Pith. sign in

REVIEW 3 major objections 6 minor 57 references

The late-time heating Green's function and improvements to distortion frequency hierarchy treatment

T0 review · 3 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A new spectral template and effective critical frequency make the fast frequency-hierarchy method match full CMB distortion calculations to within one percent in the mu era.

desk verdict Useful late-time Green's function extension; the FH improvement is a transparent calibration to CosmoTherm and its universality is untested. read the letter →

arxiv 2501.12822 v1 pith:HULKVFFE submitted 2025-01-22 astro-ph.CO gr-qchep-th

Sara Evangelista , Jens Chluba , Francesco Pace This is my paper · ORCID
classification astro-ph.COgr-qchep-th MSC 83F0585A40 PACS 98.70.Vc98.80.-k
keywords CMBspectraldistortionsmudistortionfrequencyhierarchyGreen'sfunctionvisibilitydoubleComptonBremsstrahlunganisotropies
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper aims to make the fast frequency-hierarchy (FH) method as accurate as full numerical CMB spectral distortion calculations during the mu-era, the epoch of injections at $z_h \gtrsim 3\times 10^5$ when distortions are pure chemical-potential signals. It identifies two causes of the FH mismatch: the approximate critical frequency $x_c$ that sets the rate at which a mu distortion converts into a temperature shift, and the analytic mu spectral shape $M(x)$, which misses the low-frequency photon-emission physics. The authors introduce an effective critical frequency $x_c^{\mathrm{eff}} = x_c^{\mathrm{DC},0} \hat{I}_\mu$ with $\hat{I}_\mu \approx 1.0555$, and a numerical mu template $M_*(x)$ computed from the full solution at $z_h = 5\times 10^5$. With both corrections, the FH treatment reproduces the full distortion for $z_h \gtrsim 3\times 10^5$ with high-frequency mismatch below about one percent. If correct, this makes the fast method reliable for computing spectral distortion anisotropies, which are currently too expensive for full Boltzmann solvers.

What carries the argument

The two load-bearing pieces are the effective critical frequency $x_c^{\mathrm{eff}} = x_c^{\mathrm{DC},0} \hat{I}_\mu$ (Eq. 19), with $\hat{I}_\mu$ obtained from the logarithmic derivative of the numerical distortion visibility function relative to the double-Compton-only approximation, and the numerical mu template $M_*(x)$ (Eq. 21), the CosmoTherm number-density distortion at $z_h = 5\times 10^5$ normalized by the mu amplitude. The critical frequency controls the exponential damping of the chemical potential through $d\mu_\infty/d\tau = 1.401\,\dot{Q}/\rho_\gamma - \gamma_N x_c \theta_\gamma \hat{I}_\mu \mu_\infty$; the template replaces the analytic $M(x)$ in the FH basis. Together they correct two separate errors: the rate of mu-to-temperature conversion at high frequencies and the low-frequency photon-emission shape.

What would settle it

Run a full Boltzmann solver (or CosmoTherm) for instantaneous injections at several redshifts, say $z_h = 4\times 10^5$, $10^6$, and $1.5\times 10^6$, extract the normalized number-density distortion $M_*(x)$ at each epoch, and compare the spectra; if the template shape changes by more than a few percent across this range, or if the ratio $\hat{I}_\mu$ deviates from $1.0555$ at $z_h \gtrsim 2\times 10^6$, the claimed universal improvement fails.

Watch

Extended reading notes

Core claim

The central discovery is that the FH treatment can be brought into nearly perfect agreement with the full CosmoTherm solution in the mu-era by correcting two ingredients. First, replacing the analytic critical frequency with the effective value $x_c^{\mathrm{eff}} = x_c^{\mathrm{DC},0} \hat{I}_\mu$, where $\hat{I}_\mu$ is extracted from the numerical distortion visibility function and equals about $1.0555$ at high redshifts, fixes the timing of mu-to-temperature conversion and reduces the high-frequency mismatch below one percent for $z_h \gtrsim 2\times 10^5$. Second, the analytic mu spectrum $M(x)$ is replaced by $M_*(x)$, defined by normalizing the CosmoTherm number-density distortion at $z_h = 5\times 10^5$ to the mu amplitude from the distortion visibility function. This template absorbs the low-frequency double Compton and Bremsstrahlung effects and the frequency-dependent corrections that the constant-amplitude analytic mu does not capture. After a further normalization around the mu maximum, the FH result matches CosmoTherm almost perfectly at both low and high frequencies for injections at $z_h \gtrsim 3\times 10^5$.

