{"id":"359d8114-e9ce-49f6-9e6e-dae6f4fc937d","arxiv_id":"2506.17568","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"The paper forecasts that future gravitational wave observatories could infer dark matter mass and annihilation cross-section from early matter-domination suppressions in the inflationary gravitational wave spectrum, complementing indirect detection.","lead":"Dark matter that is made by a long-lived particle during a brief early matter-dominated phase leaves a dip in the primordial gravitational wave spectrum from inflation. The authors forecast that LISA, ET, BBO, and mu-ARES could measure the dark matter mass and annihilation rate from that dip, overlapping with future gamma-ray and neutrino telescopes.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Core mapping invalid: Eqs. (2.4)-(2.5) give Tdom≈2M_N, so Tf≈Mχ/20>Tdom implies Mχ>40M_N, contradicting M_N>Mχ; benchmark freeze-out history not realized.","rationale":"The reader's weakest assumption was the benchmark n_T/r; that is a legitimate generality concern but secondary to a possible internal inconsistency in the EMD/DM mapping. The derivation of Eq. (2.12) begins with 'Tf>Tdom', yet the paper's own Tdom formula makes this impossible whenever M_N>Mχ. This is not a matter of current versus future bounds but of the model's self-consistency; if confirmed, the headline claims (e.g., ~1% ET uncertainties) do not follow even for n_T=0.5 and r=0.036. I agree with the reader's call for conditions (e.g., clearly stating benchmark assumptions), but the more load-bearing issue is the frozen-out-before-EMD prerequisite. Credit is due for the transparent derivation and the use of established transfer functions; the concern is internal, not a disagreement with consensus. The proposed test is a simple algebraic evaluation plus a rerun of the relic-density calculation without the Tf>Tdom shortcut. Given that the submitted version's central parameter mapping appears to fail for the quoted benchmarks, the verdict should move to REJECT unless the Tdom inconsistency is resolved and the forecasts are recomputed.","tokens_in":29530,"tokens_out":19296,"duration_ms":200676,"concrete_test":"Re-evaluate the benchmark (M_N=10^5 GeV, Mχ=10^5 GeV, ⟨σv⟩=10^-24 cm^3/s) of Figs. 3-6: compute Tdom from Eqs. (2.4)-(2.5) and compare with Tf≈Mχ/20=5×10^3 GeV. If Tdom≈2×10^5 GeV>Tf, then freeze-out did not occur before EMD; redo the relic-density calculation with freeze-out in the EMD-modified Hubble rate and check which, if any, points in Fig. 4 satisfy both M_N>Mχ and Tf>Tdom. If no headline benchmark survives, recompute the SNR contours, Fisher errors, and complementarity overlaps, and state explicitly which initial N abundance (g_N and dilution) Eq. (2.4) assumes.","verdict_should_be":"REJECT","load_bearing_attack":"The paper's DM reconstruction requires freeze-out before EMD: 'We consider, in this work, the scenario where Tf > Tdom' (Sec. 2.2), with Tf≈Mχ/20. However, combining Eq. (2.4), Hdom≈4[g*(Tdom)/g*(M_N)]H(M_N), with Eq. (2.5), Tdom^3≈[g_s(M_N)/g_s(Tdom)]M_N^3(Hdom/H(M_N))^{3/2}, gives Tdom≈2M_N for g*≈g_s. Then Tf>Tdom becomes Mχ/20>2M_N, i.e. Mχ>40M_N, which directly contradicts the stated requirement M_N>Mχ (needed for N→χ decay). For the headline benchmark M_N=10^5 GeV (Figs. 3-6), Tdom≈2×10^5 GeV whereas Tf≤5×10^3 GeV for all Mχ≤10^5 GeV, so every plotted point violates the assumed thermal history. If instead one uses the physical onset for a thermally populated single-species N, Tdom∼(g_N/g*)M_N≈10^-2M_N, freeze-out before EMD holds only for Mχ≳20(g_N/g*)M_N, and the viable region shrinks accordingly. Eq. (2.4) as written is thus inconsistent with the thermalization premise of Sec. 2.1 and with the footnote deferring to Ref. [101]. Since Eq. (2.12) and all SNR/Fisher maps depend on this mapping, the central claim that GW missions determine (Mχ,⟨σv⟩) is unsupported