{"id":"3ff8b58e-82ee-4f70-9661-1296a4710d55","arxiv_id":"1908.08876","paper_version":4,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"Adding a 1/frequency turbulent diffusion to the Kompaneets equation yields a stationary spectrum that fits the observed low-frequency cosmic background excess.","lead":"This paper proposes that the unexplained extra radio brightness in the cosmic background, the space roar, is a relic of nonequilibrium processes in the early universe. The authors add a turbulent frequency-diffusion term to the standard Kompaneets equation and argue the resulting spectrum can explain the low-frequency excess.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The predicted 1/ν diffusivity rests on an assumed decorrelation rate τ^{-1}∝ν_i; if the rate is set by plasma physics instead, the low-frequency enhancement and the fit to the radio excess do not follow.","rationale":"The reader identified the same load-bearing assumption as the weakest point: the decorrelation-rate scaling τ^{-1}∝ν_i in Section IV, Eqs. (6)-(8), which is the sole basis for B(ν)∝1/ν and hence for the entire low-frequency enhancement. I agree that this scaling is not derived from a plasma model and that the paper's own concessions ('details on the precise mechanism cannot be provided') mark it as an assumption. The alternative concern about the approximate stationary solution (16) is real but secondary: (16) is crafted to match the exact φ(ν) in both low- and high-frequency limits, so fits with (16) are representative of the model's predictions in the observed band; the divergence at ν→0 in Appendix B affects energy/number integrals rather than the 20 MHz–90 GHz comparison. The reduced χ^2≈2 and the fitting of ν0 and α to the same data that motivated the model further weaken the empirical claim but do not change the verdict: the theory is testable and honestly presented, and its central weakness is the unjustified τ scaling. Since my analysis does not alter the reader's CONDITIONAL verdict, no change is needed.","tokens_in":1061,"tokens_out":933,"duration_ms":79487,"concrete_test":"Derive the frequency-diffusion coefficient for a concrete microscopic model: an electron moving with velocity v through a prescribed random electromagnetic field with correlation length ℓ and correlation time τ_c, with wave frequencies as seen in the electron rest frame. Compute the mean squared photon-frequency change ⟨(Δν)^2⟩ per scattering as a function of the incoming photon frequency ν_i (e.g., from the Doppler-shifted resonance ν_i(1 - v·n̂/c) and the field spectrum). If the resulting B(ν) in the Fokker-Planck equation is not ∝1/ν over the MHz–GHz band, the proposed mechanism fails. Equivalent check: numerically extract B(ν) from the Boltzmann collision term with a stochastic force and verify the ν-scaling.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central prediction T(ν)=T_*[1+(ν/ν0)^{-α}] with α≈3 follows from the additional diffusivity B(ν)∝1/ν in Eqs. (10)-(11). This form is derived in Section IV by the central-limit argument of Eqs. (6)-(8): the variance of the photon frequency kick is proportional to the persistence time τ, so the claimed 1/ν scaling is entirely inherited from the assumed decorrelation rate τ^{-1}∝ν_i (plus v/ℓ). That assumption is not derived from any plasma model; the paper itself states that 'details on the precise mechanism cannot be provided at this point' and that the origin 'can only be thought to reside ultimately with gravitational degrees of freedom.' The physical basis is also doubtful: the force felt by an electron in a random electromagnetic field oscillates at the Doppler-shifted wave frequency, not necessarily at the CMB photon frequency ν_i, and the advective term v/ℓ can dominate unless ℓ is specifically of order c/ν_i. If τ were instead set by, e.g., the plasma frequency or the electron collision time, then ⟨(Δν)^2⟩ would be independent of ν_i, giving B(ν)=const and a diffusivity D(ν)∝ν^2 at low frequency (via the ν^2 prefactor in (10)), which eliminates the low-frequency photon enhancement. Thus the agreement with the ARCADE 2 and low-frequency data is not a consequence of the stated statistical mechanism unless the τ scaling is independently justified. The approximation (16) is not the primary issue, since it reproduces the correct asymptotic scalings of the exact stationary solution (15), but the central scaling assumption itself remains unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper