{"id":"216735a5-785a-4399-ad74-b6022563b0c9","arxiv_id":"2608.10452","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"FRB240312D is the first highly scattered FRB whose scattering screen is localized to within about 8 pc of the source, with the two new bursts implying a rate consistent with shorter FRBs.","lead":"Two newly found fast radio bursts have extremely long scattering tails, and for one of them the scattering is pinned to a region within about ten parsecs of the source rather than to the Milky Way or an intervening galaxy. This matters because it suggests strong scattering is produced by the burst's own environment, not by chance alignments, which changes how scattered FRBs are interpreted.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Screen-distance bound in Eq. (15) assumes 2D screens; the 1D orientation caveat noted in §3.1 is untested and could push D_host,src well beyond 8 pc.","rationale":"The reader's weakest assumption already identified the 1D-screen orientation caveat, and my independent reading agrees that this is the most load-bearing weakness in the paper's central claim. The paper is otherwise careful: the observational analysis is detailed, the scattering and scintillation fitting procedures are described, the host-galaxy follow-up is thorough, and the speculative discussion of pulsar wind nebulae is explicitly labelled as such. The main issue is that the abstract and the Section 5.1 conclusion present the 8 pc screen-distance bound as excluding all alternatives without carrying forward the acknowledged limitation that Eq. (15) assumes 2D screens and that the bound weakens for unfavourable 1D orientations. This does not make the paper unsound; it means the headline claim is conditional on an untested geometric assumption. The proposed concrete test—measuring the anisotropy of the scintillation pattern and propagating the relative screen orientation through the inequality—would settle whether the caveat actually relaxes the bound enough to threaten the circumsource-only conclusion. Because the reader's CONDITIONAL verdict already accounts for this and the associated lack of public data, no verdict change is needed.","tokens_in":30462,"tokens_out":6999,"duration_ms":69810,"concrete_test":"Using the recorded dynamic spectrum of FRB240312D, compute the two-dimensional intensity autocorrelation function over the on-burst window and fit the scintillation pattern's axial ratio and position angle. Combine this with the orientation of the host-screen structure (e.g., the Hα filament or spiral-arm direction at the FRB position) and re-derive the screen-distance inequality for two one-dimensional screens with the measured relative angle, following the treatment cited for Eq. (15). If the relaxed upper bound on D_host,src exceeds ~100 pc, the claim that all non-circumsource screens are excluded is unsupported; if it remains ≲10 pc, the concern is resolved.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's headline result for FRB240312D—that only the circumsource medium remains as the scattering screen—depends on Eq. (15), which yields D_host,src ≲ 8±2 pc (D_MW/kpc)^-1. The derivation of Eq. (15) assumes two-dimensional screens, as the authors acknowledge at the end of §3.1: for two one-dimensional screens the inequality holds only when the sky projections are parallel, and the constraint weakens towards perpendicular orientations. The paper does not measure the anisotropy or orientation of the scintillation pattern, nor the axis of the host screen, so the 8 pc bound is not robust to the 1D-screen geometry. If the host screen is, for instance, a filament oriented perpendicular to the Milky Way screen's scintillation axis, D_host,src could be substantially larger than 8 pc—possibly exceeding the size of the star-forming region and allowing host-ISM or intervening screens back into consideration. The abstract nevertheless states without qualification that host-centre, background-galaxy, and intervening-structure screens 'can all be excluded'. Because the circumsource-only conclusion is exactly the central claim, this unquantified caveat is the most load-bearing weak point. A secondary issue is that the quoted 8 pc value also assumes D_MW ≈ 1 kpc; if the Milky Way screen is an order of magnitude closer, the bound loosens to ~80 pc, although this alone would not invalidate a 'near-source' interpretation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper reports the discovery of two highly scattered fast radio bursts, FRB240210D and FRB240312D, with the ASKAP/CRACO backend, and presents multi-wavelength follow-up. For FRB240312D, the authors measure a scattering time of 300±48 ms at 1 GHz, a scintillation bandwidth of 1.32±0.33 MHz, and a host galaxy at z=0.04986 with the FRB offset from the galaxy centre at 6.4±1.1 kpc. Using the scintillation as arising from a Milky Way screen, they derive an upper limit on the source-screen distance of D_host,src ≲ 8±2 pc (D_MW/kpc)^-1 from Eq. (15), and conclude that the scattering must originate in the circumsource medium, most likely a pulsar wind nebula. They also construct a DM budget, analyse MUSE integral-field spectroscopy, search for a persistent radio source with ATCA, and estimate the rate of long-duration FRBs above 9 