for the benchmarks quoted.","agreement_with_reader":"disagree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes that a period of early matter domination (EMD) generated by a long-lived SM-singlet particle N leaves a characteristic frequency-dependent suppression in the inflationary primordial gravitational wave (PGW) spectrum, and that the location and depth of this feature can be used to infer the dark matter mass M_chi and annihilation cross-section <sigma v>. The authors derive a relation between the DM parameters and the N decay temperature from the observed relic abundance, then compute signal-to-noise ratios and Fisher-matrix forecasts for mu-ARES, LISA, BBO, and ET. They overlay these GW sensitivities with current and projected indirect-detection constraints from gamma-ray (HESS, Fermi-LAT, CTA) and neutrino (ANTARES, KM3NeT) telescopes, concluding that GW missions and indirect searches probe complementary and overlapping regions of the (M_chi, <sigma v>) plane, with percent-level precision on selected benchmarks.","tokens_in":29885,"tokens_out":11984,"duration_ms":122222,"significance":"If the framework is sound, the proposal is a genuinely new cosmological probe of WIMP-like DM parameters, complementary to indirect detection and potentially relevant for both the GW and particle-astrophysics communities. The paper uses standard Boltzmann equations, a well-established PGW transfer-function formalism, and transparent detector noise models, and it gives explicit numerical forecasts with clearly specified benchmark points. The main quantitative claims are, however, conditional on two strong assumptions: a blue-tilted tensor spectrum (n_T = 0.5) with the maximum currently allowed tensor-to-scalar ratio (r = 0.036), and a particular mapping between the EMD epoch and DM freeze-out. As discussed below, the latter is internally inconsistent for the benchmarks plotted, and the former is not reflected in the abstract's broad statement of detection potential.","major_comments":[{"comment":"The assumed thermal history is internally inconsistent. Combining Eqs. (2.4) and (2.5) with g_* ~ g_*s gives T_dom ~ 2 M_N. Then the freeze-out-before-EMD condition T_f > T_dom, with T_f ~ M_chi/20, requires M_chi > 40 M_N, which directly contradicts the stated requirement M_N > M_chi needed for N -> chi decay. For the benchmark M_N = 10^5 GeV used in Figs. 3-6, T_dom ~ 2 x 10^5 GeV while T_f <= 5 x 10^3 GeV for all M_chi <= 10^5 GeV, so every plotted point violates the stated thermal history. For a thermally populated single-species N, the physical onset is instead T_dom ~ (g_N/g_*) M_N ~ 10^-2 M_N, which shrinks the viable region to M_chi >~ 0.2 M_N and excludes several quoted benchmarks, e.g., (M_chi, <sigma v>) = (10^4 GeV, 2 x 10^-25 cm^3/s) with M_N = 10^5 GeV. Because Eq. (2.12) and all subsequent SNR and Fisher results depend on this mapping, the central claim that GW missions determine (M_chi, <sigma v>) is not supported for the benchmarks as quoted. The T_dom relation or the thermalization premise should be corrected, and the parameter-space reach recomputed.","section":"Sec. 2.1-2.2, Eqs. (2.4), (2.5), (2.12)"},{"comment":"All detectability and precision statements are computed at the extremal values n_T = 0.5 and r = 0.036. The left panel of Fig. 2 shows that for n_T = 0 the EMD-modified GW spectrum lies below all detector sensitivities, and the conclusions do not state this strong dependence. If future CMB experiments lower the r bound, or if the single-field consistency relation n_T ~ -r/8 is imposed, the SNR > 10 regions shrink substantially or disappear. The abstract's broad claim of 'good potential' should be qualified by this explicit condition, or the reach should be presented as a function of (r, n_T).","section":"Sec. 3.1, Sec. 4, Figs. 2-6"},{"comment":"The quantitative benchmark values quoted in the abstract are not mutually