Load-bearing premise

The numerical mu template $M_*(x)$ is taken from a single calibration epoch $z_h = 5\times 10^5$ and assumed to hold for all $z_h \gtrsim 3\times 10^5$, and the correction $\hat{I}_\mu$ is extrapolated as a constant $1.0555$ to the highest redshifts; if the mu spectral shape evolves with injection redshift, the improvement does not generalize beyond the calibrated epoch.

Editorial extensions

If this is right

  • The FH method with the corrected template reproduces full mu-era distortions to better than about one percent at high frequencies for $z_h \gtrsim 3\times 10^5$, so mu-distortion anisotropy calculations can rely on the fast method.
  • The low-frequency tail of the mu distortion is shaped by double Compton and Bremsstrahlung emission, and the new template explicitly encodes this, making the FH basis capture photon-production physics it previously missed.
  • The high-frequency energetic part, which dominates the distortion amplitude, is fixed by the effective critical frequency, so the corrected FH tracks the conversion of mu into temperature shift at the right rate.
  • The distortion visibility function at $z \lesssim 2000$ is affected by thermal decoupling, Hubble cooling, and energy branching ratios, so late-time injection scenarios require modeling beyond the analytic $J_{DC}$ approximation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the universal template continues to hold, the same calibration approach could be applied to the y-era and the intermediate era, constructing numerical templates for $Y_*(x)$ and residual distortions to extend FH accuracy to all redshift ranges.
  • The constant $\hat{I}_\mu \approx 1.0555$ at high redshifts suggests the analytic double-Compton critical frequency underestimates the effective photon-emission rate by a fixed factor; a physical derivation of this factor could replace the numerical calibration.
  • The $M_*(x)$ template is calibrated with total energy release $\Delta\rho_\gamma/\rho_\gamma = 10^{-5}$; testing at other injection amplitudes would check the linearity of the template, since the FH is linear in the energy release.
  • If low-frequency free-free and double-Compton emission can be incorporated into the FH basis analytically, the need for a numerical template per cosmology could be removed, making the method fully predictive for arbitrary thermal histories.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. The paper computes CMB spectral distortion Green's functions across cosmic time, with new emphasis on late-time (post-recombination) heating, and studies the distortion visibility function including Hubble cooling and energy branching ratio effects. The main methodological contribution is a proposed improvement of the frequency hierarchy (FH) treatment for mu-era distortions. The authors introduce an effective critical frequency x_c^eff = x_c^{DC,0} I_muhat (Eq. 19), where I_muhat is calibrated to CosmoTherm visibility output, and a new numerical mu spectral template M*(x) obtained by normalizing the CosmoTherm distortion at zh = 5e5 (Eq. 21). With these ingredients they report that the FH treatment matches CosmoTherm within about 1% at high frequencies for single injections at zh > 3e5, while a low-frequency mismatch is acknowledged.

Significance. If the claimed improvement is robust across injection redshifts and general heating histories, it is valuable: the FH treatment is orders of magnitude faster than full codes and is a practical route to spectral distortion anisotropy modeling. The paper usefully extends Green's function and visibility computations into the post-recombination era and clarifies the role of Hubble cooling and heating branching ratios. The central caveat is that the improved FH ingredients are calibrated against CosmoTherm, so the near-perfect agreement is partly a consistency check rather than an independent prediction; the universality of the template and of I_muhat across redshifts and for non-instantaneous injections is the load-bearing point that needs further support. The paper is clearly written and the comparison figures are informative, but the claimed 'almost perfect agreement' is currently demonstrated only for a limited set of instantaneous injections and without quantitative error estimates.

