{"id":"6c050a0f-3c9c-4788-90e7-0c543f9948ad","arxiv_id":"astro-ph/0403164","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"WMAP data alone set a lower bound on the lifetime of a decaying cold dark matter particle of about 123 Gyr at 68% confidence, with no evidence favoring decay.","lead":"The paper computes how a decaying component of cold dark matter would alter the cosmic microwave background, then fits that model to WMAP satellite data. The data show no preference for decay, so the fit becomes a lower bound on how long such particles must live.","discovery_kind":"new_application","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The flatness caveat is real but secondary; the more load-bearing soft spot is that the daughter-radiation fluid truncation at l=1 is asserted, not validated, and it directly controls the low-l/LISW part of the spectrum where the decay signal lives.","rationale":"The reader's verdict (CONDITIONAL, moderate confidence, medium correctness risk) is broadly right, and I would not move it. I agree with the reader that the derivation is internally consistent: Eq. (6) does reduce to the standard Euler equation, Eq. (14) is a recognizable hierarchy with a decay damping term, and the data are external so there is no circularity. I also agree that the flatness assumption is a genuine caveat, and the paper states it explicitly ('Throughout this paper we concentrate our attention on a flat universe'), so it is an acknowledged limitation rather than a hidden one. Where I differ is on which assumption is most load-bearing. The reader picks Omega_k = 0 because curvature and the decay peak-shift are partly degenerate. That is a fair point, but the paper's own Fig. 1 narrative says the low-l constraint is driven substantially by the late ISW enhancement, and the ISW is a large-scale effect that curvature also affects. So the flatness caveat and the daughter-fluid truncation are not independent: both bear on the same low-l part of the spectrum. The truncation concern is the more technical and the more checkable of the two, and it is not acknowledged anywhere in the text. The paper's claim that future experiments will not improve the bound ('errors in these scales are already limited by cosmic variance') is also asserted rather than demonstrated, but that is a forward-looking remark, not part of the central numerical claim, so I do not weight it heavily. On balance the central claim is plausible and the derivation is careful; the honest position is that the bound stands as quoted subject to two unquantified systematics (flatness and the l>=2 daughter closure), which is exactly what CONDITIONAL captures. I therefore recommend UNCHANGED and mark partial agreement with the reader, since we identify the same class of weakness but rank the specific assumptions differently.","tokens_in":26783,"tokens_out":99302,"duration_ms":1820235,"concrete_test":"Re-run the MCMC for the single-component DCDM case with the daughter-radiation hierarchy truncated at l_max = 2, 3, and 6 (instead of the fluid l=1 closure), holding all other settings fixed. If the marginalized 95.4% lower limit on Gamma^{-1} moves by less than ~10% between l_max=2 and l_max=6, the fluid truncation is validated and the concern does not land; if it moves by more, the quoted 52 Gyr bound is sensitive to an unvalidated approximation and should be quoted with that systematic attached.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is a one-sided lifetime bound from WMAP (abstract:s2, s3) plus the two-component generalization (abstract:s4). For this to hold, the modified CAMB must correctly capture how decay reshapes the CMB. The authors themselves identify three channels: peak shift to smaller l, reduced first-peak amplitude, and enhanced late ISW (Fig. 1 discussion). The third channel is the one that most directly drives the low-l constraint, and it is precisely where the treatment is weakest. In Eq. (13) the collision term is kept only through the l=0 and l=1 moments; for l>=2 the source is dropped, giving Eq. (14), a standard hierarchy with an extra damping term a*Gamma*(rho_DDM/rho_DR). The justification offered is that the DDM perturbation has only monopole and dipole. That is true for the source, but the daughter radiation's own higher moments are not sourced only by the decay term: they are sourced by the standard gravitational coupling to the l-1 and l+1 moments, and the decay damping term multiplies M_l itself. The claim that truncating the decay source at l=1 is adequate is therefore an approximation about the relative size of the decay-induced versus gravitational-induced M_2, not a consequence of the DDM having only l=0,1. The paper does not show a convergence test in l_max or a comparison against a full Boltzmann treatment of the daughter species. Since the bound is one-sided and driven by the same low-l region that the ISW enhancement populates, an error in the l>=2 daughter moments could shift the 95.4% limit. This is a correctness-risk concern, not a circularity or soundness one: the data are external and the derivation is otherwise careful.