{"id":"18c7bd46-86ad-4605-ba6f-03da36f9cc0b","arxiv_id":"2411.09203","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":8,"one_line_summary":"Local FRB rates are very sensitive to the assumed host-galaxy DM, and current localized samples show no significant host-DM difference between repeaters and non-repeaters.","lead":"Using CHIME and localized FRB samples, the authors estimate local volumetric rates for repeating and non-repeating fast radio bursts under different host-galaxy dispersion measure assumptions. They find that the assumed host DM changes the inferred rates by up to an order of magnitude, while current localized samples show no clear host-DM separation between repeaters and non-repeaters.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The claimed DMhost sensitivity of volumetric rates is set by an under-specified Vlimit boundary using the highest-DM source's 95% quantile; replacing it with a full likelihood could collapse the effect.","rationale":"I agree with the conditional verdict and with the reader's identification of the DM_IGM distribution as a fragile input, but I think the most load-bearing point is more specific: the rate calculation in Section 2.4 uses an unusual, under-specified Vlimit boundary based on the highest-DM source, and the claimed DMhost sensitivity is largely manufactured by that boundary choice. The reader flagged the DM_IGM model and the repeater efficiency, which are real concerns, but did not isolate the Vlimit construction itself. The proposed test would settle whether the central claim survives a more principled likelihood-based rate estimate. If it does not, the paper still has value as a compilation of host-galaxy data and a sensitivity study, but the headline conclusion about DMhost affecting rates would need to be substantially weakened. Since the reader's conditional verdict already allows for addressable methodological issues, no change to the verdict is needed.","tokens_in":19638,"tokens_out":4763,"duration_ms":61451,"concrete_test":"Recompute the Section 3.2 non-repeater rate with a forward-model likelihood: for each selected CHIME FRB with observed DM_ex, evaluate P(z | DM_ex, DMhost) from Eq. (5) with a prior on DMhost taken from the MCMC posterior, and estimate R by maximizing the joint Poisson likelihood with the CHIME exposure and selection function, without using the 95% highest-DM boundary to set Vlimit. Then compare R(DMhost=66.63)/R(DMhost=0) with the paper's factor-of-ten change. If the ratio drops below about 2, the headline claim that DMhost significantly affects volumetric rates is not robust to the Vlimit construction.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central rate claim follows from Eq. (8), where R is inversely proportional to Vlimit, but Section 2.4 does not specify a standard maximum-volume calculation. It states that the FRB with the highest DM_ex in the selected sample is used to set P(≤dlimit|DMex) ≈ 0.95 and that all other sources are required to satisfy P(≤dlimit|DMex) ≈ 1, using the IllustrisTNG-based DM_IGM distribution of Eq. (5). This makes a single order statistic control the entire surveyed volume, and no value of dlimit, Vlimit, or sensitivity to the 95% quantile is reported. Because changing the assumed DMhost changes dlimit and therefore Vlimit, the reported factor-of-ten rate increase with DMhost is, to a large extent, a consequence of this boundary construction rather than an independent empirical constraint. If the adopted DM_IGM distribution is misspecified at low redshift, or if a different quantile or a full likelihood treatment is used, the rates in Sections 3.2 and 3.3 could shift by an amount comparable to or larger than the claimed DMhost effect. The reader's concern about the DM_IGM distribution is therefore sharpened here: the specific Vlimit algorithm is the place where that distribution enters most forcefully, yet the algorithm itself is not described precisely enough to reproduce the numbers.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript re-examines FRB host-galaxy dispersion measures and local volumetric rates. It compiles a table of localized FRB host properties, runs a bilby MCMC to infer DMhost for 35 non-repeaters and 14 repeaters from a log-normal prior, and reports no significant median difference between the populations. It then selects CHIME/FRB Catalog 1 non-repeaters and Blinkverse repeaters, computes Vlimit from a probability statement P(≤dlimit|DMex) using the IllustrisTNG-based DMIGM distribution of Eq. (5), and derives volumetric rate densities from Eq. (8). The headline numbers are R ≈ 6.9 × 10^4 Gpc^-3 yr^-1 for non-repeaters at DMhost = 66.63 pc cm^-3 and R ≈ 630 Gpc^-3 yr^-1 for repeaters at DMhost = 64.46 pc cm^-3; the authors conclude that DMhost significantly affects the rates and compare with SN Ia, CCSN, magnetar, and GRB rates.","tokens_in":19949,"tokens_out":5692,"duration_ms":64191,"significance":"The compilation of localized host galaxies and the use of a non-Gaussian IllustrisTNG-based DMIGM distribution are useful, and the comparisons with Ravi (2019) and Shin et al. (2023) provide sanity checks. If the rate method were fully specified, the resulting local rates would be a valuable point of comparison for progenitor models. However, the central claim that DMhost significantly affects volumetric rates is currently a consequence