{"id":"d60369ba-0fc1-48d4-b95f-1d99f15d06bb","arxiv_id":"1908.07308","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"A study of 355 FIRST-SDSS quasars shows that the radio-loudness criterion RK>10 is not uniformly valid across the quasar main sequence; extreme Population A sources may be radio-loud due to star formation, while Population B sources are systematically jetted.","lead":"This paper tests whether the standard radio-loudness criterion for quasars, Kellermann's RK>10, reliably identifies jetted sources. The authors find that the criterion depends on where a quasar sits on the optical main sequence, with extreme high-accretion sources often being radio-loud because of star formation rather than jets.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The xA thermal interpretation, which drives the claim that Kellermann's RK>10 criterion fails for extreme Population A, rests on external FIR samples with no FIRST-selected xA RI/RL sources; the paper itself concedes the lack of direct evidence.","rationale":"The reader's weakest-assumption analysis identifies precisely the same load-bearing concern: the xA thermal claim is imported from external, non-FIRST-selected comparison samples, while the primary sample contributes no FIR-detected xA source. My read confirms this, and the manuscript itself contains an explicit concession in Sect. 5.3.2 that there is no direct evidence that logRK >= 1 fails to select truly jetted sources among xA. The Population B jetted conclusion is much better supported: FRII sources are confined to the RL class, their radio powers are an order of magnitude or more above xA values, the KS tests in Sect. 5.3.1 show significant differences, and the orientation analysis in Sect. 5.4 provides a coherent unification picture. The binary-BLR candidate list is also presented with appropriate caveats and an F-test, so it does not weaken the main argument. However, the headline claim about the main-sequence dependence of Kellermann's criterion depends on the xA thermal interpretation, and that interpretation is not yet tested on the actual RI/RL xA sources in this sample. The reader's CONDITIONAL verdict is therefore appropriate; I would not upgrade to ACCEPT or downgrade to REJECT. The condition should be: obtain FIR and high-resolution radio data for the Table 2 xA candidates and demonstrate that they lie on the star-forming FIR-radio correlation before asserting that RK>10 is insufficient for extreme Population A.","tokens_in":58556,"tokens_out":3692,"duration_ms":42380,"concrete_test":"Obtain Herschel/ALMA 70-160 micron photometry and e-MERLIN/VLBI 1.4-5 GHz imaging for the 27 xA candidates in Table 2, especially the five RL CD A3/A4 sources (J010123.42+005013.3, J083558.43+261444.4, J103346.39+233220.0, J132819.23+442432.9, J142549.19+394655.0). Plot their L_FIR vs P_1.4 on the Bonzini et al. (2015) FIR-radio diagram; if any RL xA object lies more than 3 sigma above the star-forming locus or shows a compact flat-spectrum core, the thermal interpretation for that object is falsified and the RK>10-insufficiency claim loses its only supporting evidence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central headline conclusion — that the validity of Kellermann's criterion depends on location along the main sequence, with extreme Population A (A3/A4) RI/RL sources possibly thermal rather than jetted — is not supported by the primary sample. Table 5 lists only seven sources with FIR data, and none is an xA source; their spectral types are A2, A1, B1, B1+, or unclassified. The xA thermal inference therefore relies entirely on the external 73-source sample used in Fig. 9, which was 'vetted to preferentially select Pop. A and especially xA' (Sect. 4.4) but is not a FIRST-selected sample, and whose xA members are 'mainly classified as RD' (Sect. 4.4). The paper itself states in Sect. 5.3.2: 'in the samples used for Fig. 9 we have no indication of thermal xA sources entering into the RL radio-loudness range, and thus no direct evidence that the condition logRK >= 1 does not select truly jetted sources.' This is an explicit, load-bearing limitation: if the RL/RI xA sources in Table 2 are actually jetted, the claim that RK>10 is insufficient for xA collapses, leaving only the well-supported Population B jetted conclusion. The concern is not that the interpretation is impossible, but that the decisive evidence is imported from a comparison sample with a different selection function and radio-loudness distribution, and the paper's own data cannot discriminate.