{"id":"5486d469-f2a0-4e0e-b764-08d506052763","arxiv_id":"2607.05289","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":4,"one_line_summary":"Combining temporal scattering, depolarization, and RM variation measurements yields source-to-magneto-active-screen distances of ~AU to ~pc for six repeating FRBs, tentatively separating binary from SNR environments.","lead":"This paper combines three radio propagation effects—scattering, depolarization, and Faraday rotation variations—to estimate the physical distance between repeating FRB sources and their surrounding magnetized plasma. A smart generalist might read it because distinguishing AU-scale (binary companion) from pc-scale (supernova remnant) environments is key to identifying what powers FRBs.","discovery_kind":"unclear","skeptic_critique":{"model":"glm-5.2","headline":"The σ_rand = 0 assumption is not independently justified and acts as a hidden prior on D_B, but the reader correctly identifies the co-location assumption as the more fundamental and consequential concern.","rationale":"The reader correctly identified the co-location assumption as the single most load-bearing concern. The paper is transparent about this limitation (§5.2, hollow symbols in Fig. 2), and the CONDITIONAL verdict with MODERATE confidence is appropriate. The framework is genuinely novel as a methods paper, the algebra is sound, and the observational compilation is thorough. The concern does not reveal an internal inconsistency or error — it identifies an insufficiently constrained premise that the authors themselves acknowledge. The paper's claims are appropriately hedged ('tentatively favor,' 'under plausible assumptions'). No adjustment to the verdict is needed. The reader's identification of the log-linear slope discrepancy (~1.5σ) and the σ_rand = 0 assumption as additional concerns is also accurate, though these are secondary to the co-location bifurcation. The paper would benefit from explicitly computing and tabulating D_B under both scenarios for all 6 sources (rather than only showing the decoupled case in the right panel of Fig. 2), which would make the range of uncertainty from the co-location assumption fully transparent. As stated, the current results are suggestive, not conclusive, which matches the CONDITIONAL verdict.","tokens_in":29152,"tokens_out":838,"duration_ms":202654,"concrete_test":"For each of the 6 sources, compute D_B under both Eq. 7 (D_S = 100 pc, decoupled) and Eq. 8 (D_S = D_B, co-located) and tabulate the ratio D_B(Eq.8)/D_B(Eq.7). Then, using the YMW16 and NE2025 Galactic scattering predictions in Table 2, compute the fraction of observed τ_scat attributable to the local screen: f_local = (τ_obs - τ_Gal)/τ_obs. If f_local < 0.5 for any source classified as SNR-scale (FRB 20190303A, 20190417A, 20190520B), recompute D_B using Eq. 7 instead and check whether the classification flips from SNR to binary scale. If it does for any source, the headline classification is not robustly supported.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The reader identifies the co-location assumption (D_S ~ D_B) as the single most load-bearing concern. I agree this is the primary issue. When co-location holds, Eq. 8 gives D_B ~ 415 AU (AU-scale); when it fails, Eq. 7 gives D_B ~ 0.45 pc (pc-scale) — a factor of ~2000. The paper itself flags that for FRB 20180916B and FRB 20121102A, Galactic scattering (YMW16 predictions) is comparable to observed values (Table 2), meaning the co-location assumption is empirically unconstrained for at least 2 of 6 sources. For FRB 20190520B, the paper acknowledges scattering may be host-galaxy dominated (§5.2, citing Ocker et al. 2022). So for 3 of 6 sources, the premise distinguishing AU from pc conclusions is either unverified or likely violated. A secondary concern: the σ_rand = 0 assumption (Eq. 6→7) is justified only by a theoretical argument (Yang et al. 2022) and acknowledged as potentially violated in active environments (§5.2). Since D_B ∝ (σ_RM^2 - σ_rand^2)^{1/2} in Eq. 6, any non-negligible σ_rand systematically inflates D_B. However, this is a second-order effect compared to the co-location bifurcation, which changes the answer by three orders of magnitude. The co-location issue is the load-bearing one.","agreement_with_reader":"agree"},"referee_report":{"model":"glm-5.2","summary":"This manuscript presents a method to infer the physical distance between repeating FRB sources and their magneto-active environments by jointly analyzing three propagation observables: temporal scattering (τ_scat), depolarization (σ_RM), and RM variation rate (|ΔRM/Δt|). The framework (Eqs. 2–8) builds on the Burn (1966) depolarization formalism and standard scattering geometry (Cordes et al. 2016). Applied to six repeating FRBs, the method yields source-to-screen distances D_B spanning AU to pc scales, which the authors interpret as tentatively favoring SNR-scale environments for three sources and allowing binary-scale structures for two