{"id":"8023ee43-3cef-415f-94c2-6c7aea8c2911","arxiv_id":"astro-ph/9506072","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"low","formal_verification":"none","parameter_count":4,"one_line_summary":"The paper gives the full linear perturbation equations for all cosmic components in two gauges and produces the first accurate microwave background power spectra for mixed cold-plus-hot dark matter models.","lead":"Ma and Bertschinger write down the complete linear equations for how gravity, dark matter, ordinary matter, photons, and neutrinos evolve together in an expanding universe, in two coordinate systems, and solve them numerically. The paper's public code and its first accurate mixed dark matter maps of the microwave background made it a standard reference for testing cosmology models against sky surveys.","discovery_kind":"new_method","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The 10^-3 accuracy claim rests on convergence tests that are asserted but not tabulated, and the 0.1% two-gauge agreement — the paper's headline cross-check — is blind to the truncation and tight-coupling approximations that both codes share.","rationale":"The reader's weakest_assumption correctly targets the truncation closures and tight-coupling switches, and I agree that this is where the accuracy claim is least secure. Where I would refine the reader's framing is the weight given to the two-gauge comparison: the rationale lists 'cross-checks the two gauges to 0.1 percent' as evidence of care, but that check is not independent with respect to the approximations under suspicion, because both codes share the same closures, q grid, switch temperatures and recombination history. The gauge comparison validates the transformation of eqs. (18) and (27) and the metric evolution of eqs. (21) and (23); it does not validate eqs. (51), (58), (65) or the switches in §5.7. I also note that 0.1% and 10^-3 are the same number, so the gauge discrepancy is not a small correction to the error budget but is of the same order as the claimed total accuracy. Beyond that, the paper is internally consistent where I checked it: the super-horizon initial conditions in eqs. (96) and (98) reproduce the constraint equations (23a) and (23d) when the neutrino shear of eq. (98) is inserted, including the sign and magnitude of phi - psi = (2/5) R_nu psi; the isentropic relations delta_c = delta_b = (3/4) delta_gamma follow from delta_rho/(rho+P) being species-independent; and the truncation forms in eqs. (51), (58) and (65) match the standard Bessel-recursion and streaming closures. The COBE normalization Q_rms-PS = 17.6 microkelvin is a cited published value (Bennett et al. 1994) used only for display in Fig. 4, so it is not a soft point. The priority claim ('first accurate') is not testable in the way a numerical result is, but the paper's substantive content does not depend on it. Overall this is a careful methods paper with released code and a genuine internal cross-check; the residual concern is about the completeness of the error-budget reporting, not about correctness, and it does not move the verdict. I would keep ACCEPT with high confidence and, if anything, ask for one table of convergence numbers rather than a change in assessment. No circularity is present: the initial conditions are derived from the same linear equations that are then integrated, and the only data-facing input is a display normalization. The concern I raise is a reporting gap that a single table would close, which is why the verdict stays UNCHANGED rather than moving to CONDITIONAL.","tokens_in":32793,"tokens_out":9149,"duration_ms":169332,"concrete_test":"Re-run the Omega_nu = 0.2 model and tabulate the fractional change in C_l at l = 200, 500 and 800, and in the z = 0 matter transfer function at k = 0.1, 1 and 10 Mpc^-1, under four perturbations applied one at a time: (i) lmax for massive neutrinos raised from 50 to 150; (ii) lmax for massless neutrinos and photons raised from 2000 to 4000; (iii) the theta_b tight-coupling switch moved from Tb = 2x10^4 K to 5x10^4 K; (iv) the F_gamma_l switch moved from 2x10^5 K to 5x10^5 K. If any single change shifts C_l or the transfer function by more than 10^-3, the quoted accuracy is not supported and the error estimate in §8 needs revision. If all shifts stay below 10^-3, the concern is settled in the paper's favour and the accuracy claim stands as stated.