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The paper reformulates cosmological perturbation theory for decaying cold dark matter (DCDM) with relativistic daughter particles, implements the equations in a modified CAMB, and compares the predicted CMB angular power spectrum with WMAP first-year data. The main results are a one-sided lower bound on the lifetime, Γ^{-1} > 123 Gyr at 68% C.L. (52 Gyr at 95.4% C.L.) when all CDM decays, and a two-component generalization, Ω_DDM h^2 < -0.5(Γ^{-1}/1Gyr)^{-1} + 0.12 for Γ^{-1} > 5 Gyr at 95.4% C.L. The authors argue that the bounds are channel-independent because they rely only on gravitational effects.","tokens_in":9188,"tokens_out":20153,"duration_ms":370029,"significance":"If correct, these are useful, channel-independent constraints on dark matter lifetime from CMB alone. The perturbation formulation (Eqs. 5-14) is a genuine contribution, and the MCMC analysis is standard. The paper is a solid, if modest, contribution. The main caveats are the flatness assumption and the fluid approximation for the source term.","major_comments":[{"comment":"The bound is computed under an assumed flat universe. This is load-bearing because the decay signature includes a shift of the acoustic peaks toward larger angular scales, which is also how spatial curvature shifts the peaks. With Ω_k free, part of the decay signal could be absorbed by curvature, and the quoted Γ^{-1} limit would weaken. The paper does not quantify this degeneracy. A run with Ω_k free, or at least a quantitative estimate of the curvature degeneracy, would strengthen the central claim.","section":"Introduction, 'Throughout this paper we concentrate our attention on a flat universe'; Fig. 1 discussion","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The source term for the daughter radiation is computed in a fluid approximation, replacing the full q-integral in Eq. (13) with the DDM density and velocity perturbations. The paper states that this is 'complicated to compute in general' but does not validate the approximation. Since the source term controls the decay signature, a comparison with the full integral, or an argument that the dipole contribution is suppressed by v/c, would strengthen the bound.","section":"Eqs. (9)-(10) and the discussion following Eq. (13)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The concern that the l≥2 truncation of the decay source is an uncontrolled approximation does not land on reading the paper: the source term in Eq. (13) is proportional to f_DDM0 Ψ_DDM, which has only l=0 and l=1 moments, so the l≥2 source vanishes identically; the extra term in Eq. (14) is a background-evolution damping, not a source. However, the paper does not show a convergence test in l_max, which would be a useful validation.","section":"Eqs. (13)-(14)","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"Typo: 'dose' should be 'does' in 'the parameter of the DCDM model, Γ, dose not degenerate'.","section":"Discussion of Fig. 2","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The abstract uses 'Γ^{-1} > 123 Gyr' while the text uses 'Γ^{-1} ≥ 123 Gyr'. The inequality should be consistent.","section":"Abstract vs. text","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The DCDM curves are computed with all other parameters fixed to WMAP optimal values. The caption should clarify that these are not best-fit DCDM models.","section":"Fig. 1 caption","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"No Δχ^2 or likelihood ratio is reported for the DCDM model vs ΛCDM. A quantitative comparison would support the statement that 'we do not find any signal to prefer the DCDM model'.","section":"Discussion of Fig. 3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The self-citation to 'Mon. Not. R. Astr. Soc., in press (astro-ph/0312358)' should be updated with the final reference.","section":"Ref. [3]","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The claim that 'bounds on DCDM would not be improved so much by future experiments since DCDM affects CMB fluctuation only in rather large scales (see Fig. 1)' is inconsistent with Fig. 1, which shows differences at l ~ 200-1000 (peak positions and amplitudes), not just at low l. Future experiments would improve the measurement of these scales.","section":"Discussion of future experiments","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The collision term assumes a smooth injection spectrum proportional to f_DDM0/q, whereas a 2-body decay of a non-relativistic particle gives a delta-function injection at q = m_DDM/2. The paper should justify this choice or show that the bound is insensitive to the injection spectrum.","section":"Eq. (12)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The paper should clarify the units of Γ and H in Eq. (3), since Γ is in physical time and H is the conformal Hubble parameter.","section":"Eq. (3)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The claim that these are 'the tightest constraints on DCDM cosmology from CMB anisotropies' should be supported by comparison with other CMB constraints on DCDM.","section":"Conclusion","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision"},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Bottom line: this is a careful, modest paper that produces a quotable, channel-independent lifetime bound from WMAP1, and the perturbation