consistent with the body of the paper. The abstract states that ET reaches ~1% on (M_chi, <sigma v>) = (10^5 GeV, 10^-24 cm^3/s) and mu-ARES ~7% on (10^4 GeV, 2 x 10^-25 cm^3/s); Sec. 5.2 states ET ~1% on (10^3 GeV, 10^-24 cm^3/s) and mu-ARES ~7% on (10^4 GeV, 10^-25 cm^3/s); Sec. 5.1 states mu-ARES ~7% for (10^5 GeV, 10^-24 cm^3/s); and Sec. 6 repeats the (10^3, 10^-24) + (10^4, 2 x 10^-25) pair. These numbers should be harmonized, and the selected benchmarks should be checked against the corrected EMD viability condition from Sec. 2.","section":"Abstract, Sec. 5.1, Sec. 5.2, Sec. 6"},{"comment":"The Fisher forecasts fix M_N = 10^5 GeV and T_RH = 10^11 GeV and vary only {M_chi, <sigma v>, n_T}. However, the EMD transfer function in Eqs. (3.9)-(3.15) depends directly on M_N and T_RH, and Fig. 2 demonstrates that varying these parameters changes the spectral shape in ways that can mimic changes in the DM parameters. The quoted percent-level uncertainties on (M_chi, <sigma v>) therefore assume exact knowledge of M_N and T_RH. Marginalizing over these parameters in the Fisher matrix, or at least demonstrating that the degeneracies are negligible, is needed to support the precision claims.","section":"Sec. 4, Eq. (4.5), Figs. 5-6"}],"minor_comments":[{"comment":"The figure legend includes detectors (THEIA, GAIA, ARES) that are not described in Table 1 or in the noise-model appendix; please either remove these or specify their sensitivity curves.","section":"Fig. 2"},{"comment":"The illustrative benchmark '(tau_N, M_N) = (0.1, 5000) (in GeV)' is dimensionally ambiguous: tau_N is presumably in seconds and M_N in GeV. Please correct the units.","section":"Sec. 6"},{"comment":"The text accompanying Fig. 5 says 'The results are shown for various GW observations, mu-ARES, LISA, BBO and LISA', with 'LISA' appearing twice; the second instance should name the fourth detector.","section":"Fig. 5 caption/text"},{"comment":"There are several typographical slips, including 'the the origin' in Sec. 1 and 'SRN' for 'SNR' in Sec. 5.3; a careful proofreading pass is recommended.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The issue raised in major comment 1 is not a matter of taste: if confirmed, it invalidates most of the plotted parameter space as currently presented, because the quoted benchmarks violate the freeze-out-before-EMD condition under the paper's own Eqs. (2.4)-(2.5). The manuscript also cites the same authors' earlier framework very heavily (e.g., Refs. [7,10,11,93-101]); a more balanced comparison with independent EMD-plus-GW analyses would strengthen the novelty claim. I recommend that the revised manuscript be re-refereed after the thermal-history mapping is corrected and the Fisher forecasts are recomputed on the resulting viable parameter region."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Biggest thing to know: the central mapping that turns EMD parameters into (M_ch, sigma_v) is internally inconsistent. Combining the paper's own Eqs. (2.4) and (2.5) with g* ~ g_s gives Tdom ~ 2 M_N. Their scenario requires Tf ~ M_ch/20 > Tdom (freeze-out before EMD). That forces M_ch > 40 M_N, which cannot coexist with M_N > M_ch (needed for N -> ch decay). With M_N = 10^5 GeV, Tdom ~ 2x10^5 GeV while every plotted M_ch <= 10^5 GeV has Tf <= 5x10^3 GeV, so every benchmark in Figs. 3-6 violates the assumed thermal history. A consistent estimate for a thermally populated single-species N gives Tdom ~ (g_N/g*) M_N ~ 10^-2 M_N, which would restore a viable window, but the numbers as quoted would not survive. This is load-bearing: Eq. (2.12) and all the SNR and Fisher maps follow from the mapping.