Jy ms.","tokens_in":30754,"tokens_out":12518,"duration_ms":105135,"significance":"If the screen-distance constraint is robust, FRB240312D would be the first highly scattered FRB for which host-centre, background-galaxy, and intervening-structure screens can be excluded, leaving the circumsource medium as the only viable location. That conclusion would have broad implications for the interpretation of other highly scattered FRBs and for the use of scattering as a probe of host or intervening media. The paper is strong in its transparent treatment of the scattering and scintillation fits, its detailed DM budget, and its honest discussion of model uncertainties; the rate estimate is a useful addition to the long-duration FRB population. The main weakness is that the central screen-distance bound rests on an assumption of two-dimensional screens and on an unmeasured Milky Way screen distance, as noted in the manuscript itself.","major_comments":[{"comment":"The headline conclusion that all screens other than the circumsource medium are excluded for FRB240312D rests on Eq. (15), which the authors themselves note in the final paragraph of §3.1 is valid only for two-dimensional screens; for two one-dimensional screens the bound holds only when their sky projections are parallel and weakens toward perpendicular orientations. The paper does not measure the anisotropy or orientation of the scintillation pattern, nor the axis of the host screen, so the quoted D_host,src ≲ 8±2 pc (D_MW/kpc)^-1 is not robust to the 1D-screen geometry. If the host screen is an unfavourably oriented filament, D_host,src could be substantially larger, possibly exceeding the size of the star-forming region and allowing host-ISM or intervening screens back into consideration. Because the abstract's unqualified statement that these screens 'can all be excluded' is exactly the central claim, the authors should either provide a quantitative assessment of the 1D-screen case (e.g., from the ACF shape or dynamic spectrum) or soften the abstract and conclusions accordingly.","section":"§3.1, Eq. (15); §5.1; Abstract"},{"comment":"The bound D_host,src ≲ 8±2 pc (D_MW/kpc)^-1 depends on the distance to the Milky Way screen, D_MW, which is not measured for this sightline. The scintillation screen is assumed to be in the Milky Way from the consistency of ν_s with NE2001/YMW16 predictions, but the screen could lie at a distance an order of magnitude smaller than 1 kpc, which would relax the bound to ~80 pc and weaken the exclusion of a host-ISM screen. The authors should explicitly state this dependence in the abstract and conclusions, or provide a constraint on D_MW from independent observations along this sightline.","section":"§3.1, Eq. (15); §5.1"}],"minor_comments":[{"comment":"The text in §4.1.6 states 'Assuming a constant MW halo DM DMMWhalo=30 pc cm−3 (Cook et al. 2023)' while Table 1 lists DMMWhalo = 40±10 pc cm^-3 for both FRBs; this discrepancy should be reconciled.","section":"§4.1.6, Table 1"},{"comment":"The DMhost value appears as 214+16−29 in Table 1 and as 225+17−30 in §5.1 (and as 224+17−30 in §4.1.6), without consistently specifying rest-frame versus observed-frame values; please harmonise the notation and numbers.","section":"§4.1.6, Table 1"},{"comment":"The rate estimate in Eq. (14) assumes a single fluence power-law index of −3/2 and a log-uniform width distribution; a brief discussion of the sensitivity of P_obs and R_tot to these assumptions would help the reader judge the robustness of the rate claim.","section":"§2.4"},{"comment":"The manuscript repeatedly uses 'spacial' instead of 'spatial' (e.g., in §2, §4.1.1, and §4.1.4); these typos should be corrected.","section":"Throughout"},{"comment":"In the sentence following Eq. (31), the derivation of the maximum gradient size ΔDM = 422 pc cm^-3 and its comparison to the observed DM would be clearer if the authors explicitly stated which DM value is used for the comparison.","section":"§5.2.1"}],"recommendation":"major_revision","confidential_remarks":"The paper presents a well-observed and carefully analysed pair of FRBs, and the multi-wavelength follow-up is commendable. The central scientific claim, however, depends on the screen-distance bound of Eq. (15), and the manuscript itself notes an unquantified caveat regarding one-dimensional screens. I would ask the editor to ensure that one referee with expertise in scintillation theory assesses whether the 1D-screen caveat can be closed with the existing data or whether the abstract and conclusions must be substantially softened. The dependence on D_MW should also be made explicit in the main claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, this is a solid, observation-heavy paper: two new highly scattered FRBs, with the closer one (FRB240312D) localized to a spiral arm of a face-on z=0.05 galaxy, and the scattering screen constrained to ~8 pc from the source by combining scattering time and Milky Way scintillation. Second, the central claim—that this is the first highly scattered FRB where only the circumsource medium remains—is plausible but the abstract states it more flatly than the evidence warrants. The paper itself flags at the end of §3.1 that Eq. (15) assumes two-dimensional screens; for two 1-D screens the bound weakens unless their sky projections are parallel. They don't quantify that weakening, and if it is substantial, D_host,src could be tens of pc, which changes 'circumsource only' into 'host vicinity.' That caveat belongs in the abstract.