Happy New Year to all my readers and here’s the first entry of 2021! I cover long-range neutrino interactions in cosmology, measuring the Hubble constant from surface brightness fluctuations, and constraining dark matter interactions with CMB spectral distortions. Enjoy the read and hope your 2021 got off to a good start!
#1 2101.05804: Long Range Interactions in Cosmology: Implications for Neutrinos by Ivan Esteban and Jordi Salvado
In my last post of 2020, I covered massive neutrino self-interactions in the context of the Hubble tension, which for the record these interactions aren’t able to solve without conflicting with both cosmological data and particle physics constraints (see also the very relevant 2012.11830 which appeared during the holidays). The interactions studied in the context of the Hubble tension, however, are short-ranged ones, i.e. mediated by heavy particles which one can effectively “integrate out” to get a four-point contact interaction, much like the electroweak bosons can be integrated out at energies much lower than 100 GeV to give Fermi’s 4-fermion theory with an effective strength G_F (which results from taking the limit of the mediator propagator for mediator mass much larger than the momenta involved). However, the absence of new physics in the charged lepton sector suggests that new neutrino interactions, if they exist, should be mediated by light particles and hence be long-ranged, as studied in 1905.02727. The motivation for this week’s paper is therefore to revisit the possibility that neutrinos, and more generally fermions, might be subject to additional long-range interactions, and what the effect of these on cosmological observables might be.
The rest of this week’s paper is a rather technical study of these interactions in the limit where long-range interactions (yet with range shorter than cosmological scales) mediated by a light scalar do not allow one to assume the ideal gas approximation. There are a bunch of very interesting phenomena which occur, such as scalar-induced contributions to the neutrino masses, or a transient regime between neutrinos behaving as radiation (equation of state w=1/3) in the early Universe and as matter (w=0) in the late Universe, where the scalar field dominates, and the fermion-scalar system effectively has an equation of state w<0. The main implication for cosmology is that in the presence of these long-range interactions the cosmological neutrino mass bound is considerably weakened. There are therefore at least two interesting consequences which highlight a very cool interplay between cosmological and laboratory searches for neutrino properties:
-the KATRIN experiment, nominally sensitive to the neutrino mass scale at the level Mnu>0.6 eV, detects neutrino masses and hence contradicts the current cosmological bound Mnu<0.12 eV. This apparent discrepancy could be resolved if neutrinos actually feel long-range interactions with strength of order 10^3-10^6 eV^-1.
-future large-scale structure experiments such as EUCLID and DESI fail to detect the neutrino mass despite their targeted sensitivity of sigma_Mnu~0.02 eV or better. This apparent discrepancy with oscillation experiments which require Mnu>0.06 eV could be solved if neutrinos actually feel long-range interactions with strength of order 10^3 eV^-1.
Overall, this is a very cool paper which provides an answer to a question I’ve always asked myself: “What if future cosmological probes don’t get the promised >5σ detection of neutrino masses and hence contradict oscillation experiments?”, with the long-range interaction scenario studied here definitely an interesting and valid possibility. We even had a topical discussion on precisely this question at Cosmology from Home 2020 (though unfortunately as per the code of conduct I am not allowed to share the Youtube link for the recording of the discussion). Kudos to the authors for also making their modified version of the Boltzmann solver CLASS publicly available!
#2 2101.02221: The Hubble Constant from Infrared Surface Brightness Fluctuation Distances by John Blakeslee et al.
The use of surface brightness fluctuations (SBF) to measure distances is an old one. How this works is that essentially you are measuring the small-scale spatial variance in the pixel intensity in the image of a galaxy. This variance is due to the fact that galaxies are not continuous objects, but rather are made up of discrete stars. The amplitude of the fluctuations depends on many stars fall within each pixel, a quantity which varies stochastically but which overall depends on far the galaxy is, i.e. on the metric area sampled by each pixel. If one is able to independently determine the absolute SBF magnitude (a complicated task which requires making assumptions about the underlying stellar population), the apparent SBF magnitude can then be used to infer the distance to the galaxy. Alternatively, apparent SBF measurements can be used to go into the Hubble flow and therefore construct a Hubble diagram (much like the Pantheon sample of Supernovae), with the zero-point calibrated using external calibrators (such as Cepheids, TRGB, masers, and so on). For a seminal review on SBF in this context, see 1202.0581 by Blakeslee, first author of this week’s paper.
In the context of the Hubble tension, SBF have not received as much attention, although the summary plot from the well-known KITP conference on the Hubble tension did quote a preliminary SBF measurement of H0=76.5±4.0 km/s/Mpc. Another important work on measuring H0 with SBF comes from 2001 (!) in astro-/0011288, finding H0=72.0±2.3, which I have to say is a surprisingly small error bar for the time. This week’s paper presents the most up-to-date measurement of H0 from SBF, giving H0=73.3±0.7±2.4, using a combination of the calibrators I mentioned previously to calibrate the SBF zero-point. While the error bars are relatively large compared to other measurements, the central value falls in line with other local measurements such as Cepheid-calibrated Supernovae and water masers. For discussions on why this value is lower than the 76.5±4.0 quoted at the KITP conference, which presented a preliminary number for this project, see the beginning of Section 5. This measurement had been in the making for quite a while, so it is nice to finally see it out here. Even with its comparatively large error bars, it still adds even more fuel to the Hubble tension (which in itself doesn’t need any more fuel than it already has!).
#3 2101.04070: Testing dark matter interactions with CMB spectral distortions by Yacine Ali-Haïmoud
Spectral distorsions (SDs) in the Cosmic Microwave Background (CMB), i.e. deviations in its frequency spectrum from a perfect black-body, are a promising way to test new physics, as virtually any process which injects energy in the photon-baryon plasma at z>2*10^6 is guaranteed to lead to spectral distortions (the question is of course whether they are observable or not). Current limits on spectral distortions are as old as 1996 and come from COBE/FIRAS, although future experiments such as PIXIE are expected to improve these limits by several orders of magnitude. One interesting use of SDs is to probe interactions of dark matter (DM) with Standard Model particles. In fact, scattering of a non-relativistic DM particle with photons, electrons, and nuclei, effectively imprints a negative chemical potential on the CMB frequency spectrum. In general SDs are a largely untapped resource of cosmological information. If you want to learn more, I recommend you watch Deanna Hooper’s excellent talk which featured in Shaun Hotchkiss’ Cosmology Talks Youtube series (whose channel you should anyhow subscribe to if you are a fan of my blog!).
An early detailed study of this effect was presented by Ali-Haïmoud, Chluba, and Kamionkowski in 1506.04745. However, this study made several simplifying assumptions on the heat-exchange rate between DM and the photon-baryon plasma, which in itself depends on the DM particles velocity distribution and thermal decoupling, themselves depending on the DM interactions with both other particles and itself. This previous paper assumed instantaneous decoupling at a certain redshift. Clearly this is a crude approximation, and the formalism for improving on this approximation was presented by Ali-Haïmoud in 1811.09903. Armed with these tools, in this week’s paper Ali-Haïmoud revisits previous constraints and forecasts on DM interactions from SDs, finding that the previous approximations systematically and significantly underestimated the amplitude of SDs, meaning that the previous constraints were actually too pessimistic. More details on the actual numbers are presented in the paper. This paper also comes with the public code DMDIST, which allows to compute SDs for generic particle DM models, specified by their cross sections for elastic scattering with and annihilation into Standard Model particles. Ali-Haïmoud writes that this code “should provide a valuable addition to the dark matter hunter toolkit”. I couldn’t agree more.