\n\nWhat the paper does well: the Fisher machinery is standard and cleanly applied; the transfer functions are the usual ones; the critical branching-ratio condition in Eq. (2.13) is a genuine new element; and the overlay with CTA, ANTARES, and KM3NeT projections is a useful framing for complementarity. The figures are readable and the parameter choices are mostly transparent. The abstract, however, does not say that every reach number assumes n_T = 0.5 and r = 0.036 at the current upper bound; Fig. 2 left shows that n_T = 0 gives no detectable signal. That is a reporting gap rather than an error. There is also unresolved overlap with the same authors' ref. [11], and the indirect-detection limits are pulled from a github repository without a version pin.\n\nWho this is for: phenomenologists working on PGW backgrounds and non-standard pre-BBN histories. The machinery is reusable, but the headline numbers should not be cited until the EMD onset is treated consistently. My recommendation: send to peer review, but with an explicit request to reconcile Eqs. (2.4)-(2.5) with the Tf > Tdom condition, or to restrict to the parameter region where freeze-out actually precedes EMD. It is fixable, but as written, the quantitative claims do not stand. For a reading group it is a useful case study of a benchmark inconsistency quietly invalidating a forecast; otherwise, maybe.","headline":"The EMD-to-DM mapping at the core of this forecast is internally inconsistent—their own Tdom formulas make the required freeze-out-before-EMD history impossible for M_N > M_ch—so the quoted GW reach numbers rest on an invalid benchmark regime.","tokens_in":30499,"tokens_out":6476,"would_cite":false,"duration_ms":62275,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Upcoming gravitational-wave missions can measure dark matter mass and annihilation cross-section from the dip that an early matter-dominated epoch leaves in the inflationary gravitational-wave spectrum.","keywords":["primordial gravitational waves","early matter domination","dark matter indirect detection","non-thermal dark matter","Fisher forecast","stochastic gravitational wave background","tensor spectral index","complementarity"],"falsifier":"Measure the tensor spectral index and tensor-to-scalar ratio with next-generation CMB experiments: if n_T is found at or below zero (as slow-roll inflation predicts) or r is found well below 0.036, then the EMD-modified spectrum plotted in the paper falls below all detector noise curves and the claimed dark-matter reach is falsified; conversely, detection of the predicted frequency-dependent dip in LISA, µ-ARES, BBO, or ET data at the forecast frequencies would confirm it.","tokens_in":29274,"feed_emoji":"🌌","tokens_out":10675,"duration_ms":103646,"temperature":0.7,"pith_summary":"The paper sets out to show that the thermal history of the universe before Big Bang nucleosynthesis leaves a measurable mark on the gravitational waves produced during inflation, and that this mark can be used to weigh dark matter. In the specific scenario studied, a long-lived particle dominates the universe for a while (early matter domination) and its decay both dilutes and then replenishes the dark matter relic; the resulting frequency-dependent dip in the primordial gravitational-wave spectrum encodes when that era began and ended. Because the decay time is linked to the dark matter mass and annihilation cross-section by the relic-abundance condition, the dip position and width become a cosmological ruler for dark matter parameters. Using signal-to-noise and Fisher forecasts, the paper identifies benchmark points where future detectors—especially the Einstein Telescope and µ-ARES—can measure those parameters to roughly one and seven percent precision, overlapping the projected reach of gamma-ray and neutrino indirect searches.","feed_headline":"Dark matter's mass may hide in a gravitational-wave dip","feed_subtitle":"Future detectors could pin down dark matter mass and annihilation rate to a few percent through an early-universe dip.","key_machinery":"The load-bearing object is the EMD-modified tensor