addresses the low-frequency excess in the cosmic background radiation (the 'space roar') by proposing a nonequilibrium modification of the Kompaneets equation. The authors add a purely diffusive term in photon-frequency space, B(ν)∝1/ν, argued from a central-limit scaling argument for stochastic acceleration in a random force field. The resulting stationary occupation number yields an effective brightness temperature T(ν)=T_*[1+(ν/ν0)^(-α)] with α≈3, which is fitted to an updated compilation of absolute temperature measurements from 22 MHz to about 600 GHz. The authors report good qualitative agreement with α fixed to 3 (reduced χ²≈2.1) and slightly better with α free (α≈3.3, χ²≈1.9), and interpret the excess as a nonequilibrium echo of the primordial plasma. The paper provides a useful data compilation and a transparent, testable functional form, but the physical derivation of the 1/ν diffusivity is the key unsupported step.","tokens_in":23595,"tokens_out":14565,"duration_ms":141722,"significance":"If the proposed mechanism were correct, the paper would offer a novel explanation for a long-standing observational anomaly, with a falsifiable prediction (a power-law temperature excess T∝ν^(-α), α≈3) that can be tested by future low-frequency measurements. The updated data compilation is a useful community resource, and the connection to nonequilibrium statistical mechanics (Einstein-relation violation, blowtorch theorem) is conceptually interesting. However, the central claim is conditional: the predicted 1/ν scaling is inherited from an assumed decorrelation rate τ^{-1}∝ν_i that is not derived from a plasma model, and the fits use an approximation to the model's stationary solution. The model's contribution is therefore best regarded as a proof-of-concept that a broad class of non-equilibrium frequency-diffusion processes can mimic the radio excess, rather than as an established explanation.","major_comments":[{"comment":"The central result B(ν)∝1/ν is obtained by assuming the decorrelation rate τ^{-1}∝ν_i+v/ℓ and then retaining only the ν_i term. This is the load-bearing step: without τ∝1/ν_i, Eq. (8) gives a variance independent of ν_i and the low-frequency enhancement disappears. The manuscript does not derive this rate from any concrete plasma or field model; the Introduction states that 'details on the precise mechanism cannot be provided at this point' and that the origin 'can only be thought to reside ultimately with gravitational degrees of freedom.' The spatial term v/ℓ is also not estimated: for the ν_i term to dominate, one needs ℓ ≫ v/ν_i, which is a nontrivial assumption about the force-field correlation length. The authors should either supply a physical model for τ or explicitly identify the 1/ν scaling as an assumption; in the latter case, the data fit is a test of the assumed form, not of the proposed statistical mechanism.","section":"Section IV, Eqs. (6)-(8)"},{"comment":"The fits do not use the model defined by Eqs. (10)-(14). The exact stationary solution of (10)-(14) has φ(ν)=∫dν' γ(ν')/D(ν'), given by a hypergeometric function, while Eq. (16) replaces it with φ(ν)=(hν/k_BT_e)(ν/ν0)^α/[1+(ν/ν0)^α], which has the same asymptotics but differs in the crossover region. Because the crossover region is precisely where the data (roughly 0.4-3 GHz) constrain the model, the best-fit values of ν0 and α and the reported reduced χ² may change when the exact model is used. The authors should fit the exact stationary solution, or quantify the error introduced by (16), before claiming quantitative agreement.","section":"Section V, Eq. (16) and Fig. 1"},{"comment":"The fixed-α=3 fit has reduced χ²≈2.1 and the two-parameter fit has reduced χ²≈1.9. For a model presented as predicting the observed spectrum, these values indicate a statistically poor fit: with the quoted 1σ errors, the probability of obtaining χ²_red≥2.1 is very small if the model is correct. The paper attributes the excess to systematic differences between data sets, but no quantitative treatment is provided. The authors should include a residual analysis, allow for a variance floor or nuisance parameters to absorb relative systematics, and state explicitly whether the agreement is qualitative or quantitative. As written, the claim that the model reproduces the data 'down to about 20 MHz' is not supported at the quoted error level.","section":"Section V, Fig. 1 and Appendix A"}],"minor_comments":[{"comment":"Some rows of Table I are