major comments (3)
  1. [§4.4, Eq. (21)] The new template M*(x) is defined by normalizing the CosmoTherm delta-function response at zh = 5e5 with the CosmoTherm-derived visibility Jd(zh), and is then assumed to represent the mu-era spectral shape for all zh > 3e5. Because the validation target is the same code used to construct the template, the agreement at the calibration epoch is partly by construction; the paper does not report the normalized shape M*_{zh}(x) at other redshifts or quantify its evolution. Please show the normalized CosmoTherm spectra at several redshifts (for example zh = 3e5, 1e6, 2e6) and state the maximum shape variation consistent with the claimed sub-percent high-frequency accuracy.
  2. [§4.3, Eqs. (18)-(19)] The correction I_muhat is extracted from the ratio of the numerical d ln Jd/dz to the analytic d ln JDC/dz and then fixed to a constant 1.0555 for the highest redshifts. This is a calibration to CosmoTherm, and no uncertainty on I_muhat or on the constancy of this ratio is provided. The claim that the numerical and analytic visibility curves scale the same at early times should be substantiated by plotting I_muhat(z) over the full range with error estimates, and by a sensitivity test showing how the final residual changes when I_muhat is varied within the expected uncertainty.
  3. [§5, Fig. 9] The stated motivation is spectral distortion anisotropy modeling, where the heating history is a convolution of injections over many redshifts. The paper validates only single narrow-Gaussian injections at discrete zh and explicitly leaves out the low-redshift regime (zh < 3e5) and low frequencies (ν < 20 GHz) from the improved agreement. Because the FH system is linear, demonstrating delta-function completeness at every zh would be sufficient, but the present validation neither covers zh < 3e5 nor tests a non-instantaneous heating history. Please add at least one extended heating test (for example a smoothly decaying injection or a two-redshift superposition) and quantify the residual as a function of zh, or explicitly restrict the scope claim.
minor comments (6)
  1. [§4.2, Eq. (13)] The definition of the normalized profile is written as \hat f = \hat f/f0, which is a self-referential notational error; it should use the unhatted f/f0 or a distinct symbol for the normalized function.
  2. [Fig. 9 caption] The caption says 'as in Fig. 9' but should refer to Fig. 6; the two captions are otherwise nearly identical and should be distinguished.
  3. [§2.2] The phrase 'this process is less inefficient' should read 'less efficient' or 'more inefficient'.
  4. [§2.1] The sentence 'where x = hν/kBTγ is the with dimensionless frequency' contains a typo and should read 'is the dimensionless frequency'.
  5. [§4.3/§4.4] The text would benefit from a single consistent statement about the frequency range of the claimed improvement: §4.3 says the low-frequency mismatch remains unchanged, while §4.4 reports improvements at both low and high frequencies, and the final claim restricts the range to x > 0.1.
  6. [§6 Data availability] For a methods paper, providing the FH code or at least the tabulated M*(x) template and I_muhat values would substantially improve reproducibility; the current statement that data are 'available upon request' is weaker than the standard for numerical methods papers.

Circularity Check

2 steps flagged · score 6.0 of 10

The two central FH improvements, M*(x) and x_c^eff, are calibrated to CosmoTherm, so the near-perfect agreement is partly a consistency check rather than an independent prediction.

  1. self definitional [Section 4.4, Eq. (21), Fig. 9]
    "However, one thing that can be done rather easily is to use the numerical result of CosmoTherm for the number density spectral distortion at zh = 5×10^5 and normalize it with the µ amplitude computed from the distortion visibility function. In this way, we obtain a new numerical spectral shape, M∗(x), as illustrated in the equation below: ∆nCosmoTherm≈ 1.401Jd(zh)∆ργ/ργ M∗(x). (21) Once computed, M∗(x) can be added to the FH and used instead of the analytical version M(x) to achieve better precision."

    M*(x) is defined, by Eq. (21), as the CosmoTherm output at zh = 5×10^5 divided by the CosmoTherm-derived visibility normalization 1.401 Jd(zh) Δρ/ρ. Inserting this template into the FH treatment at the same epoch makes the FH reproduce the CosmoTherm distortion by construction. The claimed 'almost perfect agreement' at zh > 3×10^5 is therefore a consistency check of the assumed redshift universality of M*, not an independent reproduction of the target. The paper does not quantify how the normalized shape varies with zh or test non-instantaneous heating histories, so the predictive content beyond the calibrated epoch is not established.