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing claim is accuracy: 'the first accurate calculations of the angular power spectra' (abstract:s6), with 'relative accuracy better than 10^-3' (§8). Three pieces of evidence are offered: (a) the two gauges agree to better than 0.1%; (b) grids, q-points, l-values and timesteps were doubled; (c) the code is released. Concern: (a) is not independent with respect to the dominant approximations. Both codes use the same Boltzmann closures, eqs. (51), (58) and (65); the same 128-point q grid for massive neutrinos; the same tight-coupling switch temperatures Tb = 2x10^4 K and 2x10^5 K (§5.7); and the same pre-computed recombination and baryon-temperature splines (§5.8). The 0.1% agreement therefore bounds gauge-transformation and metric-evolution errors, not truncation or switch errors. Moreover 0.1% equals 10^-3, the same order as the quoted accuracy, so the gauge difference alone consumes the entire stated error budget rather than sitting comfortably below it. (b) is stated but never quantified: no table gives C_l versus lmax, q-grid size, or switch temperature, so a reader cannot see how much margin exists. A specific gap in the stated reasoning: lmax = 50 for massive neutrinos is justified by 'higher multipole moments decay rapidly once the neutrinos become nonrelativistic,' but the high multipoles are populated while the neutrinos are still relativistic (a < a_nr ~ 10^-4, with tau_nr of order twice tau_eq), where k tau reaches ~10^3 for k = 10 Mpc^-1. The stated justification covers only the late phase; the early phase is supported solely by the unquantified empirical check. This is a reporting and justification gap, not a demonstrated error, and the released code plus the two-gauge machinery are genuine strengths.","agreement_with_reader":"partial"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript gives a systematic, side-by-side derivation of the linear theory of scalar cosmological perturbations in the synchronous and conformal Newtonian gauges for flat models containing CDM, baryons, photons, massless neutrinos, and massive neutrinos, with a cosmological constant allowed through the background. It derives the gauge transformation between the two coordinate systems (§3, Eqs. 11–18), writes the perturbed Einstein equations in both gauges (§4, Eqs. 21 and 23), derives the Boltzmann hierarchies for massless and massive neutrinos and for photons including polarization (§5.3–5.5, Eqs. 49–50, 56–57, 63–64), gives the fluid equations for CDM and baryons with Thomson coupling and a tight-coupling expansion (§5.2, 5.6, 5.7), describes recombination with helium (§5.8), derives isentropic super-horizon initial conditions analytically in both gauges (§7, Eqs. 96–98), and integrates the coupled system for a pure CDM model and two CDM+HDM models with Omega_nu = 0.2 and 0.3, presenting metric and density evolution and angular power spectra including polarization, higher neutrino moments, and helium recombination (§8, Figs. 1–4). The headline claim is that these are the first accurate C_l calculations for CDM+HDM models including all of that physics (abstract:s6).","tokens_in":33242,"tokens_out":5168,"duration_ms":94406,"significance":"If the results hold, this is a substantial and durable contribution: a complete, self-contained reference for the linear perturbation equations in two gauges, with every species relevant to CDM, HDM, and mixed models treated at the same level of detail. Three strengths deserve explicit credit. First, the parallel presentation in two gauges with an explicit transformation (Eqs. 18, 27) is not decoration; it supplies a genuine internal cross-check, and the reported agreement of better than 0.1 percent between the two independent integrations is a meaningful validation of the metric evolution and gauge transformation. Second, the super-horizon initial conditions in Eqs. (96)–(98) are derived analytically from the same Einstein and Boltzmann equations rather than fitted to output, and the growing-mode amplitude is fixed by a single constant C, so the initial-value problem is parameter-free once the mode is chosen. Third, the numerical programs for both gauges are released, which makes the calculation reproducible and is exactly the practice this field should reward. The massive-neutrino phase-space treatment (128-point q grid, per-q Legendre hierarchy, Eq. 58) is a real advance over cr","major_comments":[{"comment":"The stated accuracy, 'we estimate that our final results have a relative accuracy better than 10^-3' (§8), is load-bearing for the abstract's claim of 'the first accurate calculations' (abstract:s6), but the supporting evidence is not quantified anywhere in the paper. The convergence checks are described in one sentence ('increasing the grids of k, q, and l values as well as decreasing the integration timestep') with no numbers attached. More importantly, the 0.1 percent two-gauge agreement, which is the paper's most visible cross-check, is not independent with respect to the approximations most likely to control