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper cross-matches the FIRST radio survey with the SDSS DR14 quasar catalog to build a sample of 355 radio-detected type-1 AGNs, classifies them along the quasar main sequence (spectral types A1-A5, B1-B3) and by radio morphology (core-dominated vs. FRII-like) and radio-loudness (RD, RI, RL). It analyzes composite optical spectra per spectral type and radio class, measuring H-beta and [O III] profile parameters, estimating black hole masses and Eddington ratios, and comparing radio-derived pseudo-SFRs with FIR-derived SFRs. The central claims are that (i) radio-loud and radio-intermediate sources in extreme Population A (A3/A4) may have thermal, star-formation-dominated radio emission rather than relativistic jets, so that Kellermann's RK>10 criterion may not identify truly jetted sources in this region of the main sequence; (ii) radio-loud Population B sources are characteristically jetted, with some broad H-beta profiles consistent with a binary broad-line region; and (iii) a list of binary black hole AGN candidates is presented.","tokens_in":58881,"tokens_out":4449,"duration_ms":46547,"significance":"If the thermal-origin interpretation for extreme Population A were established, the paper would provide an observationally grounded refinement of the standard radio-loudness classification and would identify a population of quasars where the Kellermann criterion fails. The paper also delivers useful resources: a systematically selected sample, a FIRST atlas of FRII candidates, detailed spectral decompositions, and a candidate list for SMBBH searches. The Population B jetted classification is well supported by the radio morphology and power distributions. However, the headline claim that RK>10 is insufficient for extreme Population A is not established by the primary sample: as the authors explicitly concede in Sect. 5.3.2, there is no direct evidence that thermal xA sources enter the RL range. The external FIR comparison samples are differently selected and their xA members are mostly RD. This weakness is load-bearing because the main-sequence-dependent validity of Kellermann's criterion rests on it. The binary-BLR F-test evidence is suggestive but based on composite spectra and not decisive.","major_comments":[{"comment":"The central claim that RI/RL sources in extreme Population A (xA, spectral types A3/A4) are thermal rather than jetted is not supported by the present sample. Table 5 lists only seven sources with FIR data, and none of them is an xA source; their spectral types are A2, A1, B1, B1+, or unclassified. The xA thermal inference therefore rests entirely on the external 73-source sample of Fig. 9, which was 'vetted to preferentially select Pop. A and especially xA' (Sect. 4.4) but is not FIRST-selected, and whose xA members are 'mainly classified as RD' (Sect. 4.4). The paper itself states in Sect. 5.3.2: 'in the samples used for Fig. 9 we have no indication of thermal xA sources entering into the RL radio-loudness range, and thus no direct evidence that the condition logRK >= 1 does not select truly jetted sources.' This is an explicit, load-bearing limitation: if the RL/RI xA sources in Table 2 are actually jetted, the claim that RK>10 is insufficient for xA collapses. The authors should either obtain or identify FIR data for the RL/RI xA sources in their own sample, or explicitly reframe the conclusion as hypothesis-generating rather than established.","section":"§4.4, §5.3.2, Table 5, Fig. 9"},{"comment":"The flat trend of logP_nu versus logRK' for xA sources is used as evidence of a different radio emission mechanism, but the xA subsample spans a very narrow range in logRK': in Table 1, xA sources are predominantly RI (15 out of 25) with only 5 RL and 5 RD, and the RI class is limited to 1.0 <= logRK' < 1.8 by definition. A regression slope that is not significantly different from zero over such a compressed dynamic range (slope 0.136 +/- 0.168) provides almost no discriminating power between a thermal and a jetted origin. The contrast with the well-constrained Population B slope is expected even if all sources share the same mechanism but have different luminosity/radio-loudness covariance. This point should be acknowledged and the evidential weight of Fig. 11 reduced accordingly.","section":"§5.3.1, Fig. 11"},{"comment":"The F-test comparison between the BC+VBC and BC+BC models is performed on composite (median) spectra, not on individual spectra, so the quoted probabilities (e.g., P = 8.3e-10 for RD CD B1+) quantify the improvement of fit to the composite profile, not the frequency of binary-BLR signatures among the individual sources. The authors themselves note alternative explanations (unresolved outflows, accretion disk instabilities, changing-look variability) and state that it is 'obviously not possible to demonstrate' a sub-pc binary from these data. Given that the candidate list is presented as a main result, the p-values should be interpreted