\n\nWhat's genuinely new: the two detections, the first scintillation-based screen-distance constraint applied to a highly scattered FRB, and the thorough follow-up—MUSE integral-field spectroscopy, ATCA PRS limits, a careful DM budget, and an intervening-galaxy search. The rate estimate (R_tot ~ 210, consistent with shorter FRBs) is a useful addition. The treatment of the host galaxy environment is honest: they exclude a supermassive black hole, cast doubt on HII regions, and argue a PWN is the least-bad model while admitting it needs fine-tuned orientation. Credit where due: the scattering and scintillation fits are careful, the uncertainties are quoted, and the analysis of the group environment and intervening galaxies is rigorous.\n\nSoft spots, in proportion: (1) the 1D-screen issue just described—this is the main technical weakness; (2) no data or code in the preprint (the Data Availability line says they'll come with the accepted article), which makes independent verification harder; (3) the DM budget leans on assumed Milky Way halo DM and electron-density models, but that's standard practice and the ranges are shown. None of these sink the paper. For FRB240210D, they appropriately say much less.\n\nWho this is for: anyone working on FRB scattering, host galaxy associations, or the rate of long-duration bursts. It deserves a serious referee, with the request that the authors quantify the 1D-screen effect or soften the abstract. I'd be surprised if the main conclusion dies in review, but it may need reframing from 'excluded' to 'strongly disfavored.'","headline":"A strong, careful FRB-scattering paper whose lead claim is credible but slightly overreaches the quantified evidence because of an unquantified 1D-screen caveat.","tokens_in":31450,"tokens_out":2955,"would_cite":true,"duration_ms":28399,"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":"FRB 240312D is the first highly scattered burst whose scatter broadening must come from its own circumsource medium.","keywords":["fast radio bursts","scatter broadening","scintillation","circumsource medium","pulsar wind nebula","ASKAP-CRACO","host galaxy"],"falsifier":"A future repeating burst from FRB 240312D whose scattering time and dispersion measure vary on timescales of days to years would support a dynamic circumsource screen; conversely, a burst whose scattering tail is resolved by VLBI to originate at a distance greater than the 8 pc bound, or a persistent radio source with luminosity above the quoted ATCA limit, would falsify the circumsource-medium conclusion.","tokens_in":30239,"feed_emoji":"📡","tokens_out":7393,"duration_ms":57513,"temperature":0.7,"pith_summary":"Two highly scattered fast radio bursts were found during commissioning of the CRACO backend on ASKAP, with 1 GHz scattering times of 34±6 ms and 300±48 ms. For the nearer burst, FRB 240312D, the authors combine the measured scatter broadening with scintillation produced by a screen in the Milky Way to bound the distance from the burst source to the screen responsible for broadening: at most about 8 pc times (1 kpc/D_MW). Because the burst sits in a spiral arm, far from its host galaxy's centre, and no foreground galaxy or intervening structure can account for the scattering, the paper concludes the screen lies in the circumsource medium. This makes FRB 240312D the first highly scattered FRB where every alternative screen location is excluded. The most plausible origin, according to the authors, is refractive scattering by a filamentary pulsar wind nebula a few hundred years old, though the required sightline is fine-tuned.","feed_headline":"Scattering screen of fast radio burst pinned to within 10 pc of source","feed_subtitle":"FRB 240312D is the first highly scattered burst whose broadening must come from its own circumsource medium.","key_machinery":"The load-bearing device is the two-screen constraint of equation (15), which converts the presence of Milky Way scintillation into an upper bound on the distance from the FRB source to the scattering screen: the scattered image must be smaller than the Milky Way screen's resolution, so the product $D_{\\mathrm{host,src}}D_{\\mathrm{MW}}$ is limited by observable quantities $\\nu_s$, $\\tau_s$, and frequency. This inequality is what excludes the host galaxy centre, foreground galaxies, and intervening structures, leaving only the circumsource medium. A secondary mechanism is the comparison of refractive and diffractive scattering theories, together with density limits from dispersion measure and free-free absorption, which the authors use to discriminate among physical screens such as pulsar wind nebulae, H II regions, and hypernebulae.","core_discovery":"The central discovery is that scattering in FRB 240312D is produced within roughly 10 pc of the FRB source. The evidence is a two-screen argument: observed scintillation with bandwidth $\\nu_s = 1.32\\pm0.33$ MHz at 808 MHz is consistent with a screen in the Milky Way, while the scatter broadening $\\tau_s \\approx 0.7$ s at 