of the Vlimit construction in Section 2.4 rather than an empirical measurement, and the missing algorithm details prevent the reader from assessing the magnitude of the effect. The MCMC host-DM result also lacks prior and likelihood details. The paper is promising but needs substantial clarification and reframing.","major_comments":[{"comment":"The Vlimit calculation is not specified sufficiently for reproduction. The text states that the FRB with the highest DMex is assigned P(≤dlimit|DMex) ≈ 0.95 and all others ≈ 1, but it does not give the functional form of P, the parameters of Eq. (5), the value of dlimit or Vlimit, or the sensitivity to the 95% quantile. Because Eq. (8) scales all rates inversely with Vlimit, every rate in Sections 3.2 and 3.3 depends on this single order statistic. Please provide the full algorithm, the adopted parameter values from Zhang et al. (2021), and a robustness test (e.g., 68% vs 99% quantile, or a full likelihood approach).","section":"Section 2.4, Eq. (8)"},{"comment":"The abstract and Section 4 state that DMhost significantly affects rates, but this is a direct consequence of substituting assumed DMhost values into Eq. (8): increasing DMhost decreases the allowed DMIGM, hence dlimit and Vlimit, and increases R by construction. The paper does not identify an independent constraint that would make this an empirical claim. I recommend reframing these curves as a sensitivity study of a model-dependent estimator and showing whether the differences survive when the uncertainty in the DMhost distribution is propagated jointly rather than by point substitutions.","section":"Sections 3.1–3.3, Figure 6"},{"comment":"The MCMC inference of DMhost is underspecified: no priors for µhost and σhost are given, the likelihood (presumably Eq. (5) plus a log-normal host term) is not written down, and only sources with positive inferred DMhost are retained. Dropping negative or zero DMhost values biases the medians upward, and the selection of sources from Table 1 is not described. Without these details the conclusion of no significant difference between repeaters and non-repeaters cannot be evaluated.","section":"Section 2.3, Figure 2"},{"comment":"The repeater rate estimate applies ε = 0.468, a detection efficiency derived from non-repeater injection tests in CHIME/FRB Catalog 1, while the repeater sample is drawn from Blinkverse over a 4-year exposure and counts each source once. Equation (8) as written combines a per-burst efficiency with a source count, so the repeater rate is likely biased. Please derive an appropriate source-detection efficiency and effective exposure for repeaters, or state explicitly why the non-repeater efficiency applies.","section":"Section 3.3, Eq. (8)"},{"comment":"The rates for z < 0.2 are set by the low-redshift tail of the adopted DMIGM distribution, but the parameters A, α, β, σDM, and C0 from Zhang et al. (2021) are never listed, and Figure 1 shows that the choice of distribution changes distance estimates. If the IllustrisTNG distribution is misspecified at low redshift, the resulting shift in Vlimit could be comparable to the claimed DMhost effect. Please report the parameter values, show the behavior of Eq. (5) at low z, and quantify the systematic uncertainty.","section":"Section 2.2, Eq. (5)"}],"minor_comments":[{"comment":"The title contains a typo: 'F ast Radio Burst' should read 'Fast Radio Burst'.","section":"Title, page 2"},{"comment":"The text states that observations at 'low radio frequencies (∼100 GHz)' can constrain the local environment; the unit should presumably be MHz, not GHz.","section":"Section 2.3"},{"comment":"FRB 20200120E is excluded from the repeater rate estimate but the sample-selection section does not justify this exclusion; please state the reason in Section 2.5.","section":"Section 3.3"},{"comment":"The text mixes DMhost and DMsource: Eq. (3) lists them separately, but Section 2.3 says they are combined for the analysis; please clarify how DMsource is handled in Eqs. (4) and (7).","section":"Section 2.3, Eq. (3)"},{"comment":"Several figure captions are informal and do not fully define the shaded regions or the meaning of the vertical and horizontal lines; please make the captions self-contained.","section":"Figures 5–7"},{"comment":"The compiled Table 1 is valuable, but no machine-readable version or data availability statement is provided; please include a link to the table in electronic form.","section":"Appendix A"}],"recommendation":"major_revision","confidential_remarks":"The paper sits on the borderline between major revision and rejection. The central 'DMhost significantly affects rates' claim is not an empirical result as presented; it follows from the assumed Vlimit construction. However, the host-galaxy compilation and the comparative rate discussion are useful, and the issues are fixable by specifying the method, adding parameter values, and reframing the claims. I do not see evidence of deliberate misrepresentation, but the current level of detail is below what a reader needs to reproduce the rates."