{"id":"1fecba97-e41c-4544-9648-a0fc84cb40e1","arxiv_id":"2608.11968","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"MeerTRAP discovers 27 new rotating radio transients in the Galaxy, localizes 14 of them, and obtains coherent timing solutions for 4.","lead":"The MeerTRAP survey at South Africa's MeerKAT telescope found 27 new rotating radio transients, brief repeating radio pulses from neutron stars in our Galaxy. Fourteen of them were pinned down to arcsecond positions, and four turned out to show ordinary pulsar-like pulses, blurring the line between transient and steady radio emitters.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Systematic cross-match against known pulsar catalogues is not reported; unlocalized single-pulse sources could duplicate known pulsars and change the census.","rationale":"I read the paper as an observational catalog paper. The core deliverable is the list of 27 sources and the updated MeerTRAP census, so the most load-bearing assumption is that each candidate is genuinely new and astrophysical. The pipeline described in Section 2.1 (zero-DM excision, iqrm masking, DM greater than 20 pc cm^-3 rejection, frbid classification, and manual inspection) addresses RFI, and the two detailed comparisons in Section 3.1 address known-source confusion for two specific cases, but no systematic catalogue cross-match is described. Given that 13 of the 27 sources have no arcsecond localization and some are single pulses just above the S/N threshold, this gap is the point where the central claim is least secure. The reader's weakest_assumption already names 'known pulsars at other positions' as part of the central risk, so I agree with that identification. The reader's rationale, however, spends most of its weight on the reduced chi-squared values of the four timing solutions; those are important for the timing claims but do not change whether the 27 sources exist. My proposed check would settle the source-provenance question directly. I do not see an internal inconsistency that would warrant rejection; the appropriate verdict remains conditional pending the cross-match and ideally public release of the candidate list and dynamic spectra.","tokens_in":19575,"tokens_out":8245,"duration_ms":86619,"concrete_test":"Run a catalogue cross-match for all 27 sources: query ATNF psrcat and RRATalog within twice the positional uncertainty stated in Table 2 (about 1 arcmin for coherent-beam sources, about 1 degree for incoherent-beam sources, and the quoted arcsecond errors for imaging-localized sources). For every candidate match with DM within 20% of the reported discovery DM, re-reduce the stored discovery observation to test whether the pulse is recoverable at the known source's position and DM, and check whether the known source's period and phase predict the detected pulse times. If any match is confirmed, remove the source from the new-discovery list and recompute the census. This single check determines whether the '27 new' claim is exactly right or is contaminated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that all 27 sources are new Galactic radio transients. The weakest support for this is the provenance of the 13 sources without image-domain localization, especially the three incoherent-beam discoveries (MTP0093, MTP0102, MTP0103) whose positional uncertainty is about 1 degree and which were seen in only one or two pulses. Section 3.1 argues in detail only for MTP0093 and MTP0100 that a specific known pulsar is not the same source; it does not report a systematic cross-match of the full sample against the ATNF pulsar catalogue or the RRATalog. Without such a cross-match, a single S/N around 8 to 15 pulse seen at a compatible DM from a known pulsar inside the beam would be counted as a new discovery. This is a missing support rather than a demonstrated error, but it is load-bearing because the headline number 27 and the updated MeerTRAP census of 95 would both shrink if any unlocalized source is a re-detection. The timing issues identified by the reader are real but affect the secondary timing-solution claims, not the discovery list itself.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports 27 new Galactic rotating radio transients discovered by the commensal MeerTRAP single-pulse search on MeerKAT, raising the MeerTRAP census to 95. It provides arcsecond image-domain localizations for 14 sources, preliminary period estimates for 8 sources, coherent timing solutions for 4 sources (with period derivatives for 3), together with fluences, burst rates, and follow-up observations with Effelsberg and Murriyang. Four of the followed-up sources show persistent pulsar-like periodic emission, and PSR J1410−4804 shows two bright single pulses. The results are used to discuss burst-rate distributions, the blurring distinction between RRATs and canonical pulsars, and the position of the sources in the P–Pdot diagram.","tokens_in":19682,"tokens_out":4264,"duration_ms":42468,"significance":"If the new-source list is correct, this is a substantial contribution to the growing RRAT census and includes the first three S-band Galactic transients found by MeerTRAP. The paper's strengths are its direct single-pulse measurements, the imaging localizations that enable follow-up, the independent cross-checks already presented for a few sources (MTP0084, MTP0106, and the case-by-case arguments for MTP0093 and MTP0100), and the follow-up detection of persistent periodic emission from several