derivation holds up. The one caveat worth pressing is flatness, which the authors state up front rather than hide. What is new here is real but bounded: they write down the background and perturbation equations for a decaying CDM component plus relativistic daughter (Eqs. 1–14), implement them in CAMB, and run a seven-parameter MCMC against WMAP1. The prior decaying-neutrino work (Kaplinghat et al.; Bharadwaj–Sethi) is cited, and the difference is genuine: CDM decay with both abundance and lifetime free. The result — Γ⁻¹ > 123 Gyr at 68%, > 52 Gyr at 95.4%, plus the two-component fit — is a real limit, and they are honest that WMAP shows no preference for decay. On the stress-test note: I do not think the l=1 truncation concern lands. The decay source term in Eq. (13) feeds only the monopole and dipole because the DDM perturbation has only those moments; that is exact for a CDM fluid, not an approximation. For l≥2 the paper keeps the full standard hierarchy (Eq. 14) with the extra damping term aΓ(ρ_DDM/ρ_DR)M_l, which follows from the first right-hand term of Eq. (13) and multiplies every moment. So the daughter radiation is not truncated at l=1; only the source is, and correctly. A convergence test in l_max would be nice, but I do not see a correctness risk there. The flatness caveat is the real one. The decay signature includes a peak shift toward larger scales, which is also how Ω_k shifts the peaks; with curvature free, part of the signal could be absorbed and the bound would weaken. The authors say they assume flatness, so it is acknowledged, not concealed — but they do not quantify the degeneracy, and that is the number I would want. Smaller things: the modified CAMB and chains are not released, so reproducing the exact numbers means reimplementation; the claim that future experiments will not improve much rests on a cosmic-variance argument that is plausible but asserted rather than shown; and the two-component bound is a fitting formula, not derived. Who should read it: anyone quoting a dark matter lifetime limit from the CMB, or working on non-standard dark sectors. It deserves a referee.","headline":"A careful, modest paper with a genuinely quotable channel-independent lifetime bound; the perturbation treatment holds up, and the flatness caveat is the one worth pressing.","tokens_in":9970,"tokens_out":2541,"duration_ms":45415},"pith_extraction":{"msc":[],"pacs":[],"model":"selfhosted/deepseek-v4.1-flash","headline":"By re-deriving how decaying cold dark matter imprints on the microwave background, the authors turn WMAP's temperature map into a lower limit on how long dark matter particles can survive: about 123 billion years at 68% confidence, and no证据","keywords":[],"falsifier":"placeholder","tokens_in":10188,"feed_emoji":"🌌","tokens_out":1055,"duration_ms":19777,"temperature":0.7,"pith_summary":"The paper asks a simple question with a hard answer: if the cold dark matter that fills the universe is slowly decaying into relativistic particles, how long must it live before the cosmic microwave background would look wrong? To answer it the authors write down, from scratch, the perturbation equations for a decaying dark matter fluid plus the radiation it produces, patch them into a standard Boltzmann code, and run a Markov-chain fit against the first-year WMAP temperature data. The data show no preference for decay, so the result is a one-sided bound rather than a detection: $","feed_headline":"Dark matter must outlive 123 billion years","feed_subtitle":"WMAP's microwave map sets a gravity-only lower limit on how long cold dark matter can decay, independent of what it decays into.","key_machinery":"placeholder","core_discovery":"placeholder","pith_inferences":[],"forward_implications":[],"supporting_citations":[],"fun_headline_variants":["WMAP sets a 123-billion-year lower bound on decaying cold dark matter lifetime","WMAP alone does not favor decaying cold dark matter over stable dark matter","Gravity-only WMAP bound puts decaying dark matter lifetime above 123 Gyr","Decaying dark matter lifetime exceeds 123 Gyr under WMAP, decay products unspecified"],"cache_read_input_tokens":0,"weakest_assumption_plain":"placeholder","fun_headline_variants_meta":{"raw":{"variants":["WMAP sets a 123-billion-year lower bound on decaying cold dark matter lifetime","WMAP alone does not favor decaying cold dark matter over stable dark matter","Gravity-only WMAP bound puts decaying dark matter lifetime above 123 Gyr","Decaying dark matter lifetime exceeds 123 Gyr under WMAP, decay products unspecified"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1694,"prompt_tokens":738,"completion_tokens":956,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":880}},"tokens_in":738,"tokens_out":956,"duration_ms":18099,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":880,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-13T08:51:21.277381+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"pipeline_version":"pith-pipeline@v1.5.0-archive-selfhost","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"placeholder","supporting_citations":[],"lane_accuracy":{"status":"not yet measured","lane":{"model":"selfhosted/deepseek-v4.1-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}