[2007.03381] Early recombination as a solution to the $H_0$ tension

Authors:  Toyokazu Sekiguchi, Tomo Takahashi
Abstract:  We show that the $H_0$ tension can be resolved by making recombination earlier, keeping the fit to cosmic microwave background (CMB) data almost intact. We provide a suite of general necessary conditions to give a good fit to CMB data while realizing a high value of $H_0$ suggested by local measurements. As a concrete example for a successful scenario with early recombination, we demonstrate that a model with time-varying $m_e$ can indeed satisfy all the conditions. We further show that such a model can also be well fitted to low-$z$ distance measurements of baryon acoustic oscillation (BAO) and type-Ia supernovae (SNeIa) with a simple extension of the model. Time-varying $m_e$ in the framework of $\Omega_k\Lambda$CDM is found to be a sufficient and excellent example as a solution to the $H_0$ tension, yielding $H_0=72.3_{-2.8} ^{+2.7}\,$km/sec/Mpc from the combination of CMB, BAO and SNeIa data even without incorporating any direct local $H_0$ measurements. Apart from the $H_0$ tension, this model is also favored from the viewpoint of the CMB lensing anomaly.
[PDF]  [PS]  [BibTex]  [Bookmark]

Discussion related to specific recent arXiv papers
Post Reply
Cosmo Comments
Posts: 20
Joined: September 13 2019
Affiliation: N/A

[2007.03381] Early recombination as a solution to the $H_0$ tension

Post by Cosmo Comments » July 26 2020

This paper was commented on through Cosmo Comments. The following comments can also be viewed as annotations on the paper via Hypothesis.

This paper looks at the possibility of addressing the $H_0$ tension by having recombination occur earlier, which in practice is achieved by varying the electron mass $m_e$ within a curved Universe ($\Omega_k\Lambda$CDM). Combining CMB, BAO, and uncalibrated SNeIa measurements, within an ($\Omega_k\Lambda$CDM+$m_e$ Universe, the authors find $H_0=72.3\pm2.8$, which is consistent with the local measurement of Riess et al. (R19). To the best of my knowledge, this is one of the highest values of $H_0$ ever obtained from the CMB+BAO+SNeIa combination (if not the highest altogether). Therefore, if correct, these results would be extremely interesting as they would clearly represent a very compelling solution to the H0 tension.

Nonetheless, I have a few concerns regarding the accuracy of the results, and in particular whether the ingredients required for the authors' solution to work can arise from a realistic theory. I would love to hear the authors' thoughts on this.
  1. My main concern regards $m_e$. First of all, it is unclear to me how the authors have treated $m_e$ in the MCMC. They say that they use CosmoMC "modified to incorporate varying $m_e$". Have they actually varied $m_e$, some function thereof, or something else altogether? And, most importantly, what is the value of $m_e$ they recover from the MCMC? Unless I have missed something obvious, I have not seen this value quoted anywhere, neither in the main text nor in the table. It would be very instructive to quote this value. From the discussion I'm guessing the authors recover a value of $m_e$ higher by $\sim 4-5\%$ compared to the standard value of 0.511 MeV (so probably about 0.535 MeV). If my understanding of the text is correct this raises the question whether such a value of $m_e$ is allowed by any experiments? I have not checked this in detail but I am guessing we have *very precise and accurate* lab constraints on the electron mass? Could the authors therefore please clarify the following:
    • what value of $m_e$ they obtain from the MCMC,
    • what are current external constraints on $m_e$,
    • whether external constraints are compatible with the constraints the authors find are required to solve the $H_0$ tension, and if they aren't consistent, how big of a concern this is?
  2. From Fig. 3, it looks to me as if $m_e$ is actually not helping much in terms of addressing the $H_0$ tension, but most of the work is being done by curvature (compare dark blue against orange curves, or simply the various columns in table I). Could the authors please comment on this?
    Also, it would have been nice to see the 1D and 2D posteriors on $m_e$ and $\Omega_k$, or at least quote the values of $m_e$ and $\Omega_k$ obtained from the MCMC in the main text and in the table. The table only contains the values of $H_0$ and $\Delta \chi^2_\mathrm{eff}$, but I think it would be instructive to include $m_e$ and $\Omega_k$ as well.
  3. Finally, in the Hubble hunter's guide, https://arxiv.org/abs/1908.03663, in Section VC, the authors discuss the fact that high-temperature (which presumably is the same scenarios considered in this paper) is an unlikely solution to the $H_0$ tension. Admittedly, the discussion in this Section is rather qualitative and mostly focused on the fine structure constant, $\alpha$, rather than $m_e$, although I tend to agree with the qualitative conclusion that:

    "It seems highly unlikely that new physics alters $r_s$ by changing recombination, while having an acceptably small impact on the shape of the CMB damping tail. The unlikeliness is also underscored by the fact that recombination occurs out of chemical equilibrium, therefore the relevant atomic per-particle reaction rates are not much faster than the Hubble rate. The particular details of the ionization history resulting from this out-of-equilibrium recombination are marvelously consistent with the shape of the damping tail. Thus the task is more challenging than simply reproducing a generic equilibrium ionization history at a higher temperature."

    Could the authors please comment on this earlier statement? It is not clear to me whether the authors have only focused on equilibrium processes at recombination? Certainly some of their statements around Eq.(9), in particular the reference to [10] (the 2015 Planck paper on variations of the fundamental constants https://arxiv.org/abs/1406.7482) appear to suggest so. For the record, this reference obtains constraints on variations of the electron mass of order $10^{-3}$, which is way less than the $5\%$ the authors quote to solve the $H_0$ tension.

[These comments were shared with us by a member of the cosmology community. They do not necessarily reflect the opinion of the Cosmo Comments team.]

Luke Hart
Posts: 61
Joined: July 13 2015
Affiliation: University of Manchester

Re: [2007.03381] Early recombination as a solution to the $H_0$ tension

Post by Luke Hart » July 28 2020

If the author of this comment is interested, the isolated details for specifically [math] can be found in this paper submitted at the end of last year written by myself and Jens Chluba (https://arxiv.org/abs/1912.03986). Here we looked at updating the fundamental constant picture which was discussed in the Hubble Hunter's Guide by Know and Millea. We also noticed that there was very little discussion of [math], however on its own, it really traces out a vast degeneracy line with the Hubble constant. One should also be careful to notice that the authors here neglected the effects of [math] outside of thermal equilbirium, which they say 'relatively minor'. Though this is roughly correct, it is an approximation of the scenario. One which may have implications for the intense degeneracy between [math] and [math].

If the OP is interested, please email me at [Log in to view email] about the implicit effects of [math] which I suspect, the authors may be asked to expand upon in the updated paper. Jens and I have had discussions with the authors about some of these points and hopefully they will be clarified in later versions.

Post Reply