the error: both codes use the same truncation closures of Eqs. (51), (58), and (65), the same 128-point q grid, the same tight-coupling switch temperatures T_b = 2x10^4 K and 2x10^5 K (§5.7), and the same pre-computed recombination and baryon-temperature splines (§5.8). The gauge comparison therefore bounds gauge-transformation and metric-evolution errors, not truncation or switch errors. Note also that 0.1 percent equals 10^-3, the same order as the quoted accuracy, so the gauge difference alone consumes the entire stated error budget rather than sitting comfortably below it. I ask the authors to add a short table reporting C_l at a few multipoles (say l = 10, 100, 200, 500, 1000) for variations of l_max, q-grid size, and the two switch temperatures, so that the reader can see the margin rather than take the estimate on trust.","section":"§8, accuracy claim","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The justification given for truncating the massive-neutrino hierarchy at l_max = 50 is that 'the higher multipole moments decay rapidly once the neutrinos become nonrelativistic.' That argument applies at a > a_nr, but the high multipoles are populated while the neutrinos are still relativistic, and for the largest wavenumber in the integration (k = 10 Mpc^-1) the product k*tau reaches of order 10^3 by tau_nr. The truncation scheme of Eq. (58) is a Bessel-like extrapolation and may well handle this, but the paper does not demonstrate it. Because the massive-neutrino moments feed the shear stress that enters the Einstein equations and hence the potentials shown in Fig. 2, this is a correctness risk for the central result rather than a cosmetic one. A concrete test is available and cheap relative to the runs already performed: repeat the k = 1 and k = 10 Mpc^-1 integrations with l_max = 100 or 200 for the massive neutrinos and report the change in delta_h, sigma_h, and C_l. If the change is below 10^-3, the choice is vindicated and the sentence in §5.4 can be replaced by a quantitative statement.","section":"§5.4, Eq. (58) and l_max = 50 for massive neutrinos","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The paper's headline product is the angular power spectrum, yet no numerical values of C_l, and no peak positions or peak heights, are tabulated anywhere; the results exist only as figures. This makes the accuracy claim difficult to check independently and makes the paper less useful as the reference it is clearly intended to be. I recommend a small table of C_l (or of the acoustic peak positions and amplitudes) for the three models, in the same normalization as Fig. 1, so that later calculations can be compared against this work without digitizing a plot.","section":"§8, Figs. 1 and 4","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The polarization-dependent Thomson collision terms are attributed jointly to Bond & Efstathiou (1984, 1987) and to Kosowsky (1995), the latter listed in the references as a preprint (astro-ph/9501045). Since the polarization term is central to the paper's novelty claim, either cite a published derivation for the specific form of Eq. (60) or sketch the two-line derivation from the differential cross section d(sigma)/d(Omega) = 3 sigma_T (1 + cos^2 theta)/(16 pi) that is already quoted in the text.","section":"§5.5, Eqs. (59)–(60)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Two different truncation criteria are used without explanation of how they relate: l_max = 1.5 k tau_max + 10 for the temperature-fluctuation runs, and l_max = 2000 for the massless species in the potential and density runs. A sentence stating why the potential/density calculation needs the larger value (presumably because it is run to k = 10 Mpc^-1 rather than 0.5 Mpc^-1) would help the reader judge the computational cost and the adequacy of each choice.","section":"§8, truncation criteria","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The statement that the switches at T_b = 2x10^4 K and 2x10^5 K 'occur early enough to preserve good accuracy in the resulting photon phase space distribution' is asserted rather than shown. This is folded into major comment 1, but even a single sentence giving the size of the change in C_l when the switch temperature is halved would settle it.","section":"§5.7, switch temperatures","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The baryon sound speed neglects the slow time variation of the mean molecular weight and of d ln T_b / d ln a. The text says the approximation is adequate because the baryons contribute little to the pressure of the photon-baryon fluid; a rough numerical bound on the resulting error in c_s^2 during recombination would make that statement