This week’s entry discusses a new measurement of the Hubble constant from Type II Supernovae, a map from Poincaré gauge theories to bi-scalar-tensor theories useful for cosmological applications, and a map between fluid and field approaches to interacting dark energy valid at the background and first-order perturbation level.
#1 2006.03412: A measurement of the Hubble constant from Type II supernovae by Thomas de Jaeger et al.
In this blog I have written quite a bit about the Hubble tension, the inconsistency between local (direct) and inverse distance ladder estimates of the Hubble constant H0. However, I have so far exclusively focused on proposed theoretical solutions, which likely operate prior to recombination and only affect the inverse distance ladder estimate, without touching the local distance ladder bit. It is nonetheless worth questioning whether there might be problems with the most widely discussed local distance ladder measurements, based on Type Ia Supernovae (SNeIa) in the Hubble flow calibrated via Cepheid variables. In other words, it is important to develop as many independent local distance ladder measurements of H0 as possible: after all, this is how science works. In the search for independent local distance ladder measurements, one possibility is the use of alternative calibrators: worth mentioning in this sense is the use of the Tip of the Red Giant Branch in place of Cepheids to calibrate SNeIa in the Hubble flow (see e.g. 1907.05922). Besides using a different calibrator, another possibility is to use a different standard(izable) candle in first place, i.e. something other than SNeIa. One possibility is to use Type II Supernovae (SNe II) as standard(izable) candles to measure H0. Unlike SNeIa which arise from accretion onto a white dwarf, SNe II arise from red supergiant stars, and are characterized by strong hydrogen features. While SNeIa have been more widely studied as standardizable candles, SNe II have other advantages in that they are more abundant than SNeIa and have better understood progenitors and environments.
In this week’s paper, de Jaeger et al. use the largest SNe II sample in the Hubble flow (consisting of 125 SNe II from various surveys) to measure H0. These SNe II are calibrated via Cepheid variables or the Tip of the Red Giant Branch (in particular, at least 7 calibrators are available). Much as SNeIa, SNe II peak luminosities need to be standardized, and various methods for doing so have been proposed. In this work, de Jaeger and collaborators use the so-called “standard candle method” (SCM), first developed by Hamuy & Pinto in astro-ph/0201279 (Hamuy was de Jaeger’s PhD advisor), which exploits an empirical correlation between luminosity and expansion velocities. Constructing the local distance ladder in this way, de Jaeger and collaborators find H0~75.8±5.8 km/s/Mpc (statistical and systematics, I’ll suppress these units in the following). This is a 7.6% measurement of H0 (the 6.5% quoted in the abstract is based on statistical errors only), and represents the most precise ever measurement of H0 from SNe II. One of the previous most precise measurement was performed in 1812.04982 which finds H0=71±8 (using a different standardization method known as PMM). This week’s paper’s improvement is mostly due to a larger number of calibrators and the use of the largest SNe II sample in the Hubble flow. The central value of the inferred H0 is consistent with the local Cepheid-calibrated SNeIa value of H0=74.03±1.42, which would appear to suggest that SNeIa are not at the source of the H0 tension. However, since the error bars are very large, this measurement is also nominally consistent (at 1.4 sigma) with the Planck value assuming LCDM, H0=67.4±0.5. This type of measurement certainly holds a huge potential moving forward. With upcoming surveys such as VRO (formerly LSST), both the number of calibrators and SNe II in the Hubble flow will increase, improving the precision of the value of H0 one can infer. At that point, one could imagine that SNe II might play an important role in unraveling the origin of the Hubble tension.
#2 2006.03581: Mapping Poincaré cosmology to Horndeski theory for emergent dark energy by Will Barker et al.
We are used to thinking about coordinate transformations as being an external global symmetry. But what happens if one decides to gauge these? Let me elaborate a bit more. Given a certain manifold with a metric on it, one can identify coordinate transformations which do not change the form of the metric, and are therefore referred to as the isometry group of the given manifold. In the case of Minkowski space M4, the isometry group is the so-called Poincaré group, a 10D non-compact Lie group characterized by 10 constant parameters describing Lorentz rotations, translations, and boosts. This symmetry is global in that the group parameters are constant across the whole manifold. Now, anyone who has taken a basic QFT course knows that one of the simplest “upgrades” to a global symmetry is to gauge it, i.e. make it local by promoting the constant group parameters to being function of spacetime. In the case of internal symmetries (such as the SU(3)xSU(2)xU(1) symmetry of the Standard Model) it is well known that this leads to the promotion of partial derivatives to covariant derivatives, which requires introducing gauge fields, which then become the carriers of the resulting forces, and so on and so force (pun and not typo).