others. The paper also presents a log-linear consistency check (Eq. 15, Figure 3) and simulation-based validation using the RM variation curve of FRB 20201124A (Appendix B).","tokens_in":30172,"tokens_out":1623,"duration_ms":139126,"significance":"The central derivation is algebraically clean and connects three independently measurable observables into a single geometric constraint on D_B, which is a genuinely useful diagnostic for the FRB community. The presentation of both the co-located (Eq. 8) and spatially decoupled (Eq. 7) cases is appropriate, and the authors are commendably transparent about the limitations of their assumptions. The simulation appendix (Appendix B) provides a constructive cross-check on the analytical framework. The method is falsifiable and motivates specific observational priorities (contemporaneous wideband polarimetry, scintillation-based screen distances). However, the practical applicability of the framework is currently limited by the co-location assumption and by data quality for several sources, as discussed below.","major_comments":[{"comment":"§5.2 and Abstract: The central claim distinguishing AU-scale from pc-scale conclusions rests on the co-location assumption (D_S ~ D_B, Eq. 8 vs. Eq. 7). The paper itself flags that for FRB 20180916B and FRB 20121102A, Galactic scattering (YMW16 predictions in Table 2) is comparable to observed values, and for FRB 20190520B, host-galaxy scattering may dominate (§5.2, citing Ocker et al. 2022). This means the co-location premise is either unverified or likely violated for at least 3 of 6 sources. The abstract's statement that distances 'tentatively favor SNR-scale magneto-environments for FRB 20190303A, FRB 20190417A, and FRB 20190520B' should be qualified more prominently, especially for FRB 20190520B where the scattering is acknowledged as likely host-dominated. The paper should explicitly state which sources have secure local scattering and which do not, rather than presenting all six D","section":null},{"comment":"Table 2 and §5.2: For FRB 20180916B, the YMW16-predicted Galactic scattering at the depolarization frequency (43 ms) is comparable to the observed value (38 ms), while the NE2025 prediction (280 ms) exceeds it. This discrepancy between the two models means the fraction of scattering attributable to the local environment is model-dependent and could range from negligible to dominant. The paper should discuss this model discrepancy explicitly and its impact on D_B for this source, since the difference between Eq. 7 and Eq. 8 changes D_B by ~3 orders of magnitude.","section":null},{"comment":"Eq. 6 and the σ_rand = 0 assumption (stated after Eq. 6): The justification references Yang et al. (2022), but the manuscript acknowledges in §5.2 and Appendix A (Eq. A3) that σ_rand may be non-negligible in active environments. Since D_B ∝ (σ_RM^2 - σ_rand^2)^{1/2} in Eq. 6, any non-negligible σ_rand systematically inflates D_B. For sources like FRB 20190520B with |RM| ~ 10^4 rad/m², the stochastic component could be substantial. A brief quantitative argument for why σ_rand << σ_RM is expected (or a sensitivity analysis showing how D_B shifts for plausible σ_rand/σ_RM ratios) would strengthen the framework's applicability.","section":null},{"comment":"§4.3, Eq. 15, and Figure 3: The log-linear consistency check yields a fitted slope of 0.31 ± 0.19, which is inconsistent with the predicted slope of unity at ~1.5σ. The paper attributes this to 'dynamic evolution process or host multi-class origins,' but an alternative explanation is that the co-location assumption fails for some sources, causing them to scatter off the Eq. 15 relation. The paper should discuss whether the shallow slope is itself evidence that the framework's assumptions are violated for a subset of sources, rather than attributing it solely to source diversity.","section":null}],"minor_comments":[{"comment":"Abstract: 'supernvae' should be 'supernovae'; 'has the potential to systematic discrimination' should be 'has the potential to systematically discriminate.'","section":null},{"comment":"§1: 'only a small fraction of which are associated with persistent radio counterparts' — this sentence appears before the text discusses source models and could be better integrated into the context about source models.","section":null},{"comment":"Eq. (4): The factor of 2 in the first term is explained as accounting for 2D scattering, and the 1/√2 factor for directional sampling is mentioned in the text following Eq. 4. A cross-reference to Appendix A (Eq. A4) where this is derived would help the reader.","section":null},{"comment":"Figure 1 caption: 'velosity' should be 'velocity.'","section":null},{"comment":"§4.1: The statement 'adopting a fiducial transverse velocity of 100 km/s flexible from 10-1000 km/s' is grammatically awkward. Consider rephrasing.","section":null},{"comment":"Table 1: The footnote symbols (†, ¶, *) are inconsistent with the footnote text. The ¶ symbol appears