sources, which supports the idea that single-pulse-discovered objects can be ordinary weak or nulling pulsars. However, the novelty of the full sample and the reliability of the timing solutions are not yet fully established, as detailed below.","major_comments":[{"comment":"The manuscript does not report a systematic cross-match of the 27 new sources against the ATNF pulsar catalogue or the RRATalog. This is most consequential for the 13 sources without image localization, especially the three incoherent-beam discoveries MTP0093, MTP0102, and MTP0103, whose positional uncertainty is about 1 degree and which were seen in only one or two pulses. The individual arguments in Section 3.1 for MTP0093 and MTP0100 are useful but do not establish novelty for the full sample. Please add a cross-match with search radii set by the beam size or positional uncertainty, and report for each source the closest known pulsar or RRAT, its DM, and the angular separation. Without this, the headline claim of 27 new sources and the updated MeerTRAP census of 95 are not fully supported.","section":"Section 3.1, Table 1"},{"comment":"The reduced chi-squared values of the four timing fits are 66.5, 132.9, 11.3, and 985. These values are far above 1 and imply that the formal TOA uncertainties are severely underestimated or the timing model is incomplete. The quoted period derivatives (2.94(2), 159.89(3), and 2.46(2) × 10^-15 s/s) therefore cannot be taken at face value. The paper should either apply EFAC/EQUAD or an appropriate noise model, discuss profile-evolution or DM-variation systematics, or explicitly present these as provisional timing solutions. The current presentation overstates the security of the rotational properties and of the P–Pdot placement in Figure 3.","section":"Section 3.2, Table 3"},{"comment":"The claimed phase coherence for PSR J1454−3338 is not adequately demonstrated. The text says that the UHF-band TOAs obtained after a 517-day gap aligned 'within 1% of the period' before the final refit; aligning within 1% of the period is not equivalent to phase connection across such a long gap. Please show the number of acceptable period-solution candidates, the timing residuals with the gap explicitly visible, and justify why a 1% alignment is sufficient for coherent timing. Without this, the period derivative for this source should be treated as preliminary rather than as a secure measurement.","section":"Section 3.2, PSR J1454−3338"},{"comment":"PSR J1750−4450 is plotted as a point in the P–Pdot diagram, but Table 3 reports its period derivative as 1.2(9) × 10^-15 s/s from an unweighted fit with reduced chi-squared 985 and the text describes only an upper limit. A value with this uncertainty and fit quality should not be plotted as a firm measurement. Replace it with an upper-limit marker or omit it from the population comparison.","section":"Section 4.2, Figure 3"}],"minor_comments":[{"comment":"The reference in the text to 'Manch ester et al. 2005' is a typo; it should read 'Manchester et al. 2005'.","section":"Section 4.2"},{"comment":"For sources without imaging localization, the table lists 'CB' or 'IB' in the localization-method column, but this is not explained in the table caption. Please add a footnote clarifying that these entries denote the discovery beam and that the positional uncertainty is the beam size.","section":"Table 2"},{"comment":"The 27 discovery plots are small, and the source names in the top-left corner are difficult to read. Consider enlarging the panels or using a separate label for each source.","section":"Figure 1"},{"comment":"The statement that burst rates are lower limits is clear, but for sources detected only once (e.g., MTP0089, MTP0090, MTP0091) no burst-rate value is given in Table 1; adding a formal upper limit or a 'not applicable' entry would make the table more interpretable.","section":"Section 3.3"}],"recommendation":"major_revision","confidential_remarks":"This is a valuable survey paper, but the novelty check for the unlocalized sources and the statistical treatment of the timing solutions need to be strengthened before publication. The high reduced chi-squared values in Table 3 are the main technical concern and should be addressed head-on rather than by reporting the fits as secure. I recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: this is a solid discovery paper, and the 27 new sources are real. The paper does exactly what a catalog paper should: it reports direct measurements — 14 arcsecond positions from imaging, 8 period estimates, fluences, burst rates, and four follow-up detections showing ordinary pulsar-like emission. The three S-band transients are genuinely new territory for MeerTRAP, and the follow-up program with Effelsberg and Murriyang strengthens the emerging picture that many RRATs are just faint or nulling pulsars. Credit where due: the source list, positions, and period estimates are internally consistent, and the authors are careful when arguing that MTP0093 and MTP0100 are not known pulsars. The methods are inherited from the group's earlier pipeline papers, which is fine — self-citation here is not a red flag because the results are direct measurements, not fits feeding back into the claim. The soft spots are real but not disqualifying. First, the four coherent timing solutions have reduced chi-squared values of 66, 133, 11, and 985. The quoted TOA uncertainties do not describe the residuals, so these fits are not phase-connected timing