as model-selection diagnostics for the composite, and the candidate list should be presented as a sample for follow-up rather than a statistically significant detection of binaries.","section":"§5.5, Table 6"}],"minor_comments":[{"comment":"The word 'exagons' appears to be a typo for 'hexagons'; please correct it.","section":"Fig. 11 caption"},{"comment":"The units for MBH are listed as 'M⊙/yr' but should be 'M⊙'; the table footnote should be corrected.","section":"Table 4 footnote"},{"comment":"The abstract defines RD as RK' < 10 while the text in §2.2 uses log RK < 1.0, which corresponds to RK' < 10; the notation should be made consistent so that the reader does not confuse the threshold with the value of the physical ratio.","section":"Abstract and §2.2"},{"comment":"In the description of Figure 5, the sentence 'This translates in a redshifted line' should read 'This translates into a redshifted line'.","section":"§4.1.2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is original and within the scope of A&A. The major concern is the gap between the headline conclusion about the Kellermann criterion and the evidence actually presented in the paper; the authors' own Section 5.3.2 disclaimer is appropriately honest but should determine how the conclusion is framed. No concerns about citation practice or novelty disclosure."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nQuick take: this paper is worth engaging seriously. It is the first systematic spectral-type-resolved analysis of a FIRST-selected radio sample along the quasar main sequence, and the most robust part is the Population B side. The RI/RL Population B sources look genuinely jetted by multiple independent indicators: FRII morphology, radio power distributions, and the radio-power vs. RK' correlation. The suggestion that extreme Population B sources, especially B1++ and B1+, are good targets for binary broad-line region searches is concrete and testable, and the candidate list has practical value.\n\nWhat is actually new here is the atlas-like breakdown: 355 objects classified by spectral type, radio-loudness, and morphology, with the FRII maps in the appendix. The flat P_radio vs. RK' slope for xA sources compared with the steep slope for Population B is an interesting observational fact that does not depend on the FIR interpretation. The paper is also honest about alternatives, for example outflow vs. binary BLR for the H-beta profiles.\n\nThe soft spot is the xA thermal claim. In the primary sample, only seven sources have FIR data and none is xA. The supporting Fig. 9 relies on external samples that were vetted to over-select Population A and xA sources, not on the FIRST-selected sample itself. The paper states this almost explicitly in Sect. 5.3.2: there is no direct evidence that logRK >= 1 fails to select truly jetted sources for xA. So the abstract's \"we show\" is stronger than the data justify. This is not a fatal flaw because the conclusion is ultimately framed as a suggestion and Table 2 is proposed for follow-up, but a revised version should either label the xA thermal interpretation as an externally motivated hypothesis or back it with new FIR and high-resolution radio data for the Table 2 candidates.\n\nThe binary BLR F-test is done on composite spectra and is suggestive rather than decisive. Minor concerns include the fixed FIR K-correction assumptions and the use of FIRST peak fluxes for compact sources, but these are not load-bearing. The citation pattern is fine and the core classification is not circular.\n\nBottom line: this deserves a serious referee, not a desk reject. The Population B results and the target list stand on their own; the xA thermal claim needs to be scaled down or further tested. I would send it to review and ask for that revision.\n\nBest,\n[You]","headline":"A systematic, useful MS-resolved radio-loudness study whose Population B jetted conclusion is solid, but whose xA thermal headline rests on external FIR data and is explicitly under-supported in the paper itself.","tokens_in":59431,"tokens_out":3322,"would_cite":true,"duration_ms":40859,"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":"The standard radio-loudness criterion $R_K>10$ picks out jetted quasars only on part of the main sequence; in extreme Population A it can select star formation instead.","keywords":["quasar main sequence","radio-loudness","Kellermann criterion","relativistic jets","star formation","broad-line region","binary black holes","FIR-radio correlation"],"falsifier":"Measure infrared brightness at 70–160 microns for the radio-intermediate and radio-loud extreme-A quasars in the paper's candidate list. If they fall below the star-forming infrared-radio relation and show compact flat-spectrum core-jet