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. First, the paper's real contribution is a clean compilation and an honest MCMC comparison: on the current localized sample, repeater and non-repeater DMhost medians are statistically indistinguishable, and the authors say so without overclaiming. Second, the rate-density analysis is a direct extension of Ravi (2019) and Zhang et al. (2020a,b), and the central 'DMhost significantly affects rates' result is largely baked into the method. Equation (8) divides by Vlimit, and Vlimit is set by a single high-DM source at the 95% quantile of the adopted DM_IGM distribution. The paper does not report dlimit or Vlimit values, nor the sensitivity to the quantile choice, so the claimed factor-of-ten rate swing with DMhost is not an independent empirical measurement. That does not make the paper worthless; it makes the rate numbers conditional on the DM_IGM model and the boundary construction.\n\nWhat is actually good: the host galaxy table in Appendix A is a useful inventory; the MCMC analysis with bilby is reproducible in principle; the rate comparison with SN Ia, CCSN, magnetar, and GRB rates is informative. The authors are also explicit that no origin model is excluded, which is the right degree of humility given the uncertainties.\n\nSoft spots, in proportion: the MCMC priors are not specified, which matters because the log-normal prior on DMhost directly shapes the posterior; only sources with positive inferred DMhost are kept, which biases the median upward; and the repeater rate uses the non-repeater detection efficiency and a roughly estimated observing time. None of these is fatal on its own, but together they weaken the quantitative claims. The stress-test note is correct and should be taken seriously: replacing the single-order-statistic boundary with a full likelihood could shift the rates by an amount comparable to the claimed DMhost effect. The authors should at least report dlimit and Vlimit values and show how the rates change under different quantiles or a full likelihood treatment.\n\nWho is this for: FRB phenomenologists and rate modelers who want a current inventory of localized hosts and a set of local rate estimates under stated assumptions. It deserves peer review because the questions are relevant and the analysis is checkable, but it needs a revision that documents the Vlimit calculation and tests its sensitivity. I would not cite the rate numbers in my own work until that revision lands; the host table and the MCMC comparison are the more durable parts.","headline":"A useful, honest compilation with a modest MCMC result, but the central rate-versus-DMhost claim is largely built into an under-specified Vlimit boundary and needs a careful revision before the numbers can be taken at face value.","tokens_in":20495,"tokens_out":1559,"would_cite":false,"duration_ms":19429,"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 paper argues that the unknown ionized-gas content of fast radio burst host galaxies is a dominant uncertainty in volumetric rate estimates, and that repeaters and non-repeaters currently show indistinguishable host-gas contributions.","keywords":["fast radio bursts","dispersion measure","host galaxies","volumetric event rate density","repeaters","non-repeaters","FRB progenitors","radio transients"],"falsifier":"A falsifying check would be to compare the assumed intergalactic DM distribution with a sample of FRBs that have measured redshifts and host galaxies at $z<0.2$. If the observed distribution of extragalactic DM minus the inferred host contribution has a different median or tail than the simulation-based model, the limiting-volume calculation and the quoted local rates would have to be revised by an amount comparable to the host-DM effect.","tokens_in":19428,"feed_emoji":"📡","tokens_out":13631,"duration_ms":115460,"temperature":0.7,"pith_summary":"This paper argues that the amount of ionized gas in fast radio burst (FRB) host galaxies is a first-order uncertainty in measuring how often these bursts occur per cosmic volume, and that this uncertainty directly affects which stellar-death and neutron-star models can match the data. The paper collects the currently localized FRB sample, fits a log-normal host-galaxy gas contribution to repeaters and non-repeaters, and finds no significant difference between their medians (66.63 vs 64.46 pc cm$^{-3}$). It then uses a survey catalog from a drift-scan radio telescope to derive local volumetric rates of about $6.9\\times10^4$ Gpc$^{-3}$ yr$^{-1}$ for non-repeaters and about $630$ Gpc$^{-3}$ yr$^{-1}$ for repeaters at those medians, and shows that the rates swing by an order of magnitude as the assumed host contribution changes. If these estimates hold, future rate comparisons with supernovae, magnetars, and gamma-ray bursts must treat host-galaxy gas as a dominant systematic, not a minor correction.","feed_headline":"Host-galaxy gas can change FRB rate estimates tenfold","feed_subtitle":"How much gas sits in FRB host galaxies decides whether rates match supernovae, magnetars, or GRBs.","key_machinery":"The mechanical core is the dispersion-measure (DM) budget, the relation that divides an FRB's measured signal delay into Milky Way, halo, intergalactic, and host-galaxy contributions. The paper treats the host contribution as a log-normal random variable, fits its median to 35 non-repeaters and 14 repeaters with Markov-chain Monte Carlo, and then uses the simulation-calibrated probability distribution $P(\\Delta)$ of Eq. (5) to convert each remaining extragalactic DM into a maximum comoving distance. That distance sets $V_{\\rm limit}$ in the volumetric-rate formula $R = N/(\\epsilon V_{\\rm limit}\\Omega_{\\rm sky}t_{\\rm obs}f_b)$, so every assumption about the IGM or host DM propagates directly into the rates.","core_discovery":"On its own terms, the paper's discovery is that the host-galaxy term in the dispersion-measure budget, not just the intergalactic term, controls what volumetric rates imply about FRB origins. The Markov-chain Monte Carlo fit to 35 localized non-repeaters and 14 localized repeaters yields overlapping host DM medians, so the paper concludes that repeater and non-repeater environments cannot be distinguished with present data. It then shows that changing the assumed $\\mathrm{DM_{host}}$ from zero to values typical of galaxies raises the inferred non-repeater local rate by about an order of magnitude; at the fitted medians the rates are $R\\approx6.9\\times10^4$ Gpc$^{-3}$ yr$^{-1}$ (non-repeaters) and $R\\approx630$ Gpc$^{-3}$ yr$^{-1}$ (repeaters). These numbers are the paper's quantitative case that $\\mathrm{DM_{host}}$ significantly affects volumetric rates.","pith_inferences":["An untested consequence is that if the intergalactic DM model is recalibrated with localized FRBs at known redshifts, the local rates may shift by as much as the host-DM effect, which would blur the paper's central contrast.","A natural next step, not taken here, is to reclassify FRBs by host-galaxy properties or local environment rather than by repetition, since the similar $\\mathrm{DM_{host}}$ medians suggest repetition status may not track the surrounding gas.","A testable extension is to run injection simulations tailored to repeater exposure to check whether the single detection efficiency assumed for both populations changes the repeater rate outside its quoted Poisson errors."],"forward_implications":["The derived local non-repeater rate of about $6.9\\times10^4$ Gpc$^{-3}$ yr$^{-1}$ sits near the Type Ia supernova rate and the magnetar upper limit, so neither model is excluded at the fiducial host DM.","At higher assumed host DM, the non-repeater rate exceeds the magnetar and soft-gamma-repeater upper limits, so the host correction determines whether those progenitor channels remain viable.","The repeater rate of about $630$ Gpc$^{-3}$ yr$^{-1}$ is consistent with long and short gamma-ray burst rates within the quoted uncertainties, leaving both magnetar and merger interpretations open.","Because the median host DM is similar for repeaters and non-repeaters, the data do not support a clean environmental separation between the two populations, though small samples and orientation effects could mask a real difference."],"supporting_citations":[{"why":"Supplies the baseline volumetric-rate method and the characteristic host DM values used for comparison.","marker":"Ravi (2019)"},{"why":"Provides the rate-estimation approach the paper adapts for repeaters from merger scenarios.","marker":"Zhang et al. (2020a)"},{"why":"Supplies the simulation-calibrated parameters for the intergalactic DM distribution that set the limiting volume.","marker":"Zhang et al. (2021)"},{"why":"Gives the analytic form of the intergalactic DM probability distribution used as Eq. (5).","marker":"Macquart et al. (2020)"},{"why":"Introduces the log-normal host-DM model used in the Markov-chain Monte Carlo fits.","marker":"Zhang et al. (2020b)"},{"why":"Contributes the non-repeater catalog, effective exposure time, and detection efficiency.","marker":"CHIME/FRB Collaboration et al. (2021)"},{"why":"Provides the bilby package used for the Markov-chain Monte Carlo host-DM analysis.","marker":"Ashton et al. (2019)"},{"why":"Offers the comparison non-repeater rate whose consistency the paper checks.","marker":"Shin et al. (2023)"}],"fun_headline_variants":["Host-galaxy gas shifts FRB rates tenfold","FRB rate estimates hinge on host-galaxy gas","Host gas is the wildcard in FRB rate estimates"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The estimates rest on assuming that the adopted probability model for how much signal delay intergalactic electrons add, taken from a cosmological simulation, is correct for every line of sight; if that model is wrong, the inferred distances and rates shift by as much as the host-galaxy effect the paper highlights.","fun_headline_variants_meta":{"raw":{"variants":["Host-galaxy gas shifts FRB rates tenfold","FRB rate estimates hinge on host-galaxy gas","Host gas is the wildcard in FRB rate estimates"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000498,"raw_usage":{"total_tokens":2446,"prompt_tokens":955,"completion_tokens":1491,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":571,"completion_tokens_details":{"reasoning_tokens":1440}},"tokens_in":571,"tokens_out":1491,"duration_ms":37342,"temperature":1.0,"reasoning_tokens":1440,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T20:56:18.111050+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A falsifying check would be to compare the assumed intergalactic DM distribution with a sample of FRBs that have measured redshifts and host galaxies at $z<0.2$. If the observed distribution of extragalactic DM minus the inferred host contribution has a different median or tail than the simulation-based model, the limiting-volume calculation and the quoted local rates would have to be revised by an amount comparable to the host-DM effect.","supporting_citations":[],"review_version":1}