solutions in the usual sense. The paper should say that, or the authors should inflate the uncertainty until chi-squared is acceptable. The J1454−3338 solution is especially shaky: 14 pulses, a 517-day gap, and phase coherence based on a 1% period match. That claim needs a caveat or a proper cycle-count justification. Second, and more important for the headline number: I do not see a systematic cross-match of all 27 sources against the ATNF catalog or the RRATalog. The paper argues individually for two unlocalized sources, but the three incoherent-beam discoveries with ~1 degree uncertainties and one or two pulses each could in principle be re-detections of known pulsars. A table listing the nearest known pulsar, DM, and position offset for each source would settle this. It is likely that all 27 are new, but the paper currently assumes rather than demonstrates it. That is missing support, not a demonstrated error, but it is load-bearing for the census of 95. Who is this for? Anyone working on RRAT population statistics, neutron-star transient searches, or the pulsar/RRAT boundary. It is not a methods paper; the value is in the data and the follow-up results. A serious referee should get this, and the requested revisions should focus on the systematic cross-match and on treating the timing solutions with appropriate humility. I would cite it for the source list and the follow-up detections, and I would bring it to reading group if anyone in the group works on radio transients.","headline":"A credible catalog paper: 27 new MeerTRAP RRATs with real localizations, periods, and follow-up detections; the timing solutions and the missing systematic cross-match are the soft spots, but the discoveries stand.","tokens_in":758,"tokens_out":1055,"would_cite":true,"duration_ms":23642,"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 commensal single-pulse search with MeerKAT reports 27 new rotating radio transients, bringing the MeerTRAP census to 95 and showing that several such sources behave as ordinary pulsars in follow-up observations.","keywords":["rotating radio transients","RRATs","pulsars","single pulse search","MeerKAT","timing solutions","neutron star populations","Galactic plane survey"],"falsifier":"Independently re-check the 27 candidates against the ATNF pulsar catalogue and the published RRATalog: if any candidate matches a known source in position and DM outside the pairs already examined (MTP0084/CRAFTS, MTP0106/MPIfR, MTP0093/J0149+29, MTP0100/PSR J1832−0901t), one discovery would be erased. For the timing solutions, a decisive test is whether the published periods and period derivatives predict pulses at later epochs not used in the fits; the high reduced chi-squared values reported (11–985) already hint that the single-component timing model is incomplete.","tokens_in":19324,"feed_emoji":"📡","tokens_out":6262,"duration_ms":57639,"temperature":0.7,"pith_summary":"The paper reports the discovery of 27 new Galactic rotating radio transients (RRATs) from the commensal MeerTRAP single-pulse search on MeerKAT, raising the MeerTRAP census of such sources to 95. Of these, 14 were localised to arcsecond accuracy through imaging of the transient-buffer data, eight yielded preliminary spin periods between 0.78 s and 4 s, and four received phase-coherent timing solutions, three with measured period derivatives. Long follow-up observations with other telescopes found continuous, ordinary pulsar-like periodic emission from four of the new sources, indicating that many RRAT-like objects detected in single-pulse searches are actually weak or heavily nulling pulsars. A sympathetic reader would care because this sharpens the picture of an incomplete and blurred neutron-star census, showing that single-pulse and periodicity searches are complementary routes to the same underlying population.","feed_headline":"27 new rotating radio transients found in MeerKAT data","feed_subtitle":"Follow-up shows four are ordinary pulsars, blurring the line between RRATs and pulsars.","key_machinery":"The load-bearing mechanism is the MeerTRAP real-time single-pulse search pipeline running on MeerKAT's beamformed data: coherent and incoherent beams are searched for dispersed pulses, with zero-DM RFI excision, iqrm channel masking, rejection of events with DM below 20 pc $cm^{-3}$, a S/N > 8 threshold, frbid classification, and manual inspection. Positions come from the transient buffer, where raw voltages are correlated, imaged, and the source is identified in the detection beam; periods come from applying rratsolve to multi-pulse arrival times, and coherent timing comes from tempo2 fits to all detection epochs. Follow-up periodicity searches use fast folding as implemented in riptide, which is sensitive to long-period, low duty-cycle, high-nulling sources.","core_discovery":"The MeerTRAP programme has found 27 previously unknown Galactic radio transients, bringing its total to 95. Fourteen sources were positioned to arcsecond accuracy by imaging the voltage data held in the transient buffer, eight had periods estimated from arrival times of multiple pulses (0.78–4 s), and four sources were given coherent timing solutions with tempo2, three of them with a measured period derivative. Follow-up observations with the Effelsberg and Murriyang telescopes detected continuous periodic pulsed emission from four of the sources, including two with signs of nulling, showing that sources found through single pulses can be ordinary weak or nulling