  2. fitted input called prediction [Section 4.3, Eqs. (18)-(19), Fig. 9 upper panels]
    "Iˆµ≈ (d lnJd/dz)/(d lnJDC/dz) ... The numerical value for Iˆµ, including higher-order corrections, can then be computed using CosmoTherm outputs. For the FH treatment, an improved effective critical frequency xeff_c can now be defined as xeff_c = xDC,0_c · Iˆµ."

    The effective critical frequency is not derived independently: I_muhat is fitted from the ratio of the CosmoTherm visibility derivative to the analytic JDC derivative, and x_c^eff is then set equal to x_DC,0 times that fitted ratio. Through Eq. (16), this forces the FH mu-amplitude decay to follow the CosmoTherm Jd. The improved high-frequency agreement (below ~1% in Fig. 9 upper panels) is therefore partly enforced by construction from the target data, and the text acknowledges 'By matching directly to CosmoTherm, we improved the agreement.' This is a calibrated consistency improvement rather than a parameter-free first-principles prediction.

full rationale

Sections 2 and 3 present numerical Green's functions, visibility functions, and physical effects (Hubble cooling, branching ratios) computed with CosmoTherm; these are self-contained numerical explorations, not circular. The FH basis and Kompaneets matrix in Section 4.1 come from earlier papers and are external inputs, so no load-bearing self-citation chain is present. However, the two new elements that produce the claimed improvement are both calibrated to the code that also defines the validation target. M*(x) is exactly the normalized CosmoTherm distortion at zh = 5×10^5 (Eq. 21), so agreement at that epoch is definitional; and x_c^eff is fitted from CosmoTherm's visibility function (Eqs. 18-19), so the improved mu-to-T timing is not an independent prediction. The remaining non-tautological content is that a single M* template and a constant I_muhat ~1.0555 extrapolate to other zh > 3×10^5 and still match CosmoTherm reasonably well. That is a useful consistency test, but it is tested only against the same code, only for delta-like injections, and the paper itself acknowledges the persistent low-frequency and low-redshift mismatch. Overall, the central claim partially reduces to its inputs by construction, giving partial circularity.

Assumptions & free parameters 5 free parameters · 6 assumptions · 0 invented entities

The central numerical claims rest on the CosmoTherm solver as ground truth, on prior analytic fits for the Bremsstrahlung critical frequency, and on two newly calibrated quantities (I_muhat and M*(x)). No new physical entities are introduced.

free parameters (5)
  • I_muhat ratio = 1.0555 at z > 3e4, extrapolated to high z
    Defined via Eq 18 from the ratio of numerical Jd to analytic JDC; used to define x_c^eff in Eq 19. This is fitted to CosmoTherm output.
  • M*(x) numerical spectral template = CosmoTherm output at zh = 5e5, normalized by 1.401 Jd Delta rho / rho
    The new mu spectral shape introduced in Eq 21 is not derived but tabulated from CosmoTherm, then used in the FH basis.
  • x_c^BR fit coefficients = 1.23e-3 and exponent -0.672
    Adopted from Chluba (2014) numerical fit for the Bremsstrahlung critical frequency (Eq 15), used to compute the total x_c.
  • Gaussian injection width sigma = 0.01 for FH, 0.001 for CosmoTherm
    Choice for approximating a delta injection; the paper argues the effect is small, but the value affects the comparison.
  • Fit amplitudes mu, y, Theta at zh = 5e5 = mu = 1.37e-5, y = 1.63e-9, Theta = -5.96e-8
    Least-squares fit of the CosmoTherm output in the FH basis (Sect 4.4), used to motivate the template.
assumptions (6)
  • standard math The Kompaneets equation and the spectral basis decomposition (Eq 10) are valid for the thermalization problem.
    The paper builds on the FH matrix formulation from Chluba et al. 2023a without re-deriving its validity.
  • domain assumption CosmoTherm solves the thermalization problem accurately and is treated as ground truth for the FH calibration.
    The whole of Sect 4 measures FH errors against CosmoTherm and extracts calibration quantities from it.
  • domain assumption The Chen-Kamionkowski branching ratios chi_h = (1 + 2 Xe) / 3 are valid for the injected energy.
    Section 3.1 uses Eq 8 to redistribute injected energy into heating, excitation, and ionization.
  • domain assumption The free-free and double Compton emission coefficients and the combination x_c^2 = (x_c^DC)^2 + (x_c^BR)^2 describe the critical frequency.
    Section 4.3 adopts these standard approximations and adds the numerically calibrated I_muhat.
  • ad hoc to paper The FH critical frequency is set to zero below z = 4.5e4.
    Section 4.3 states this cutoff with the justification that little mu to T conversion occurs at lower z.
  • ad hoc to paper M*(x) computed at zh = 5e5 applies over the whole mu era zh > 3e5.
    Section 4.4 assumes redshift universality of the new template when adding it to the FH basis.