checkable.","section":"§5.6, Eq. (68)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The initial conditions retain only the fastest-growing physical mode and drop the D-mode, and the two residual gauge modes A and B are removed by the choice of coordinates. It would be worth stating explicitly, in one sentence, that the numerical integration starts at tau_0 = 3x10^-4 Mpc and that the neglected decaying mode contributes below the stated accuracy at that time, since the choice of starting time is what makes the truncation of the mode expansion safe.","section":"§7, Eqs. (96)–(98)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The symbol glossary omits several quantities that appear repeatedly in the text and figures: k_fs and a_nr (§8, Eq. 102), F_gamma_l and G_gamma_l (§5.5), Psi_l (§5.4), and Delta_l (§6). Adding them would make the table serve its stated purpose as a guide to the reader.","section":"Table 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"Fig. 2 shows 31 curves distinguished only by the caption's 'from left to right'; labelling three or four representative wavenumbers directly on the plot would help. In Fig. 3, marking a_nr, a_eq, and a_rec on the horizontal axis of each panel would make the discussion in items (1)–(3) of §8 much easier to follow.","section":"Figs. 2 and 3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The large-angle result C_l ≈ (8 pi^2/9) A / [l(l+1)] for n = 1 should state explicitly that A is defined by P_psi(k) = A k^{-3} in the same convention as Eq. (90), since the numerical coefficient depends on that convention and readers will otherwise compare it to the standard Sachs-Wolfe normalization and find a discrepancy.","section":"§6, Eq. (91)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The version supplied to me shows broken words and stray spacing in the title and abstract ('Co nformal', 'onu' for Omega_nu, split superscripts in the equations). This may be an artifact of the file I received, but please check the typeset version, since the abstract is the part most readers will see.","section":"Title and abstract, typesetting","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision","confidential_remarks":"This is a careful, well-organized, and likely to be heavily cited paper, and the release of both gauge codes is a genuine service to the community. My recommendation of minor revision should not be read as doubt about the physics: the derivations I checked (the gauge transformation of Eqs. 11–18, the transformation of the energy-momentum tensor in Eq. 27, and the super-horizon initial conditions of Eqs. 96–98) are correct and consistent with the standard literature. The revision I am asking for is narrow: quantify the convergence tests that currently back the 10^-3 accuracy claim, and add a table of C_l values so the headline result is checkable without digitizing a figure. The paper is long and equation-dense, but the symbol glossary and the parallel gauge presentation make it usable, and I would not ask for cuts."},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"This is the paper people cite when they need the linear equations written down correctly in both gauges, and it earns that. The real content is the parallel derivation: Einstein equations, Boltzmann hierarchies for photons and both neutrino species, fluid equations for CDM and baryons, all laid out side by side in synchronous and conformal Newtonian form, plus the explicit transformation between them (eq. 18) and the super-horizon isentropic initial conditions (eqs. 96–98). The massive neutrino treatment is the part that was genuinely missing before — full momentum dependence on a 128-point q-grid instead of the l=2 truncation everyone else used. The released code for both gauges is not a gesture; it is why the paper became infrastructure. The two-gauge agreement to 0.1% is a real cross-check on the transformation and the metric evolution, and the initial conditions are derived from the same equations rather than fitted, so there is no circularity to worry about. The COBE quadrupole appears only as a normalization for Fig. 4 and does no argumentative work. On the soft side, the stress-test note is right about the accuracy claim, though it is a reporting gap rather than an error. The 0.1% gauge agreement is not independent with respect to the things most likely to be wrong: both codes share the same Boltzmann closures (eqs. 51, 58, 65), the same q-grid, the same tight-coupling switch temperatures, and the same pre-computed recombination splines. So it bounds gauge error, not truncation error — and 0.1% is the same order as the quoted 10^-3, which leaves no visible margin. The convergence checks are described but never tabulated; a reader cannot see how Cl moves