in the footnote but no corresponding marker is visible in the table. Consider standardizing.","section":null},{"comment":"§5: The claim that the framework is 'model-independent' is somewhat overstated given the co-location assumption and the σ_rand = 0 prior. Consider softening to 'less model-dependent.'","section":null},{"comment":"Figure 3: The two non-repeating FRBs from Uttarkar et al. (2025) are included but their role in the analysis is unclear. A brief comment on why they are shown and whether they are expected to follow the same relation would help.","section":null},{"comment":"Appendix B: The simulation methodology is described but the connection between the simulated screen size (in hours) and the physical D_S is not fully transparent. The conversion formula t_recon = D_B/v × √(2c τ_scat/D_S) is given but the assumptions entering this step should be stated more explicitly.","section":null},{"comment":"References: Wang et al. (2025a) and (2025b) both cite arXiv:2507.15790. If these are the same paper, they should be consolidated.","section":null}],"recommendation":"major_revision","confidential_remarks":"The reader's report and stress-test note correctly identify the co-location assumption as the load-bearing issue. The paper is transparent about this limitation, which is appropriate, but the abstract and conclusions do not sufficiently flag which specific source-level conclusions are secure versus contingent on this assumption. The manuscript is a solid methods paper that needs better hedging and a clearer separation of robust vs. assumption-dependent results before publication. The simulation appendix is a nice addition but currently feels disconnected from the main analysis; the authors could either integrate it more tightly or present it as a standalone validation."},"author_rebuttal":{"model":"glm-5.2","summary":"We thank the referee for a careful and constructive report. The referee's central concern—that the co-location assumption is unverified or likely violated for several sources—is well-taken, and we agree that the manuscript needs to be more transparent about which sources have secure local scattering. We address each comment below and describe revisions we will make.","responses":[{"response":"We agree with this comment. The manuscript already flags these issues in §5.2 and uses hollow symbols in Figure 2 for sources with unknown circum-source scattering (FRB 20121102A, FRB 20180916B, and FRB 20201124A), but the abstract does not adequately convey the distinction between sources with secure and insecure local scattering. We will revise the abstract to explicitly note that for three of the six sources the local scattering origin is uncertain, and that the SNR-scale conclusion for FRB 20190520B in particular is contingent on the co-location assumption despite its scattering being likely host-dominated (as acknowledged in §5.2, citing Ocker et al. 2022). We will also add a summary table or explicit statement in §4.1 classifying each source as having (a) secure local scattering, (b) ambiguous scattering origin, or (c) likely non-local scattering, so the reader can immediately assess the reliability of each D_B estimate. For FRB 20190520B specifically, we will state that the SNR-scale inference should be treated as an upper limit on D_B under the co-location assumption, and that the spatially decoupled scenario (right panel of Figure 2) provides a more conservative estimate. We note that the manuscript already presents both the co-located (Eq. 8) and decoupled (Eq. 7) results in Figure 2, so the framework itself does not depend on the co-location assumption; the issue is one of presentation and emphasis, which we will correct.","revision_made":"yes","referee_comment":"§5.2 and Abstract: The co-location assumption is unverified or likely violated for at least 3 of 6 sources. The abstract should be qualified more prominently, especially for FRB 20190520B. The paper should explicitly state which sources have secure local scattering and which do not."},{"response":"This is a fair and important point. The discrepancy between YMW16 and NE2025 predictions for FRB 20180916B is indeed large: YMW16 predicts 43 ms (comparable to the observed 38 ms), while NE2025 predicts 280 ms (exceeding the observed value). Under YMW16, the local scattering contribution could be negligible, placing FRB 20180916B in the decoupled-screen regime (Eq. 7, D_B ~ pc scale). Under NE2025, the Galactic contribution exceeds the observed scattering, which is unphysical and suggests either that the NE2025 model overestimates scattering along this particular sightline or that the observed scattering has a local origin. We will add an explicit discussion of this model discrepancy in §5.2, noting that the two models bracket the range from 'Galactic-dominated' to 'locally dominated' and that the resulting D_B for FRB 20180916B is correspondingly uncertain by ~3 orders of magnitude. We will also note that FRB 20180916B is already shown