how reviews work

0 comments
Cite this review

Pith. "Pith review of The late-time heating Green's function and improvements to distortion frequency hierarchy treatment." pith.science (2026). https://pith.science/paper/HULKVFFE

@misc{pith2026250112822,
  author       = {Pith},
  title        = {Pith review of: The late-time heating Green's function and improvements to distortion frequency hierarchy treatment},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HULKVFFE}},
  note         = {Machine review of arXiv:2501.12822}
}
abstract

Early energy injection leaves an imprint on the observed blackbody spectrum of the CMB, allowing us to study the thermal history of the Universe. For small energy release, the distortion can be efficiently computed using the quasi-exact Green's function method. For pre-recombination injections, the Green's function has already been studied previously. Here we reconsider the pre- and post-recombination periods, showcasing both the spectral distortion intensity and the relative temperature difference, which encrypt precious information about physical processes such as free-free interactions and thermal decoupling. We present the associated distortion visibility function, investigating the impact of various physical effects. We then study improvements to the so-called frequency hierarchy (FH) treatment, a method that was developed for the modelling of anisotropic distortions, which like the average distortion signals encode valuable cosmological information. Specifically, the FH treatment has shortcomings even in the $\mu$ era, that in principle should be easy to overcome. In this paper, we introduce a new approach to reduce the mismatch, concluding with a redefinition of the $\mu$ spectral shape using CosmoTherm. This solution takes into account double Compton and Bremsstrahlung effects in the low tail, which can be included in the FH. This opens the path towards a refined modeling of spectral distortion anisotropies.

Figures

Figures reproduced from arXiv: 2501.12822 by the authors.

Figure 1
Figure 1. CosmoTherm results for injections at zh ∈ (2 × 103 , 5 × 106 ) and total energy release ∆ργ/ργ = 10−5 . In the upper panel, the variation of the Green’s function according to the injection redshift can be observed. In the lower panel, the relative temperature difference is illustrated, highlighting the full range of frequencies and transition of the early-type µ- to late-type y￾distortions. Here, negative parts are … view at source ↗
Figure 2
Figure 2. As in [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Relative temperature difference for instantaneous energy injections at three redshifts. Heavy lines are used for the scenario for which the Hubble cooling effect is not included, while light lines show the cases with Hubble cooling. The main difference is observed at low frequencies, where the T(x) drops drastically. equilibrium state, photons have to lose some of their energy, gener￾ating a negative µ ≃ −3 × 10−9 (… view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: Low-redshift distortion visibility functions with different branch￾ing ratios for heating. The heating functions affect both the Te and Ne evolu￾tion, requiring a combined treatment. 3.1 The effects of energy branching ratio Another aspect that should be considered is …
Figure 6
Figure 6. Figure 6: Comparison of the full CosmoTherm results (solid lines) with the approximate FH treatment (dashed lines) for single injections at various heating redshifts and ∆ργ/ργ = 10−5 . For the FH we used a spectral basis up to Y15. The left panel illustrates the distortion and …
Figure 7
Figure 7. Figure 7: Illustrations of how well a Gaussian energy injection history ap￾proximates a Dirac-δ as a function of redshift and for various values of the relative width. At z ≲ 105 , the mismatch is negligible for all cases, while the precision can be hampered significantly early …
Figure 8
Figure 8. Figure 8: Critical frequencies in function of redshift compared to the effec￾tive critical frequency corrected with the numerical values of Iµˆ . The main difference is observed for z ≲ 106 [PITH_FULL_IMAGE:figures/full_fig_p007_8.png]
Figure 9
Figure 9. Figure 9: Spectral distortion signals as in [PITH_FULL_IMAGE:figures/full_fig_p008_9.png]
Figure 10
Figure 10. Figure 10: The low-frequency spectral distortion for various choices of xc in the FH treatment. It shows how using the precise value for Iµˆ in the code influences the approximate solution, but not solving the problem. 4.4 The Numerical µ-distortion template In this section, we …
Figure 11
Figure 11. Figure 11: Distortion for a single ∆ργ/ργ = 10−5 divided by x to highlight the low-frequency difference between the (number-conserving) CosmoTherm solution, the analytical µ-distortion and a direct fit using µ, y and Θ. We fur￾thermore also show the analytical µ-distortion with …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