with lmax or with the switch temperature. The lmax=50 justification for massive neutrinos is stated for the non-relativistic phase, but the high multipoles are populated while the neutrinos are still relativistic, so that early phase rests entirely on the unquantified empirical check. None of this undermines the physics; it means the word 'accurate' in the abstract is doing slightly more work than the evidence shown. Who should read it: anyone building a Boltzmann code, anyone who needs the gauge transformation, and anyone working on mixed dark matter. It is a methods paper, not a controversy, so it is a reference rather than a reading-group session. Send it out; the referees will ask for the convergence table and the paper will be better for it.","headline":"The standard reference for linear perturbation theory in two gauges, and it holds up — the only thing worth pressing is that the 10^-3 accuracy is an estimate, not a bound.","tokens_in":33933,"tokens_out":951,"would_cite":true,"duration_ms":14644,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"pith_extraction":{"msc":[],"pacs":["98.80.-k","98.70.Vc","95.35.+d"],"model":"selfhosted/deepseek-v4.1-flash","headline":"The paper derives the linear theory of cosmic structure formation twice over — once in the synchronous gauge and once in the conformal Newtonian gauge — for every particle species that matters, and uses the agreement between the two to make","keywords":["cosmological perturbation theory","cosmic microwave background anisotropy","synchronous gauge","conformal Newtonian gauge","gauge transformation","massive neutrinos","Boltzmann hierarchy","cold plus hot dark matter"],"falsifier":"A converged solution of the same equations that does not share the paper's truncations — for instance direct integration of the photon and neutrino phase space without a Legendre hierarchy, or the same code with $ell_{max}$, the $q$-grid, and the timestep refined far past the paper's values and the tight-coupling switches removed — would settle it: a shift in any $C_ell$ larger than the quoted $10^{-3}$ relative accuracy would break the claim, while agreement to that level would confirm it.","tokens_in":34081,"feed_emoji":"🌌","tokens_out":4015,"duration_ms":84498,"temperature":0.7,"pith_summary":"The paper asks a bookkeeping-shaped question with real physical stakes: if you write the linear theory of gravitational instability in two different coordinate choices — the synchronous gauge used since Lifshitz, and the conformal Newtonian gauge — do you get the same universe? Ma and Bertschinger answer by deriving side by side the complete set of Einstein, Boltzmann, and fluid equations for every species that matters in a flat cold dark matter, hot dark matter, or mixed model: CDM, baryons, photons, massless neutrinos, and massive neutrinos, with a cosmological constant allowed. The payoff is a numerical pipeline that can be run twice, once per gauge, and checked against itself; the two independent integrations agree to better than 0.1 percent. On that foundation they produce the first angular power spectra for mixed dark matter models that retain photon polarization, the full neutrino multipole hierarchy, and helium recombination. For a reader who wants to know why cosmic microwave background predictions can be trusted at the percent level, the paper's real contribution is the demonstration that a gauge choice is a computational convenience rather than a physical assumption.","feed_headline":"Two gauges, one answer for how cosmic structure grew","feed_subtitle":"Independent synchronous and Newtonian integrations agree to 0.1 percent, giving the first full neutrino CMB spectra.","key_machinery":"Three pieces carry the argument. The first is the gauge transformation of equation (18), which converts between $(h, eta)$ and $(psi, varphi)$ and is what makes a two-code cross-check possible at all. The second is the parallel set of Boltzmann hierarchies: the photon and massless-neutrino distribution functions are expanded in Legendre polynomials and reduced to coupled ordinary differential equations in equations (49), (50), (63), and (64), while the massive neutrinos keep an explicit momentum dependence and are tracked on a grid of 128 $q$-points per wavenumber with their own hierarchy in equations (56) and (57). The third is the set of truncation rules — equations (51), (58), and (65) — which close the infinite hierarchies by extrapolating the spherical-Bessel-like behavior of the high multipoles rather than simply setting them to zero, plus the tight-coupling approximation of §5.7 that replaces the stiff Thomson-scattering terms with a sound-speed equation for the combined photon-baryon fluid. Together these turn an intractable