as a hollow symbol in Figure 2 precisely because of this ambiguity, and that the binary-scale interpretation for this source should be understood as applying only under the co-location assumption. We cannot resolve the model discrepancy with current data; this is a genuine limitation that requires independent scattering measurements (e.g., from scintillation bandwidth analysis) to address.","revision_made":"yes","referee_comment":"Table 2 and §5.2: For FRB 20180916B, YMW16 predicts 43 ms vs observed 38 ms, while NE2025 predicts 280 ms. This model discrepancy means the local scattering fraction is model-dependent and could range from negligible to dominant. The paper should discuss this explicitly and its impact on D_B, since Eq. 7 vs Eq. 8 changes D_B by ~3 orders of magnitude."},{"response":"We agree that a quantitative sensitivity analysis would strengthen the paper. The theoretical justification for σ_rand << σ_RM comes from Yang et al. (2022), who showed that for a turbulent screen with outer scale l_s and thickness ΔR, the stochastic RM contribution σ_RM,clump scales as (ΔR/l_s)^{1/2} × δRM(l_s), where δRM(l_s) is the RM fluctuation on scale l_s. For the large-scale gradient to dominate, the RM variation must be coherent over scales larger than the scattering disk, which is expected when the RM variation is driven by bulk motion through a structured medium rather than by small-scale turbulence. However, the referee is correct that for FRB 20190520B, with |RM| ~ 10^4 rad/m² and dramatic RM reversals, the stochastic component could be non-negligible. We will add a sensitivity analysis showing how D_B shifts for plausible σ_rand/σ_RM ratios. Specifically, since D_B ∝ (σ_RM^2 - σ_rand^2)^{1/2}, a ratio σ_rand/σ_RM = 0.3 would reduce D_B by ~5%, while σ_rand/σ_RM = 0.5 would reduce it by ~13%, and σ_rand/σ_RM = 0.7 would reduce it by ~29%. These are sub-dominant compared to the order-of-magnitude uncertainties from the velocity and co-location assumptions, but they are systematic in the direction of inflating D_B. We will include this calculation in §5.2 and note that for the most active sources (FRB 20190520B, FRB 20201124A), σ_rand/σ_RM ~ 0.3–0.5 is plausible, leading to a modest (~10–15%) overestimate of D_B that does not change the SNR vs. binary classification but should be acknowledged as a systematic bias.","revision_made":"yes","referee_comment":"Eq. 6 and the σ_rand = 0 assumption: D_B ∝ (σ_RM^2 - σ_rand^2)^{1/2}, so non-negligible σ_rand inflates D_B. For sources like FRB 20190520B with |RM| ~ 10^4 rad/m², σ_rand could be substantial. Need quantitative argument or sensitivity analysis."},{"response":"The referee raises a valid alternative interpretation that we did not discuss. A shallow slope in the log(√τ_scat |RM|)–log(σ_RM) relation could indeed arise if the co-location assumption fails for a subset of sources. Specifically, if the scattering screen is more distant than the magneto-active region (D_S >> D_B), then Eq. 8 overestimates D_B relative to the true value, and the data point would fall below the unity-slope relation. This would flatten the fitted slope. We will add a discussion in §4.3 noting that the shallow slope is consistent with two non-exclusive explanations: (1) genuine diversity in source environments (as currently stated), and (2) violation of the co-location assumption for sources where scattering is dominated by the host ISM or Milky Way rather than the local environment. We note that the three sources with the most uncertain local scattering origin (FRB 20121102A, FRB 20180916B, FRB 20201124A) are also among those that deviate most from the unity-slope relation, which is consistent with the referee's interpretation. We will state this explicitly and note that distinguishing between these explanations requires independent constraints on D_S, such as from scintillation analysis. We will also note that with only 6–8 data points and large observational uncertainties, the slope measurement is not yet statistically powerful enough to discriminate between these scenarios, but the referee's interpretation is a plausible and important one that should be discussed.","revision_made":"yes","referee_comment":"§4.3, Eq. 15, and Figure 3: The fitted slope of 0.31 ± 0.19 is inconsistent with the predicted slope of unity at ~1.5σ. The paper attributes this to 'dynamic evolution process or host multi-class origins,' but should also discuss whether the shallow slope is evidence that the co-location assumption fails for some sources."