difficult to parse because the columns run together (for example, the 0.022 GHz row); the table should be reformatted with clearly separated columns or presented in machine-readable form.","section":"Appendix A, Table I"},{"comment":"The parameter ν1 is introduced as the lower limit of validity of B(ν)∝1/ν, but the bound ν1<10^{-2} GHz is stated without derivation; a brief explanation of how this bound is obtained from the data or from the model would improve clarity.","section":"Section IV, after Eq. (11)"},{"comment":"The mapping of the early-universe stationary solution to the present epoch under cosmological expansion should be made explicit; in particular, the parameter ν0 in Eq. (17) should be identified as the present-day crossover frequency, and the redshift dependence of ν0 should be stated.","section":"Section V, after Eq. (17)"},{"comment":"The phrase 'best fit, ν^{-1.3}' in the Introduction anticipates the fit results before the model is derived; rephrasing this to avoid the appearance of post hoc tuning would strengthen the presentation.","section":"Introduction, p. 2"}],"recommendation":"major_revision","confidential_remarks":"The paper is conceptually interesting and the data compilation is valuable, but the central physical assumption (τ^{-1}∝ν_i) is unproven and the quantitative fit is weaker than the abstract suggests. If the authors can either derive the decorrelation rate from a concrete model or clearly reframe the work as a phenomenological study, and if they fit the actual stationary solution of their equation, the manuscript could be suitable for publication. In its present form, the central claim of a nonequilibrium mechanism rests on an assumption that is stated rather than justified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this paper has a genuinely new mechanism and a useful data compilation, but the load-bearing step — where the 1/ν diffusion law comes from — is an assumed decorrelation rate, not a derived plasma result. To its credit, the paper says so itself: the detailed mechanism \"cannot be provided at this point\" and its origin \"can only be thought to reside ultimately with gravitational degrees of freedom.\"\n\nThe genuinely new piece is the extra turbulent frequency-diffusion term B(ν) ∝ 1/ν in the Kompaneets equation, which breaks the Einstein relation and gives a stationary T(ν) = T_*[1+(ν/ν0)^{-α}] with α ≈ 3 that tracks the low-frequency cosmic background excess down to ~20 MHz. The blowtorch framing — low-frequency localization plus a broken Einstein relation pumps photons toward low ν — is a clean way to see why such a term would do this. The updated data compilation in Table I, with two extragalactic subtraction models for the low-frequency points, is the most reusable part. The fixed-α=3 fit is reported honestly (reduced χ² ≈ 2.1), and Appendix B gives a quantitative treatment of the low-frequency divergence and the cutoff it requires.\n\nThe soft spots, in proportion. The big one: the derivation of B(ν) ∝ 1/ν rests entirely on the assumed decorrelation rate τ^{-1} ∝ ν_i in Eqs. (6)-(8). If that rate is set by plasma frequency or collision time instead, you do not get the 1/ν law and the match to the data stops being a consequence of the mechanism. The stress-test note is right about this, though its \"eliminates the enhancement\" phrasing is a step too strong — a constant B would give a shallower ν^{-2}-type tail, so the fit would fail rather than the mechanism vanish. Second: the power-law family is data-informed (the introduction already cites a \"best fit ν^{-1.3}\"), so the agreement is not an independent prediction; the α=3 anchor is a real prediction, but a loose one. Third: the fit is adequate, not excellent, and uses the approximation (16) rather than the exact stationary solution, though the asymptotics match. Minor, as the reader says.