structure, the thermal interpretation fails; if they stay on it, the dichotomy holds.","tokens_in":58378,"feed_emoji":"📡","tokens_out":10925,"duration_ms":95381,"temperature":0.7,"pith_summary":"This paper tests whether the standard criterion for calling a quasar radio-loud—the radio-to-optical flux ratio $R_K>10$—actually identifies quasars whose radio emission comes from a relativistic jet. Using 355 radio-detected SDSS quasars arranged along the quasar main sequence, it finds that the answer depends on where a source falls on that sequence. In the most extreme Population A spectral types (A3 and A4), sources classed as radio-intermediate and even radio-loud reach radio powers compatible with star formation, not with a jet. In Population B, radio-loud sources are characteristically jetted, and their broad H$\\beta$ profiles are often consistent with a binary broad-line region. The paper concludes that the Kellermann criterion is not universally sufficient, and that main-sequence position must be considered before classifying a source as jetted.","feed_headline":"Star formation can masquerade as a radio-loud quasar","feed_subtitle":"The usual radio-loudness cutoff finds jets only in part of the quasar main sequence, not among the extreme accretors.","key_machinery":"The organizing device is the quasar main sequence, the plane defined by FWHM(H$\\beta$) versus the FeII strength $R_{\\rm FeII}$, subdivided into spectral types from A1 to A5 and B1 to B1+++. Within each spectral type, the analysis compares three radio-loudness classes ($R_K<10$, $10\\le R_K<70$, and $R_K\\ge70$) and two radio morphologies, compact core-dominated and extended FRII-like. The decisive comparison is between the radio pseudo-star-formation rate derived from the 1.4 GHz power and the star-formation rate derived from 70/100 $\\mu$m infrared photometry; sources that fall inside the star-forming, radio-quiet zone of the FIR-radio correlation are interpreted as thermally powered, while sources outside it are read as jetted.","core_discovery":"The central discovery is that radio-loudness is not one physical condition along the quasar main sequence. The authors build a sample of 355 type-1 quasars from a FIRST-SDSS cross-match, split into compact core-dominated and extended FRII-like radio morphologies, and into radio-detected, radio-intermediate, and radio-loud classes inside each optical spectral type. Extreme Population A (xA: A3 and A4, strong FeII and narrow H$\\beta$) contains a notable prevalence of radio-intermediate and radio-loud sources, but their radio powers overlap those expected from star formation and their infrared-radio positions match radio-quiet quasars. Population B radio-loud sources, especially the FRII double-lobed ones, are jetted, and several of their H$\\beta$ composites are better described by two displaced broad components than by the usual broad-plus-very-broad model, pointing to a binary broad-line region. The paper therefore argues that $R_K>10$ is sufficient to identify jetted sources in Population B and in spectral bin A1, but not in extreme Population A.","pith_inferences":["If the thermal interpretation holds, radio-loudness-selected samples that ignore main-sequence position will systematically overestimate the number of jetted quasars at the highest Eddington ratios, and may inflate binary black hole candidate lists by mixing starbursts into them.","A direct extension would be to obtain high-resolution radio maps of the RL CD A3 candidates: compact symmetric or GPS/CSS morphologies combined with steep spectra would argue for a young jetted source, while diffuse or resolved disk-like radio emission would support a star-formation origin.","The binary-broad-line-region interpretation predicts that the two H$\\beta$ components should change radial velocity on decade timescales; single-epoch double-component fits alone cannot confirm a binary, so monitoring campaigns on the B1++ candidates would settle the matter.","If xA radio power is substantially thermal, then feedback at the most extreme accretion rates is probably driven by outflows and star formation rather than by radio jets, which would shift how high-accretion quasars are treated in galaxy evolution models."],"forward_implications":["In Population B and in spectral bin A1, the condition $R_K>10$ (or $\\log R_K>1$) remains a workable identifier of jetted sources, so those bins can be used as a clean jetted sample.","In extreme Population A (A3/A4), radio-intermediate and radio-loud sources should not be counted as jetted without infrared or high-resolution radio follow-up; they are prime targets for testing the thermal interpretation.","FRII double-lobed quasars appear almost exclusively in Population B, with a few A1 objects interpretable as the