transfer function $F(k)|_{\\rm EMD} = T_1^2(k/k_{\\rm eq})\\,T_2^2(k/k_{\\rm dec})\\,T_3^2(k/k_{\\rm dec,S})\\,T_2^2(k/k_{\\rm RH,S})$, built from the standard fitting functions $T_1$, $T_2$, $T_3$. The two characteristic scales $k_{\\rm dec}$ and $k_{\\rm dec,S}$ mark, respectively, the end of the early matter era and the entropy-dilution period caused by the decay of N, so the frequency and width of the suppression encode $T_{\\rm dec}$ and the entropy dilution factor $\\Delta_s$. The second piece is the relic-abundance link $T_{\\rm dec} = (3\\times10^{-26}\\,{\\rm cm^3/s}/\\langle\\sigma v\\rangle_{\\rm ann})\\,T_f$ with $T_f \\approx M_\\chi/20$, which converts the dip into a constraint on the dark matter mass and annihilation cross-section; the critical branching ratio $Br^c_{N\\to\\chi}$ of Eq. (2.13) guarantees that enough non-thermal dark matter is produced for residual annihilation to set the final abundance.","core_discovery":"The paper's central claim is that an epoch of early matter domination, driven by a heavy metastable particle N that decays partly into dark matter, imprints a characteristic two-step suppression on the inflationary gravitational-wave spectrum: a high-frequency step from reheating and a lower-frequency dip set by the decay temperature $T_{\\rm dec}$. The transfer function for the tensor modes acquires an extra factor $T_3^2(k/k_{\\rm dec,S})$, and the dip's frequency and width are fixed by the comoving scales $k_{\\rm dec}$ and $k_{\\rm dec,S}$, which depend on $T_{\\rm dec}$ and the entropy dilution factor. The relic-abundance requirement then connects $T_{\\rm dec}$ to the dark matter parameters through $T_{\\rm dec} = (3\\times10^{-26}\\;{\\rm cm^3/s}/\\langle\\sigma v\\rangle_{\\rm ann})\\,T_f$, with $T_f \\approx M_\\chi/20$, so a measured dip translates directly into constraints on the dark matter mass and annihilation cross-section. On these grounds the authors forecast that LISA can reach masses from roughly 200 GeV to $10^5$ GeV, that the Einstein Telescope can measure a benchmark $(M_\\chi, \\langle\\sigma v\\rangle_{\\rm ann}) = (10^5\\,{\\rm GeV}, 10^{-24}\\,{\\rm cm^3/s})$ to about one percent, and that µ-ARES can measure $(10^4\\,{\\rm GeV}, 2\\times10^{-25}\\,{\\rm cm^3/s})$ to about seven percent; these regions overlap the projected sensitivities of CTA, ANTARES, and KM3NeT, defining the claimed complementarity.","pith_inferences":["If future CMB experiments tighten the tensor-to-scalar ratio below about 0.01 or exclude a blue tilt, the dark-matter reach quoted here would shrink proportionally, so the precision numbers are best read as upper limits set by the current maximal $r$.","The dip frequency and width are two independent observables determined by $T_{\\rm dec}$ and the entropy dilution, while the relic-abundance relation ties $T_{\\rm dec}$ to $M_\\chi/\\langle\\sigma v\\rangle$; a joint fit could separate mass from cross-section more cleanly than either observable alone.","The same EMD fingerprint should also appear in the scalar-induced secondary gravitational-wave background and in pulsar-timing bands if the epoch lasted long enough; detecting a dip there would corroborate the scenario before the space-based interferometer missions fly.","Fixing $M_N$ and $T_{\\rm RH}$, as the Fisher analysis does, likely overstates the precision; treating them as free parameters in a full Markov-chain Monte Carlo analysis would broaden the reported error bars."],"forward_implications":["For the benchmark choice $M_N = 10^5$ GeV, $n_T = 0.5$, and $r = 0.036$, LISA is sensitive to dark matter masses in the range $2\\times10^2$ to $10^5$ GeV with annihilation cross-sections near $10^{-26}$ to $4\\times10^{-24}$ cm$^3$/s.","The Einstein Telescope can measure a $10^5$ GeV dark matter candidate with cross-section $10^{-24}$ cm$^3$/s to about 1% precision, and a $10^5$ GeV candidate with