\n\nThis is a paper for people working on the ARCADE 2/EDGES excess and CMB spectral distortions — the data table will get reused, and the mechanism deserves to be in the conversation of proposed explanations. It deserves a serious referee, not a desk reject: the referee's key question is whether the decorrelation scaling can be anchored in plasma physics, and if not, the paper should remain a clearly-labeled phenomenological proposal. I would send it out.","headline":"A genuinely new mechanism (1/ν turbulent diffusion in the Kompaneets equation) wrapped around a useful data compilation, but the central scaling rests on an assumed decorrelation rate that no plasma model grounds — worth refereeing, not desk-rejecting.","tokens_in":24157,"tokens_out":11148,"would_cite":true,"duration_ms":106878,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"The low-frequency cosmic background excess may be a nonequilibrium echo of the primordial plasma.","keywords":["cosmic microwave background","low-frequency excess","space roar","Kompaneets equation","stochastic acceleration","nonequilibrium statistical mechanics","Einstein relation violation","ARCADE 2"],"falsifier":"A clean measurement of the absolute sky brightness in the currently unobserved 0.1–0.4 GHz band, using the same extragalactic foreground subtraction as the paper, would settle it: if the background follows $T(\\nu)\\simeq T_*[1+(\\nu/\\nu_0)^{-\\alpha}]$ with $\\nu_0\\simeq 0.3$–$0.4$ GHz and $\\alpha\\simeq 3$, the temperature near 0.2 GHz should be tens of kelvin, while a synchrotron-limited or flatter spectrum would fall well below that. Likewise, finding no rise or a flattening below 20 MHz would contradict the predicted $1/\\nu$ diffusion tail.","tokens_in":22964,"feed_emoji":"🌌","tokens_out":6313,"duration_ms":59930,"temperature":0.7,"pith_summary":"The paper argues that the unexplained low-frequency rise in the cosmic background radiation, often called the 'space roar', is a real cosmological signal rather than a foreground artifact. Its cause, the authors propose, is a stochastic frequency-diffusion process in the turbulent primordial plasma: random force kicks make photons drift toward lower frequencies, an effect absent from the standard equilibrium treatment. Adding this effect to the Kompaneets equation breaks the Einstein relation between diffusion and friction and yields a stationary spectrum with temperature $T(\\nu)=T_*[1+(\\nu/\\nu_0)^{-\\alpha}]$, $\\alpha\\simeq 3$. With only $\\nu_0$ (and optionally $\\alpha$) as free parameters, this curve fits the absolute temperature measurements of the cosmic background from about 20 MHz upward. If correct, the low-frequency excess is a nonequilibrium imprint left by the early universe's plasma.","feed_headline":"Cosmic radio excess fits a nonequilibrium echo of the early universe","feed_subtitle":"A stochastic frequency-diffusion term reproduces the observed low-frequency radio excess down to 20 MHz.","key_machinery":"The load-bearing object is the modified Kompaneets equation, the standard kinetic equation that describes how photons thermalize via Compton scattering off hot electrons. The paper adds to it a frequency-space diffusion term of the form $\\partial_\\tau n \\supset \\frac{1}{\\nu^2}\\partial_\\nu\\{\\nu^2 \\frac{k_B T_*}{h}B(\\nu)\\partial_\\nu n\\}$ with $B(\\nu)\\propto \\nu^{-1}$ for $\\nu\\gg \\nu_1$. The power $-1$ comes from a central-limit argument: if the random force on an electron decorrelates at a rate $\\tau^{-1}\\propto \\nu_i + v/\\ell$, the variance of the photon frequency kick scales as $1/\\nu_i$, so low-frequency photons diffuse the most. Combined with the inherent $\\nu^4$ low-frequency localization of the standard Kompaneets equation, this drives photons into a non-Planckian stationary distribution whose low-frequency density follows a modified Rayleigh-Jeans law $\\rho_{mRJ}(\\nu)\\propto \\nu^{1-\\alpha}$. The same structure, diffusion without compensating friction, is the mechanism that populates the soft-photon tail.","core_discovery":"The central proposal is that the standard Kompaneets equation, which relaxes photons toward a Planck blackbody by Compton scattering, is incomplete at low frequencies. The authors add a purely diffusive term with diffusivity $B(\\nu)\\propto \\nu^{-1}$, the analogue in frequency space of stochastic acceleration of charged particles by turbulent fields. Because this extra diffusion does not carry a matching friction, the Einstein relation is violated, and the stationary occupation number becomes non-Planckian: $n_s(\\nu)=1/(e^{\\varphi(\\nu)}-1)$ with $\\varphi(\\nu)\\simeq \\frac{h\\nu}{k_B T_*}\\frac{(\\nu/\\nu_0)^\\alpha}{1+(\\nu/\\nu_0)^\\alpha}$, corresponding to an effective temperature $T(\\nu)=T_*[1+(\\nu/\\nu_0)^{-\\alpha}]$. The predicted exponent $\\alpha\\simeq 3$ reproduces the measured excess, with best-fit $\\nu_0\\simeq 0.35$–$0.42$ GHz and $\\alpha\\simeq 3.3$ when left free. The paper reads the observed 'space roar' as a nonequilibrium echo of the primordial plasma.","pith_inferences":["If the mechanism is generic stochastic