But what happens if one chooses to gauge an external spacetime symmetry instead of an internal symmetry (I’m pretty sure anyone taking a QFT course must have asked themselves this question at some point)? Is that even possible? It is, in fact, possible to gauge the Poincaré group, something which was done back in the 50s and 60s by Kibble, Sciama, and Utiyama. The result is the so-called Poincaré gauge theory (PGT), which importantly introduces torsion onto spacetime, which on the other hand has to be of the Riemann-Cartan type in order for torsion itself to be accommodated. This all sounds extremely cumbersome and, in fact, it is, beginning with the fact that there are 58 PGTs which are power-counting renormalizable and free from ghost/tachyonic instabilities (i.e. theoretically sound). Several of these were classified in the earlier 2003.02690 by the same authors, a technical tour de force to which a quick glance is enough to give most a headache. As this class of theories is so cumbersome, it is of no surprise that cosmological studies thereof have been quite limited and discontinuous. The question therefore is: can one recast the cosmology of PGTs into a more manageable form, perhaps in terms of an analogue theory which can mimic PGTs at least at the background level?
This is the question Barker and collaborators set out to address in this week’s paper. They focus on the 10-parameter quadratic PGT, usually known as PGTq,+ (keeping in the theme of current events, the name sounds like a dangerous new virus). They then try to construct a so-called metric analogue (MA) which reproduces the spatially-flat background cosmology of PGTq,+. The Lagrangian of PGTq,+ is given in Eq.(4) and the MA Lagrangian turns out to be given by Eq.(12). In the Jordan frame, this is a non-canonical bi-scalar-tensor theory, sometimes referred to as bi-galileon (even though this is a slight abuse of nomenclature). The MA is simpler to deal with than the original PGTq,+ theory, and makes it easier to study cosmological implications of the latter. For instance, it is found that the Class 2A* of PGTq,+ (for what this means, see the top of Page 4) not only generates emergent dark energy, but also contains a dark radiation component which could have something to say about the Hubble tension. This map from Poincaré gauge theories to bi-scalar-tensor theories (I prefer not to call them Horndeski theories as in the paper’s title, as I find this a slight abuse of nomenclature) at the cosmological level will certainly prove very interesting and useful for cosmological studies of the former, and helps making an extraordinarily rich but otherwise cumbersome theory more accessible for phenomenological studies. The only concern I have at this point is that this map only holds (as far as I have understood) at background level, but not at the level of linear perturbations (in this sense, a question in this spirit is addressed in paper #3 I summarize below). However, a detailed study of the latter is unavoidable for a complete analysis of the cosmology of a given model, and I guess it is probably something the authors plan on doing for future studies.
#3 2006.04618: Cosmological perturbations in the interacting dark sector: Mapping fields and fluids by Joseph Johnson and S. Shankaranarayanan
The discussion about this paper closely relates to an earlier paper (2001.03120) I discussed previously in my Week 2 entry, in connection to the question (on which I will expand shortly): "Do phenomenological interacting dark matter-dark energy models have a well-defined field theoretical origin?” There has been lots of recent interest in so-called interacting dark energy (IDE) models, where dark matter (DM) and dark energy (DE) interact non-gravitationally. The way IDE models are usually dealt with in cosmological studies is not from first-principles (i.e. starting from a well-motivated Lagrangian) but simply coupling the DM and DE continuity equations. This leads to energy exchange between DM and DE, implying that only the sum of the DM and DE energy-momentum tensors is covariantly conserved, rather than the two individually. The DM-DE energy exchange is governed by a coupling function (usually called Q), and at the phenomenological level people usually consider functional forms such as Q~H*rho, where H is the Hubble parameter and rho is a quantity with dimensions of energy density (usually the DM or DE energy density, or some combination of them). While such couplings have some motivation from coupled quintessence models, most of the motivation for them is that these phenomenological couplings are simple to handle. The pressing question then is: can these phenomenological couplings arise from a well-motivated field theory, in other words from a well-motivated Lagrangian/action? This question was partially addressed in 2001.03120, where starting from a given phenomenological choice of Q, the authors reconstructed the specific action (and more precisely DM-DE interaction potential) which can give the same background expansion as one would get from this choice of Q. However, this begs the question: “can this Q-field theory correspondence be established beyond the background expansion, i.e. at the level of cosmological perturbations?”
In this week’s paper, Johnson and Shankaranarayanan address precisely this question, and try to find a mapping between phenomenological choices of the DM-DE coupling Q, and well-defined IDE field theories, with the mapping holding both at background and first-order perturbations level. Johnson and Shankaranarayanan work with an action which is of the f(R,h) form in the Jordan frame, where R is the Ricci scalar and h (called chi in the paper) is a scalar field which plays the role of DM. In the Einstein frame, see their Eq.(5), one manifestly sees how this action indeed leads to a IDE model. Finally, their Eqs.(20,38,39) determine the Q-action mapping at first order in perturbations. What they find is that for this specific type of theory (i.e. f(R,h) in the Jordan frame, which is anyhow very well motivated), such a mapping only exists provided Q is of a specific form. This allows them to then classify a number of forms of Q often considered in the literature depending on whether or not they allow a f(R,h) field theory description or not. Their table spanning pages 11 and 12 is very useful and impressive, considering 31 forms of IDE considered in the literature (with extensive references), and classifying these according to whether they allow a f(R,h) field theory description or not. Unsurprisingly, some well-motivated models such as coupled quintessence do indeed allow such a description, but many other models do not. Overall, this was a rather interesting paper which provides further field theoretical motivation for a number of well-studied choices of IDE coupling Q, which would otherwise be purely phenomenological.