pulsars. The sample also includes the first MeerTRAP Galactic transients found at S band, and bright broadband pulses from PSR J1410−4804 with no detected periodic counterpart, which the authors place near the canonical RRAT category.","pith_inferences":["If the blurred separation between RRATs and ordinary pulsars is general, then the 'prototype RRAT' population may be smaller than thought; a statistical comparison of nulling fractions between single-pulse-discovered and periodicity-discovered samples would test this.","The off-plane sources MTP0084, PSR J0506−5900, and PSR J2225+1243 show DM-distance discrepancies of up to roughly a factor of 25 between Galactic electron density models; future parallax or scintillation-based distances would turn them into model discriminators.","PSR J1410−4804, with two bright broadband pulses in a 45-minute follow-up and no detected periodic emission, resembles an intermittent bright-pulse emitter; a longer monitoring campaign could measure its burst-energy distribution and decide between a nulling pulsar and a genuinely sporadic emitter."],"forward_implications":["The MeerTRAP sample grows to 95 Galactic transients, with burst rates between about 0.05 and 88 pulses per hour and fluences between 0.13 and 11 Jy ms, nourishing population statistics for sporadic neutron-star emitters.","Four single-pulse discoveries showed continuous periodic emission in follow-up, so single-pulse surveys are complementing periodicity searches and will keep finding the faintest pulsars.","Three coherent timing solutions place these sources on the $P$–$\\dot P$ plane; PSR J2225+1243, with $P = 4.0077$ s and an estimated age near 31 Myr, sits close to the pulsar death valley.","The first S-band MeerTRAP transients show that high-frequency searches can uncover high-DM, shallow-spectrum transients missed at lower frequencies.","Arcsecond localizations turn single-pulse finds into targets for follow-up timing, giving the period and period-derivative measurements needed to connect RRATs to the rest of the neutron-star population."],"supporting_citations":[{"why":"Defines the RRAT class in which this paper's transients are placed.","marker":"McLaughlin et al. 2006"},{"why":"Describes the MeerTRAP commensal observing programme that produced the detections.","marker":"Stappers 2016"},{"why":"Provides details of the MeerTRAP single-pulse search pipeline used to find the 27 sources.","marker":"Rajwade et al. 2022"},{"why":"Supplies the zero-DM RFI excision technique applied to all candidates.","marker":"Men et al. 2019"},{"why":"The riptide fast-folding algorithm used to search follow-up data for faint periodic emission.","marker":"Morello et al. 2020b"},{"why":"tempo2 is the timing package used for the coherent timing solutions of four sources.","marker":"Hobbs et al. 2006"},{"why":"Earlier MeerTRAP RRAT discoveries that set the baseline for the census now reaching 95.","marker":"Bezuidenhout et al. 2022"},{"why":"Previous MeerTRAP discoveries and the period-search method this work extends; also the earlier report of MTP0040.","marker":"Turner et al. 2025"},{"why":"The NE2001 Galactic electron density model used to estimate distances from dispersion measures.","marker":"Cordes & Lazio 2002"},{"why":"The YMW16 electron density model provides the alternative distance estimates compared in Table 2.","marker":"Yao et al. 2017"}],"fun_headline_variants":["27 new radio transients, four turn out to be pulsars","MeerTRAP finds 27 RRATs; follow-up reveals 4 are pulsars","New RRATs discovered, but some are just weak pulsars","27 new transients: RRATs or pulsars? Both, says MeerTRAP","MeerTRAP adds 27 RRATs, with pulsar impostors among them"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The source list rests on the premise that every candidate passing the RFI excision, classification, and manual inspection is a genuine astrophysical pulse from a distinct rotating neutron star, rather than radio-frequency interference or a previously catalogued pulsar outside the specific coincidences the paper already discusses.","fun_headline_variants_meta":{"raw":{"variants":["27 new radio transients, four turn out to be pulsars","MeerTRAP finds 27 RRATs; follow-up reveals 4 are pulsars","New RRATs discovered, but some are just weak pulsars","27 new transients: RRATs or pulsars? Both, says MeerTRAP","MeerTRAP adds 27 RRATs, with pulsar impostors among them"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000185,"raw_usage":{"total_tokens":1313,"prompt_tokens":926,"completion_tokens":387,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":542,"completion_tokens_details":{"reasoning_tokens":280}},"tokens_in":542,"tokens_out":387,"duration_ms":3882,"temperature":1.0,"reasoning_tokens":280,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:20:51.122622+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Independently re-check the 27 candidates against the ATNF pulsar catalogue and the published RRATalog: if any candidate matches a known source in position and DM outside the pairs already examined (MTP0084/CRAFTS, MTP0106/MPIfR, MTP0093/J0149+29, MTP0100/PSR J1832−0901t), one discovery would be erased. For the timing solutions, a decisive test is whether the published periods and period derivatives predict pulses at later epochs not used in the fits; the high reduced chi-squared values reported (11–985) already hint that the single-component timing model is incomplete.","supporting_citations":[],"review_version":1}