton catastrophe) limit and sit far above equipartition for most detected sources, which directly challenges standard jet emission models and bears on the 'Doppler factor crisis' from gamma-ray observations. The paper also presents detection fraction versus projected baseline and attributes an excess of 18-cm detections at the longest baselines to interstellar scattering substructure.","feed_headline":"Two-thirds of AGN cores show ultra-compact radio structure","feed_subtitle":"Space-VLBI survey resolves microarcsecond-scale knots with brightness temperatures above Compton limits.","key_machinery":"The central instrument is the ground-space interferometer RadioAstron, whose baselines extend to 28 Earth diameters and whose detection sensitivity reaches about 6 mJy at 18 and 6 cm and 60 mJy at 1.3 cm with the largest ground telescopes. The load-bearing observable is a fringe detection: a source counts as detected when the probability of a false detection is below 0.01%, with the signal-to-noise-to-probability calibration derived by fitting a theoretical noise distribution to the low-SNR part of the empirical fringe-SNR distribution. Detection fraction versus projected baseline then maps directly to the angular-size and brightness-temperature distribution of AGN cores, because the maximum measurable brightness temperature depends on baseline length and fringe-visibility accuracy rather than on wavelength.","core_discovery":"The paper establishes that two-thirds of the observed complete sample are detected on space VLBI baselines: 30 of 108 sources at 1.3 cm, 95 of 147 at 6 cm, and 84 of 145 at 18 cm, with detection defined by a probability of false detection below 0.01%. The detected sources therefore contain structures with angular sizes of tens to hundreds of microarcseconds and brightness temperatures significantly exceeding the Compton catastrophe limit, with most far above the equipartition value. The fractional detection as a function of projected baseline resembles earlier 5-GHz VSOP and 2-cm VLBA survey curves, consistent with core-jet structure probed at smaller scales. An excess of 18-cm detections at baselines of about 25 Earth diameters or longer is interpreted as evidence of scattering substructure, while interstellar scattering is only weak at 6 cm and absent at 1.3 cm for the typical sightlines sampled.","pith_inferences":["My inference: if the two-thirds detection fraction extends to the unobserved low-declination sources, the number of AGN cores with brightness temperatures above $10^{13}$ K is large enough that the apparent 'Doppler factor crisis' could be resolved by very high Doppler factors in a substantial fraction of jets.","My inference: the short-baseline scheduling bias means the true detection fraction at baselines below 2 Earth diameters is probably higher than plotted, so the two-thirds figure is more likely a lower bound on the prevalence of ultra-compact cores.","My inference: the same PFD-to-SNR calibration could be applied to multi-epoch observations of individual sources to test whether the ultra-compact component varies, connecting the survey to intraday-variability and scintillation studies.","My inference: combining the detection-fraction curves with core-size estimates from scattering should let future space-VLBI missions predict how many AGN will be detectable at baselines beyond 28 Earth diameters."],"forward_implications":["If two-thirds of radio-strong AGN contain such compact cores, jet emission models must accommodate sustained brightness temperatures above equipartition and often above the inverse-Compton limit.","Detections at baselines of about 25 Earth diameters at 18 cm and 15 Earth diameters at 1.3 cm set upper limits of tens of microarcseconds on the angular size of the emitting regions.","The excess of 18-cm detections at the longest baselines implies that interstellar scattering substructure, not intrinsic core structure, dominates the most extreme long-baseline detections at that wavelength.","The weak scattering at 6 cm and its absence at 1.3 cm mean that brightness-temperature measurements at those bands are largely free of scattering bias for typical mid-Galactic-latitude sightlines.","The similarity of the detection histograms to earlier ground and space VLBI surveys suggests a single continuous core-jet brightness distribution from ground to space baselines, with sensitivity rather than physics setting the difference."],"supporting_citations":[{"why":"Sets the $10^{11.5}$ K inverse-Compton brightness temperature limit that the paper's detections exceed.","marker":"Kellermann and Pauliny-Toth, 1969"},{"why":"Provides the equipartition brightness temperature and the one-day Compton-cooling timescale used to interpret excess brightness.","marker":"Readhead, 1994"},{"why":"Supplies the result that ground VLBI measures AGN brightness temperatures around $10^{13}$ K and identifies cores as the compact emission region.","marker":"Kovalev et al., 2005"},{"why":"Provides the VSOP 5-GHz survey detection-fraction comparison for the baseline histograms.","marker":"Horiuchi et al., 2004"},{"why":"Supplies the fringe-fitting and SNR/PFD statistical method used to define detections.","marker":"Petrov et al., 2011"},{"why":"Supplies the theoretical noise distribution fitted to the low-SNR peak to calibrate PFD from SNR.","marker":"Thompson et al., 2017"},{"why":"Supplies the scattering-substructure estimates used to conclude that 6-cm results are weakly affected and 1.3-cm data are scattering-free.","marker":"Johnson and Gwinn, 2015"},{"why":"Reports scattering substructure discovered in pulsar observations on RadioAstron baselines, the basis for interpreting the 18-cm long-baseline excess.","marker":"Gwinn et al., 2016"}],"fun_headline_variants":["Space VLBI survey detects 2/3 of AGN cores","AGN survey: 2/3 cores have microarcsecond structures","Most AGN cores packed into microarcsecond scales","RadioAstron survey finds ultra-compact AGN cores","2/3 of AGN cores compact on microarcsecond scales"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The two-thirds detection fraction is assumed to represent the complete 163-source sample even though about 10% of targets, mostly low-declination sources, had not been observed by June 2016 and short-baseline K/L-band observations were under-scheduled; if those missing or under-scheduled sources are systematically fainter or more extended, the sample-level prevalence of ultra-compact cores is overestimated.","fun_headline_variants_meta":{"raw":{"variants":["Space VLBI survey detects 2/3 of AGN cores","AGN survey: 2/3 cores have microarcsecond structures","Most AGN cores packed into microarcsecond scales","RadioAstron survey finds ultra-compact AGN cores","2/3 of AGN cores compact on microarcsecond scales"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000186,"raw_usage":{"total_tokens":1270,"prompt_tokens":835,"completion_tokens":435,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":451,"completion_tokens_details":{"reasoning_tokens":348}},"tokens_in":451,"tokens_out":435,"duration_ms":4126,"temperature":1.0,"reasoning_tokens":348,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T05:36:12.257990+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Re-observing the roughly 16 unobserved low-declination targets with a southern tracking station and uniform short-baseline scheduling would settle the matter: if the detection fraction on baselines beyond a few Earth diameters among these sources falls below about one-third, the claim that two-thirds of the complete sample contains ultra-compact cores would not survive.","supporting_citations":[],"review_version":1}