integro-differential problem into a finite system that a Runge-Kutta integrator can carry from redshift $z sim 10^9$ to today.","core_discovery":"The central claim is that a single, complete linear perturbation theory can be written down for all the species relevant to flat CDM, HDM, and CDM+HDM cosmologies, and that solving it in two different gauges gives the same answer. The bridge between the gauges is an explicit coordinate transformation, equation (18), which expresses the conformal Newtonian potentials $psi$ and $varphi$ in terms of the synchronous fields $h$ and $eta$ and their time derivatives. With that bridge in hand, the authors derive super-horizon isentropic initial conditions in both gauges — equations (96) and (98) — showing that the same physical mode appears as a growing density perturbation in the synchronous gauge and as a constant one in the conformal Newtonian gauge, with the difference traceable entirely to the coordinate choice. They then integrate the coupled system numerically for a standard CDM model and two mixed models with neutrino mass densities $Omega_nu = 0.2$ and $0.3$ (neutrino masses 4.7 eV and 7 eV), assuming a scale-invariant adiabatic spectrum. The two gauge integrations agree to better than 0.1 percent, and the resulting spectra are, on the authors' account, the first for mixed dark matter models that include photon polarization, higher neutrino multipole moments, and helium recombination.","pith_inferences":["If two independently written Boltzmann codes ever disagree by more than about 0.1 percent on a mixed dark matter spectrum, that disagreement is more likely a truncation artifact or a coding error than new physics — the paper effectively sets a benchmark for what 'agreeing codes' should mean.","The weakest link in the numerical chain is the hierarchy closure, which rests on an asymptotic analogy to spherical Bessel functions; a genuinely independent check would come from a line-of-sight integration that never truncates the multipole expansion at all.","The sensitivity of the acoustic peak heights to $Omega_ nu$ runs in both directions: the same machinery, inverted, turns a measured peak ratio into a constraint on the sum of neutrino masses.","The gauge transformation itself is a reusable tool beyond this paper — any calculation performed in the synchronous gauge can be mapped into the conformal Newtonian gauge where the potentials have a Newtonian interpretation, which is exactly what is needed when feeding linear results into Newtonian N-body simulations."],"forward_implications":["Adding a massive neutrino at $Omega_nu = 0.2$ or $0.3$ raises the anisotropy at high multipoles by roughly 10 percent at the second and third acoustic peaks, so a percent-level measurement of peak heights separates mixed dark matter from pure CDM.","Free-streaming of the massive neutrino suppresses its own perturbation growth and slows the growth of CDM in turn, with the suppression scale set by $k_{fs}(a) propto a^{1/2} m_ u$, so the shape of the matter power spectrum becomes a handle on neutrino mass.","Because the two gauges agree to 0.1 percent, the gauge-invariant content of the calculation is confirmed, and a user may choose whichever gauge simplifies an extension — for example, integrating neutrino geodesics to build N-body initial conditions.","The released code makes the entire pipeline reproducible and extendable to a cosmological constant and to additional species without rederiving the perturbation equations."],"supporting_citations":[{"why":"Supplies the gauge-invariant variables $Phi_A$ and $Phi_H$ to which the conformal Newtonian potentials $psi$ and $varphi$ are shown to reduce, anchoring the claim that this gauge is free of gauge modes.","marker":"Bardeen 1980"},{"why":"The source of the conformal Newtonian (longitudinal) gauge adopted here, and the review against which the paper positions its treatment of scalar perturbations.","marker":"Mukhanov et al. 1992"},{"why":"Origin of the synchronous gauge and of the linear perturbation formalism that the paper rederives and extends to a fuller species list.","marker":"Lifshitz 1946"},{"why":"The prior neutrino treatment the paper improves on: it kept only the first two angular moments and approximated the neutrino as a perfect fluid, which is the gap the full hierarchy is meant to close.","marker":"Bond & Szalay 1983"},{"why":"Source of the tight-coupling method used in §5.7 to replace the stiff Thomson-scattering terms with a combined photon-baryon sound-speed equation.","marker":"Peebles & Yu 1970"},{"why":"Supplies the recombination calculation, including the reduction factor $C_r$ for net decay to the ground state, on which the free-electron density and hence the Thomson opacity