}],"tokens_in":28931,"tokens_out":2324,"duration_ms":66152,"standing_objections":[]},"desk_editor":{"model":"glm-5.2","letter":"The main thing to know: this paper combines three propagation observables — RM variation rate, depolarization (σ_RM), and scattering timescale (τ_scat) — into an algebraic estimator for the source-to-magneto-active-screen distance D_B (Eqs. 6–8). That synthesis is genuinely new. The individual ingredients (Burn 1966 depolarization, standard scattering geometry, RM structure functions) are all established, but nobody had put them together this way before. The derivation is clean and the observational compilation is thorough — they pulled together multi-telescope RM, depolarization, and scattering data for six active repeaters, which is useful by itself. Appendix B includes a simulation validating the method on synthetic screens using FRB 20201124A's RM curve, recovering ~169 AU under the co-location assumption. That is a nice internal check and gives me some confidence the algebra is doing what they claim. The paper is also refreshingly honest about its limitations — they flag every soft spot I would want flagged. The load-bearing issue is the co-location assumption (D_S ~ D_B). When it holds, Eq. 8 gives AU-scale distances; when it fails, Eq. 7 gives pc-scale distances — a factor of ~2000. The paper itself notes that for FRB 20180916B and FRB 20121102A, YMW16 Galactic scattering predictions are comparable to observed values, and for FRB 20190520B the scattering is likely host-galaxy dominated (Ocker et al. 2022). So for at least 3 of 6 sources, the premise that distinguishes AU from pc is unverified or likely violated. They show both scenarios in Figure 2, which is the right thing to do, but it means the specific source classifications should not be taken as results — they are illustrative. The σ_rand = 0 assumption is a secondary concern. The paper acknowledges it (§5.2, Eq. A3) and notes it would inflate D_B if violated, but this is a second-order effect compared to the co-location bifurcation. The log-linear consistency check (Fig. 3) yields a slope of 0.31 ± 0.19 versus the predicted unity — about 1.5σ off, not great but not damning given six sources and heterogeneous data. No code or data products are shipped, and some parameter choices involve manual inspection of RM time series, which limits exact reproducibility. This is a methods paper with a tentative first application, not a definitive classification of FRB environments. The framework has real value as a roadmap for CHORD and DSA observations, and the method itself is sound given its assumptions. It deserves a serious referee who can pressure-test the co-location issue and push the authors to sharpen the boundary between what the data currently supports and what it does not.","headline":"New method combining RM variation, depolarization, and scattering into an FRB environment distance estimator; results are suggestive but rest on an unverified co-location assumption.","tokens_in":29902,"tokens_out":1366,"would_cite":false,"duration_ms":33903,"reading_group":"no","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"glm-5.2","headline":"Three propagation signals pin FRB environments to AU or pc scales","keywords":[],"falsifier":"If independent scintillation or VLBI measurements show that the scattering screen for a given FRB is located at a very different distance from the magneto-active region (D_S >> D_B or D_S << D_B), then Eq. 7 rather than Eq. 8 applies, and the inferred D_B shifts by orders of magnitude—potentially moving a source from the binary regime to the SNR regime or vice versa. Additionally, if the observed scattering for sources like FRB 20180916B is confirmed to be dominated by the Galactic ISM, the local scattering contribution is overestimated and all inferred distances are underestimated.","tokens_in":29256,"feed_emoji":"📡","tokens_out":1230,"duration_ms":79412,"temperature":0.7,"pith_summary":"The paper introduces a method to determine the physical distance between a repeating fast radio burst source and its surrounding magnetized plasma by combining three independent propagation observables: temporal scattering (which constrains the angular scale of scattered images), depolarization (which measures RM fluctuations across those images), and the RM variation rate (which tracks how the line of sight sweeps through the medium as the source moves). When the scattering screen and the magneto-active region are co-located, the three observables close into a single equation (Eq. 8) yielding a source-to-screen distance in astronomical units; when they are spatially separated, a different relation (Eq. 7) yields a parsec-scale distance. Applied to six active repeaters with multi-epoch RM data, the method finds that three sources (FRB 20190303A, FRB 20190417A, FRB 20190520B) fall in the supernova-remnant regime (0.1–10 pc), while two (FRB 20180916B, FRB 20201124A) remain compatible with binary-companion scales (1–100 AU) under the co-location assumption, though they shift to SNR scales if the scattering screen is distant. The framework is explicitly model-independent: it does not assume a binary or SNR environment beforehand but infers the scale directly from observables, then compares against physical predictions.","feed_headline":"Three propagation signals pin FRB environments to AU or pc scales","feed_subtitle":"Combining scattering, depolarization, and RM variation reveals