acceleration, the same $1/\\nu$ frequency diffusion should appear in other strongly turbulent plasmas, so analogous soft-photon excesses might be sought in settings such as cluster radio halos or the solar corona.","The paper's own energy-integrals appendix implies that for $\\alpha\\ge 3$ the total photon energy diverges unless a low-frequency cutoff or flattening exists; measuring the spectrum well below 20 MHz would therefore probe that cutoff rather than merely confirm the tail.","Because the stationary solution depends only on frequency, the model predicts an isotropic excess; comparing sky maps at a single low frequency would separate this prediction from anisotropic astrophysical foregrounds."],"forward_implications":["The low-frequency radio excess observed by ARCADE 2 and implied by the EDGES 21-cm absorption would be a cosmological nonequilibrium signal, not an unresolved foreground.","The primordial plasma cannot be assumed to have been in global thermal equilibrium around one second after the Big Bang; the low-frequency photon modes carried a nonequilibrium, near-stationary occupation.","The background's effective temperature rises steeply below about 1 GHz, following $T(\\nu)\\approx T_*[1+(\\nu/\\nu_0)^{-\\alpha}]$, so the excess grows roughly as $\\nu^{-3}$ at the lowest frequencies.","The unobserved bands near 0.1–0.4 GHz and below 20 MHz are where the model makes its sharpest, most falsifiable predictions."],"supporting_citations":[{"why":"Supplies the standard Kompaneets equation that the paper modifies.","marker":"[12]"},{"why":"Provide the stochastic acceleration literature that motivates the added frequency diffusion.","marker":"[13–15]"},{"why":"ARCADE 2 absolute sky brightness measurement showing the 3–90 GHz excess.","marker":"[8]"},{"why":"Interpretation paper whose Table 1 data and extragalactic subtraction are used in the fits.","marker":"[10]"},{"why":"EDGES 21-cm absorption profile cited as independent evidence for an excess low-frequency background.","marker":"[9]"},{"why":"Recent 0.04–0.08 GHz measurements by Dowell and Taylor, the lowest-frequency data compared.","marker":"[34]"},{"why":"Model of unresolved extragalactic radio source counts used to subtract foregrounds from the data.","marker":"[35]"},{"why":"FIRAS high-frequency measurements that anchor the 2.7 K Planckian part of the spectrum.","marker":"[7]"}],"fun_headline_variants":["Cosmic radio excess fits stochastic photon cooling","Photon frequency diffusion cools cosmic background to fit data","Space roar is nonequilibrium echo from primordial plasma","Stochastic diffusion rewrites Kompaneets to cool photons","Cosmic background low-frequency excess is nonequilibrium echo"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole low-frequency boost rests on the assumption that the random force on an electron decorrelates at a rate that grows with the photon frequency, making the variance of each frequency kick scale as $1/\\nu$; the paper presents this as a statistical heuristic and concedes that the detailed plasma mechanism is unknown.","fun_headline_variants_meta":{"raw":{"variants":["Cosmic radio excess fits stochastic photon cooling","Photon frequency diffusion cools cosmic background to fit data","Space roar is nonequilibrium echo from primordial plasma","Stochastic diffusion rewrites Kompaneets to cool photons","Cosmic background low-frequency excess is nonequilibrium echo"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001027,"raw_usage":{"total_tokens":4334,"prompt_tokens":956,"completion_tokens":3378,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":572,"completion_tokens_details":{"reasoning_tokens":3301}},"tokens_in":572,"tokens_out":3378,"duration_ms":25151,"temperature":1.0,"reasoning_tokens":3301,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:27:36.474134+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A clean measurement of the absolute sky brightness in the currently unobserved 0.1–0.4 GHz band, using the same extragalactic foreground subtraction as the paper, would settle it: if the background follows $T(\\nu)\\simeq T_*[1+(\\nu/\\nu_0)^{-\\alpha}]$ with $\\nu_0\\simeq 0.3$–$0.4$ GHz and $\\alpha\\simeq 3$, the temperature near 0.2 GHz should be tens of kelvin, while a synchrotron-limited or flatter spectrum would fall well below that. Likewise, finding no rise or a flattening below 20 MHz would contradict the predicted $1/\\nu$ diffusion tail.","supporting_citations":[],"review_version":1}