depend.","marker":"Peebles 1968"},{"why":"Provide the linearized Thomson collision operators for the photon intensity and polarization that appear in equations (59) and (60).","marker":"Bond & Efstathiou 1984, 1987"},{"why":"The comparison case establishing the stakes of the truncation question: dropping massless-neutrino moments beyond $ell ge 2$ costs about 10 percent in the anisotropy power spectrum, motivating the full hierarchy here.","marker":"Hu et al 1995"},{"why":"Classifies the four super-horizon eigenmodes in the synchronous gauge into two gauge modes and two physical modes, which justifies discarding the $A$ and $B$ solutions in the initial conditions.","marker":"Press & Vishniac 1980"},{"why":"The companion paper whose N-body initial conditions required the metric perturbations computed here, and the stated motivation for developing the conformal Newtonian code.","marker":"Ma & Bertschinger 1994a"}],"fun_headline_variants":["Same cosmic structure emerges from two different gauges","Two gauges agree to better than 0.1 percent on cosmic growth","First full neutrino CMB spectra for mixed dark matter models","Massive neutrinos tracked in full phase space, not approximated","One linear theory covers CDM, HDM and the mixed case alike"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The quoted accuracy of the spectra rests on an estimate — grids, multipole ceilings, and timesteps doubled until the answer stopped moving — rather than on a proven error bound, so everything hangs on the assumption that the closure rules of equations (51), (58), and (65) and the tight-coupling switch temperatures of §5.7 converge for every wavenumber and multipole the paper uses.","fun_headline_variants_meta":{"raw":{"variants":["Same cosmic structure emerges from two different gauges","Two gauges agree to better than 0.1 percent on cosmic growth","First full neutrino CMB spectra for mixed dark matter models","Massive neutrinos tracked in full phase space, not approximated","One linear theory covers CDM, HDM and the mixed case alike"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":1461,"prompt_tokens":1111,"completion_tokens":350,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":276}},"tokens_in":1111,"tokens_out":350,"duration_ms":5493,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":276,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-10-29T19:29:41.249389+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"falsifier":"A converged solution of the same equations that does not share the paper's truncations — for instance direct integration of the photon and neutrino phase space without a Legendre hierarchy, or the same code with $ell_{max}$, the $q$-grid, and the timestep refined far past the paper's values and the tight-coupling switches removed — would settle it: a shift in any $C_ell$ larger than the quoted $10^{-3}$ relative accuracy would break the claim, while agreement to that level would confirm it.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the gauge-invariant variables $Phi_A$ and $Phi_H$ to which the conformal Newtonian potentials $psi$ and $varphi$ are shown to reduce, anchoring the claim that this gauge is free of gauge modes."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The source of the conformal Newtonian (longitudinal) gauge adopted here, and the review against which the paper positions its treatment of scalar perturbations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Origin of the synchronous gauge and of the linear perturbation formalism that the paper rederives and extends to a fuller species list."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"The prior neutrino treatment the paper improves on: it kept only the first two angular moments and approximated the neutrino as a perfect fluid, which is the gap the full hierarchy is meant to close."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Source of the tight-coupling method used in §5.7 to replace the stiff Thomson-scattering terms with a combined photon-baryon sound-speed equation."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the recombination calculation, including the reduction factor $C_r$ for net decay to the ground state, on which the free-electron density and hence the Thomson opacity depend."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provide the linearized Thomson collision operators for the photon intensity and polarization that appear in equations (59) and (60)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Classifies the four super-horizon eigenmodes in the synchronous gauge into two gauge modes and two physical modes, which justifies discarding the $A$ and $B$ solutions in the initial conditions."}],"review_version":1}