cross-section $10^{-23}$ cm$^3$/s to similar precision, the latter lying inside the ANTARES and KM3NeT projections.","µ-ARES can measure a $10^4$ GeV candidate with cross-section $2\\times10^{-25}$ cm$^3$/s to about 7% precision, a point that lies within the projected CTA reach.","The overlapping sensitivity regions mean the same dark matter candidate can be probed both by gravitational-wave observatories and by gamma-ray and neutrino indirect searches, providing an independent cosmological test.","If the tensor spectrum is scale-invariant ($n_T = 0$) at the maximal $r = 0.036$, the EMD-modified spectrum lies below all detector sensitivities, so the quoted reach is conditional on a blue-tilted spectrum."],"supporting_citations":[{"why":"Defines the onset and end of early matter domination from long-lived particles, including the entropy dilution factor used throughout.","marker":"[43]"},{"why":"Supplies the non-thermal dark matter production from long-lived particle decay and the residual-annihilation condition that yields the critical branching ratio.","marker":"[44]"},{"why":"Provides the transfer functions T1, T2, T3 whose fitting formulas build the EMD-modified gravitational-wave spectrum.","marker":"[137]"},{"why":"Establishes the formalism for primordial gravitational-wave spectra in modified cosmologies, the basis for the spectral-shape analysis.","marker":"[8]"},{"why":"Sets the maximum tensor-to-scalar ratio r=0.036 adopted for the benchmark forecasts.","marker":"[108]"},{"why":"Supplies Planck 2018 cosmological parameters, the relic-density constraint, and the current Delta N_eff bounds.","marker":"[6]"},{"why":"Provides the mu-ARES noise curve used in the SNR and Fisher forecast computations.","marker":"[147]"},{"why":"Supplies the projected CTA sensitivity that defines the gamma-ray complementarity region.","marker":"[185]"},{"why":"Supplies the ANTARES projected sensitivity for neutrino-based indirect dark matter searches.","marker":"[186]"},{"why":"Supplies the KM3NeT projected sensitivity for neutrino-based indirect dark matter searches.","marker":"[188]"}],"fun_headline_variants":["Dark matter's gravitational-wave dip reveals its mass","GW dip pins down dark matter mass and annihilation","Gravitational waves complement indirect dark matter probes","Early matter domination's GW dip probes dark matter"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The projected reach rests on assuming the inflationary tensor spectrum is blue-tilted with index 0.5 and has the maximum amplitude currently allowed (r = 0.036); if either is smaller, the signal falls below the detectors' sensitivity.","fun_headline_variants_meta":{"raw":{"variants":["Dark matter's gravitational-wave dip reveals its mass","GW dip pins down dark matter mass and annihilation","Gravitational waves complement indirect dark matter probes","Early matter domination's GW dip probes dark matter"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001209,"raw_usage":{"total_tokens":5132,"prompt_tokens":1250,"completion_tokens":3882,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":866,"completion_tokens_details":{"reasoning_tokens":3823}},"tokens_in":866,"tokens_out":3882,"duration_ms":28639,"temperature":1.0,"reasoning_tokens":3823,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T19:07:19.346347+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the tensor spectral index and tensor-to-scalar ratio with next-generation CMB experiments: if n_T is found at or below zero (as slow-roll inflation predicts) or r is found well below 0.036, then the EMD-modified spectrum plotted in the paper falls below all detector noise curves and the claimed dark-matter reach is falsified; conversely, detection of the predicted frequency-dependent dip in LISA, µ-ARES, BBO, or ET data at the forecast frequencies would confirm it.","supporting_citations":[],"review_version":2}