57 extracted references · 44 canonical work pages

  1. [1]

    H., Chluba J., Hill J

    Abitbol M. H., Chluba J., Hill J. C., Johnson B. R., 2017,

    work page 2017

  2. [2]

    K., Cyr B., Chluba J., 2023, , 523, 1908

    Acharya S. K., Cyr B., Chluba J., 2023, , 523, 1908

    work page 2023

  3. [3]

    K., Khatri R., 2019, , 99, 043520

    Acharya S. K., Khatri R., 2019, , 99, 043520

    work page 2019

  4. [4]

    Bianchini F., Fabbian G., 2022, , 106, 063527

    work page 2022

  5. [5]

    Bolliet B., Chluba J., Battye R., 2020, arXiv e-prints, arXiv:2012.07292

    arXiv 2020

  6. [6]

    Bull P. et al. , 2024, Rhino: A large horn antenna for detecting the 21cm global signal

    work page 2024

  7. [7]

    Burigana C., Danese L., de Zotti G., 1991, , 246, 49

    work page 1991

  8. [8]

    Rev., 48, 1107

    Burigana C., De Zotti G., Feretti L., 2004, New Astro. Rev., 48, 1107

    work page 2004

Show all 57 references
  1. [9]

    Chen X., Kamionkowski M., 2004, , 70, 043502

    2004

  2. [10]

    Chluba J., 2005, PhD thesis, LMU M \"u nchen

    2005

  3. [11]

    Chluba J., 2010, , 402, 1195

    2010

  4. [12]

    Chluba J., 2013 a , , 436, 2232

    2013

  5. [13]

    Chluba J., 2013 b , , 434, 352

    2013

  6. [14]

    Chluba J., 2014, , 440, 2544

    2014

  7. [15]

    Chluba J., 2015, , 454, 4182

    2015

  8. [16]

    Chluba J., 2016, , 460, 227

    2016

  9. [17]

    Chluba J., 2018, arXiv e-prints, arXiv:1806.02915

    2018 arXiv

  10. [18]

    Chluba J. et al. , 2021, Experimental Astronomy, 51, 1515

    2021

  11. [19]

    A., Kamionkowski M., 2015, , 446, 2871

    Chluba J., Dai L., Grin D., Amin M. A., Kamionkowski M., 2015, , 446, 2871

    2015

  12. [20]

    Chluba J., Jeong D., 2014, , 438, 2065

    2014

  13. [21]

    Chluba J., Kite T., Ravenni A., 2023 a , Journal of Cosmology and Astroparticle Physics, 2023, 026

    2023

  14. [22]

    K., 2020, , 498, 959

    Chluba J., Ravenni A., Acharya S. K., 2020, , 498, 959

    2020

  15. [23]

    Chluba J., Ravenni A., Kite T., 2023 b , Journal of Cosmology and Astroparticle Physics, 2023, 027

    2023

  16. [24]

    A., 2012, , 419, 1294

    Chluba J., Sunyaev R. A., 2012, , 419, 1294

    2012

  17. [25]

    K., Chluba J., 2024, , 534, 738

    Cyr B., Acharya S. K., Chluba J., 2024, , 534, 738

    2024

  18. [26]

    Danese L., de Zotti G., 1982, , 107, 39

    1982

  19. [27]