whether repeating FRBs live in binary or supernova-remnant environments","key_machinery":"Eq. 8 (co-located screen and magneto-active region): D_B ~ 415 AU * (sigma_RM/10 rad m^-2)^2 * (tau_scat/1 ms)^-1 * (|dRM/dt|/10 rad m^-2 day^-1)^-2 * (v/100 km/s)^2, with D_S/D_B ~ 1. Eq. 7 (spatially separated screen): D_B ~ 0.45 pc * (sigma_RM/10)^2 * (tau_scat/1 ms)^-1/2 * (|dRM/dt|/10)^-1 * (v/100 km/s) * sqrt(D_S/100 pc). The log-linear consistency check (Eq. 15) predicts a slope-unity relation between sqrt(tau_scat)|RM| and sigma_RM under a single-class origin; the observed slope is 0.31 +/- 0.19, shallower than unity but consistent within scatter.","core_discovery":"The central object is the source-to-magneto-active-screen distance D_B, derived by combining temporal scattering, depolarization width sigma_RM, and RM variation rate |dRM/dt| into a single geometric closure. The key finding is that this combination of three measurable propagation effects suffices to estimate the physical scale of an FRB's magnetized environment at order-of-magnitude level without assuming whether the environment is a binary companion wind or a supernova remnant. The application to six repeaters reveals a genuine diversity: some sources sit at AU scales consistent with binary separations, others at parsec scales consistent with SNR shells, and the assignment can flip between","pith_inferences":[],"forward_implications":["If the co-location assumption holds, the AU-vs-pc split among repeaters directly distinguishes binary-companion environments from supernova-remnant environments, constraining FRB progenitor channels.","Simultaneous wideband measurements of scattering, depolarization, and RM from upcoming instruments (CHORD, DSA) could reduce the current order-of-magnitude uncertainties to factor-of-few, enabling model discrimination for individual sources.","Independent scintillation-based screen-distance measurements (e.g., from interstellar scintillation) would provide an external anchor for D_S, directly testing whether the scattering and Faraday-active regions are physically associated.","If future data show the log-linear relation (Eq. 15) tightening toward slope unity across a larger sample, it would support a single environmental class for most repeaters; persistent scatter would suggest multi-class origins or evolutionary diversity.","The method could be extended to non-repeating FRBs with depolarization measurements, testing whether apparently one-off bursts share the same environmental scale distribution as repeaters."],"fun_headline_variants":["Three propagation signals size FRB magnetized environments","Scattering, depolarization, and RM variation size FRB environments","Sizing FRB magnetized environments with three propagation effects","Closure relation sizes FRB magnetized environments from AU to pc","Tracing FRB magnetized environments from AU to parsec scales"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The derivation of the AU-scale distance (Eq. 8) assumes the scattering screen and the magneto-active region are co-located (D_S ~ D_B) and that the observed scattering timescale is dominated by this local screen rather than by the Milky Way ISM or a distant host-galaxy structure. For at least two sources (FRB 20180916B and FRB 20121102A), Galactic scattering model predictions are comparable to the observed values, so Galactic contamination cannot be excluded. If the co-locity","fun_headline_variants_meta":{"raw":{"variants":["Three propagation signals size FRB magnetized environments","Scattering, depolarization, and RM variation size FRB environments","Sizing FRB magnetized environments with three propagation effects","Closure relation sizes FRB magnetized environments from AU to pc","Tracing FRB magnetized environments from AU to parsec scales","Propagation diagnostics reveal AU and pc scales around FRBs","Sizing FRB environments: binary winds or supernova remnants?","AU or parsec? Sizing FRB magnetized environments"]},"model":"glm-5.2","effort":"high","cost_usd":0.0,"raw_usage":{"total_tokens":1299,"prompt_tokens":607,"completion_tokens":692,"prompt_tokens_details":null},"tokens_in":607,"tokens_out":692,"duration_ms":13731,"temperature":1.0,"reasoning_tokens":626,"cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-07T19:34:38.850784+00:00","model_set":{"reader":"glm-5.2"},"falsifier":"If independent scintillation or VLBI measurements show that the scattering screen for a given FRB is located at a very different distance from the magneto-active region (D_S >> D_B or D_S << D_B), then Eq. 7 rather than Eq. 8 applies, and the inferred D_B shifts by orders of magnitude—potentially moving a source from the binary regime to the SNR regime or vice versa. Additionally, if the observed scattering for sources like FRB 20180916B is confirmed to be dominated by the Galactic ISM, the local scattering contribution is overestimated and all inferred distances are underestimated.","supporting_citations":[],"review_version":1}