same jetted population seen at smaller viewing angle, so orientation must be included when using H$\\beta$ FWHM to classify sources.","The H$\\beta$ profile fits of several Population B bins are consistent with two broad-line regions orbiting each other; if secure, these sources are binary black hole candidates, and radio-loud extreme Population B is the preferred hunting ground for such binaries."],"supporting_citations":[{"why":"Defines the radio-loudness ratio $R_K=f_\\nu(\\mathrm{radio})/f_\\nu(\\mathrm{optical})$ and the $R_K>10$ jet threshold that this paper tests along the main sequence.","marker":"Kellermann et al. 1989"},{"why":"Showed that powerful jetted sources occupy Population B and proposed the stricter $R_K>70$ condition; the paper's bin-by-bin comparison directly tests that proposal.","marker":"Zamfir et al. 2008"},{"why":"Provides the FIR-radio correlation diagram separating star-forming galaxies and radio-quiet quasars from jetted sources, the diagnostic used to infer thermal radio power.","marker":"Bonzini et al. 2015"},{"why":"Established the Eigenvector-1 correlations that define the quasar main sequence and the FeII/H$\\beta$ parameter $R_{\\rm FeII}$ used for spectral classification.","marker":"Boroson & Green 1992"},{"why":"Supplies the spectral-type grid and the optically selected prevalence statistics along the main sequence to which the radio-selected sample is compared.","marker":"Marziani et al. 2013"},{"why":"Supplies the calibration from 1.4 GHz radio power to pseudo-star-formation rate used to estimate the expected thermal radio output.","marker":"Yun et al. 2001"},{"why":"Provides the FIR-based star-formation calibration and the comparison sample linking high-Eddington, FeII-strong quasars to active star formation.","marker":"Sani et al. 2010"},{"why":"Defines the FIRST 1.4 GHz survey whose detections, peak and integrated fluxes, and radio maps are the basis of the sample and morphology classification.","marker":"Becker et al. 1995"},{"why":"Supplies the statistical pair-ranking procedure used to identify physical FRII double-lobed systems from FIRST components around each quasar.","marker":"de Vries et al. (2006)"},{"why":"Provides the core-to-lobe ratio versus H$\\beta$ FWHM relation and the relativistic beaming model used to interpret orientation effects among jetted sources.","marker":"Wills & Browne 1986"}],"fun_headline_variants":["Radio-loudness criterion fails for extreme quasars","Star formation masquerades as radio-loud in extreme accretors","Quasar jets only appear in one part of the main sequence","Extreme accretors: radio from stars, not jets"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The thermal interpretation of the extreme Population A sources assumes that the infrared-radio correlation from previous samples correctly marks star formation as the dominant radio mechanism and that those samples represent the present selection; only seven objects here have infrared data and none is an extreme-A source.","fun_headline_variants_meta":{"raw":{"variants":["Radio-loudness criterion fails for extreme quasars","Star formation masquerades as radio-loud in extreme accretors","Quasar jets only appear in one part of the main sequence","Extreme accretors: radio from stars, not jets"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000248,"raw_usage":{"total_tokens":1642,"prompt_tokens":1134,"completion_tokens":508,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":750,"completion_tokens_details":{"reasoning_tokens":439}},"tokens_in":750,"tokens_out":508,"duration_ms":5236,"temperature":1.0,"reasoning_tokens":439,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:20:00.705540+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure infrared brightness at 70–160 microns for the radio-intermediate and radio-loud extreme-A quasars in the paper's candidate list. If they fall below the star-forming infrared-radio relation and show compact flat-spectrum core-jet structure, the thermal interpretation fails; if they stay on it, the dichotomy holds.","supporting_citations":[{"cited_title":"W., Plauchu-Frayn , I., & del Olmo , A","cited_arxiv_id":null,"evidence_quote":"Supplies the spectral-type grid and the optically selected prevalence statistics along the main sequence to which the radio-selected sample is compared."},{"cited_title":"2010, MNRAS, 403, 1246","cited_arxiv_id":null,"evidence_quote":"Provides the FIR-based star-formation calibration and the comparison sample linking high-Eddington, FeII-strong quasars to active star formation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the core-to-lobe ratio versus H$\\beta$ FWHM relation and the relativistic beaming model used to interpret orientation effects among jetted sources."}],"review_version":1}