    J., 2003, , 594, L67

    Fixsen D. J., 2003, , 594, L67

    2003

  20. [28]

    J., Cheng E

    Fixsen D. J., Cheng E. S., Gales J. M., Mather J. C., Shafer R. A., Wright E. L., 1996, , 473, 576

    1996

  21. [29]

    Hu W., 1995, arXiv:astro-ph/9508126

    1995 arXiv

  22. [30]

    Hu W., Silk J., 1993, , 48, 485

    1993

  23. [31]

    H \"u tsi G., Hektor A., Raidal M., 2009, , 505, 999

    2009

  24. [32]

    F., Sunyaev R

    Illarionov A. F., Sunyaev R. A., 1975, Soviet Astronomy, 18, 413

    1975

  25. [33]

    Kamionkowski M., Kosowsky A., Stebbins A., 1997, , 55, 7368

    1997

  26. [34]

    A., 2012 a , , 9, 16

    Khatri R., Sunyaev R. A., 2012 a , , 9, 16

    2012

  27. [35]

    A., 2012 b , , 6, 38

    Khatri R., Sunyaev R. A., 2012 b , , 6, 38

    2012

  28. [36]

    A., Chluba J., 2012, , 540, A124

    Khatri R., Sunyaev R. A., Chluba J., 2012, , 540, A124

    2012

  29. [37]

    Kite T., Ravenni A., Chluba J., 2023, Journal of Cosmology and Astroparticle Physics, 2023, 028

    2023

  30. [38]

    J., Meyer S., Spergel D., 2016, in Proc.SPIE, Vol

    Kogut A., Chluba J., Fixsen D. J., Meyer S., Spergel D., 2016, in Proc.SPIE, Vol. 9904, SPIE Conference Series, p. 99040W

    2016

  31. [39]

    Kogut A. et al. , 2011, , 7, 25

    2011

  32. [40]

    Kogut A., et al. , 2024

    2024

  33. [41]

    Komatsu E. et al. , 2011, , 192, 18

    2011

  34. [42]

    Krishna D., Sathyanarayana Rao M., 2024, arXiv e-prints, arXiv:2411.18945

    2024 arXiv

  35. [43]

    Maffei B. et al. , 2021, arXiv e-prints, arXiv:2111.00246

    2021 arXiv

  36. [44]

    Masi S. et al. , 2021, arXiv e-prints, arXiv:2110.12254

    2021 arXiv

  37. [45]

    , 2020, , 641, A6

    Planck Collaboration et al. , 2020, , 641, A6

    2020

  38. [46]

    Remazeilles M., Chluba J., 2018, , 478, 807

    2018

  39. [47]

    Rotti A., Chluba J., 2021, , 500, 976

    2021

  40. [48]

    Rotti A., Ravenni A., Chluba J., 2022, , 515, 5847

    2022

  41. [49]

    Rubi \ n o Mart \' n J. A. et al. , 2020, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 11453, Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, p. 114530T

    2020

  42. [50]

    Sathyanarayana Rao M., Subrahmanyan R., Udaya Shankar N., Chluba J., 2015, , 810, 3

    2015

  43. [51]

    Seljak U., Zaldarriaga M., 1997, Physical Review Letters, 78, 2054

    1997

  44. [52]

    R., 2016, , 93, 023521

    Slatyer T. R., 2016, , 93, 023521

    2016

  45. [53]

    R., Padmanabhan N., Finkbeiner D

    Slatyer T. R., Padmanabhan N., Finkbeiner D. P., 2009, Physical Review D (Particles, Fields, Gravitation, and Cosmology), 80, 043526

    2009

  46. [54]

    A., Zeldovich Y

    Sunyaev R. A., Zeldovich Y. B., 1970, , 7, 20

    1970

  47. [55]

    Trombetti T., Burigana C., 2014, , 437, 2507

    2014

  48. [56]

    Zegeye D., Crawford T., Chluba J., Remazeilles M., Grainge K., 2024, arXiv e-prints, arXiv:2406.04326

    2024 arXiv

  49. [57]

    Zerafa D., 2021, PhD thesis, University of Manchester, U.K

    2021

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.