1 GHz rest frame is millions of times too large to come from that same screen. Because the source scintillates, the angular size of the scattered image must be smaller than the resolution of the Milky Way screen, which yields the constraint $D_{\\mathrm{host,src}}D_{\\mathrm{MW}} \\lesssim (1+z)D_{\\mathrm{src}}^2 \\nu_s / (8\\pi\\nu^2 m \\tau_s)$, bounding the source-screen distance to $D_{\\mathrm{host,src}} \\le 8\\pm2$ pc $(D_{\\mathrm{MW}}/\\mathrm{kpc})^{-1}$. Combined with the FRB's location 6.4 kpc from the centre of a face-on spiral galaxy at $z=0.04986$ and the absence of any intervening galaxy or large-scale structure along the sightline, this forces the screen to be in the circumsource medium. The authors then argue that the scattering is most likely refractive, arising from dense filaments in a pulsar wind nebula similar to the Crab, while noting that a hypernebula or other source models remain possible.","pith_inferences":["If the pulsar-wind-nebula interpretation is correct, young FRB sources may be surrounded by compact ionised nebulae that also shape their persistent radio emission; targeted searches for faint PRSs around nearby scattered FRBs could test this.","The fine-tuned orientation required suggests either a selection bias toward highly scattered sightlines or an additional, unmodeled structure (for example, a dense circumstellar shell) that could be probed by repeated bursts from the same source.","The measured rate implies that imaging searches at longer boxcar widths, such as CRACO, should uncover many more highly scattered FRBs, enabling a population-level test of scattering-origin scenarios."],"forward_implications":["FRB 240312D is the first highly scattered FRB whose scattering screen is localised to the circumsource medium; all other screen locations are excluded.","Strong scattering in other FRBs is more likely to be caused by the sources' own environments rather than chance-aligned foreground plasma.","The rate of FRBs with durations between 55.2 ms and 1 s and fluence above 9 Jy ms is $210^{+460}_{-180}$ events sky$^{-1}$ day$^{-1}$, consistent with the rate of shorter FRBs, implying a substantial population of long-duration events.","Scattering cannot be used as a reliable estimator of host-galaxy dispersion measure when the circumsource medium dominates, as it does here."],"supporting_citations":[{"why":"Established the two-screen interpretation for scattered FRBs, separating Milky Way scintillation from extragalactic scatter broadening.","marker":"Masui et al. 2015"},{"why":"Provides the derivation of the distance-product constraint (equation 15) that localises the screen.","marker":"Pradeep E. T. et al. 2025"},{"why":"Supplies the density, dispersion-measure, and free-free absorption limits used to compare possible circumsource media.","marker":"Kumar et al. 2024"},{"why":"The cloudlet model and fluctuation parameter used to rule out a diffractive screen in the host galaxy.","marker":"Cordes et al. 2022"},{"why":"Describes the CRACO detection and imaging pipeline that produced these two detections.","marker":"Wang et al. 2025"},{"why":"Provides the statistical association of FRB 240312D with a spiral arm in its host galaxy.","marker":"Gordon et al. 2025"},{"why":"Observational evidence that pulsar scattering often arises from filamentary structures, motivating the pulsar wind nebula model.","marker":"Stinebring et al. 2001"}],"fun_headline_variants":["FRB 240312D scattering traced to circumsource medium","First highly scattered FRB pinned to 10 pc from source","Scattering screen within 10 pc of fast radio burst source","Circumsource scattering identified for FRB 240312D","Fast radio burst scattering localized to immediate surroundings"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The distance bound assumes the Milky Way screen and the host screen are both two-dimensional; if the host screen is a one-dimensional filament oriented unfavourably, the scattering screen could lie farther than about 10 pc from the source.","fun_headline_variants_meta":{"raw":{"variants":["FRB 240312D scattering traced to circumsource medium","First highly scattered FRB pinned to 10 pc from source","Scattering screen within 10 pc of fast radio burst source","Circumsource scattering identified for FRB 240312D","Fast radio burst scattering localized to immediate surroundings"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00032,"raw_usage":{"total_tokens":1925,"prompt_tokens":1190,"completion_tokens":735,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":806,"completion_tokens_details":{"reasoning_tokens":652}},"tokens_in":806,"tokens_out":735,"duration_ms":6552,"temperature":1.0,"reasoning_tokens":652,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:20:46.948699+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A future repeating burst from FRB 240312D whose scattering time and dispersion measure vary on timescales of days to years would support a dynamic circumsource screen; conversely, a burst whose scattering tail is resolved by VLBI to originate at a distance greater than the 8 pc bound, or a persistent radio source with luminosity above the quoted ATCA limit, would falsify the circumsource-medium conclusion.","supporting_citations":[],"review_version":1}