The Cosmological Principle May Be False

The cosmological principle holds that at large enough scales, the universe is homogeneous and isotropic (i.e. spherically symmetrical). But, there is meaningful evidence from astronomy observations of anisotropy at the largest observable scales in the universe. 

Unsurprisingly, this also restates a problem the the ΛCDM "standard model of cosmology".

On the assumption that quasars (QSO) and gamma-ray bursts (GRB) represent standardisable candles, we provide evidence that the Hubble constant H0 adopts larger values in hemispheres aligned with the CMB dipole direction. The observation is consistent with similar trends in strong lensing time delay, Type Ia supernovae (SN) and with well documented discrepancies in the cosmic dipole. Therefore, not only do strong lensing time delay, Type Ia SN, QSOs and GRBs seem to trace a consistent anisotropic Universe, but variations in H0 across the sky suggest that Hubble tension is a symptom of a deeper cosmological malaise.

Orlando Luongo, et al., "On Larger H0 Values in the CMB Dipole Direction" arXiv:2108.13228 (August 30, 2021).

The introduction to the paper lays out the issues well:

Persistent cosmological tensions suggest that it is timely to reflect on the success of the flat ΛCDM cosmology based on Planck values. In particular, a ∼ 10% discrepancy in the scale of the Hubble parameter in the post Planck era, if true, belies the moniker “precision cosmology”. Recently, the community has gone to considerable efforts to address these discrepancies, but proposals are often physically contrived. Great progress has been made in cosmology through the assumption that the Universe is isotropic and homogeneous, namely the Cosmological Principle or Friedmann-Lemaˆıtre-Robertson-Walker (FLRW) paradigm. Nevertheless, cosmological tensions point to something being amiss. Here, we present evidence that FLRW is suspect. 

The Cosmic Microwave Background (CMB) dipole is almost ubiquitously assumed to be kinematic in origin, i. e. due to relative motion. By subtracting the dipole, the CMB is defined as the rest frame for the Universe. Some of the CMB anomalies have been documented in and refer to anomalies with directional dependence, for example the (planar) alignment of the quadrupole and octopole and their normals with the CMB dipole. In addition, it has been argued that an anomalous parity asymmetry may be traced to the CMB dipole, so a common origin for CMB anomalies is plausible. 

Separately, attempts to recover the CMB dipole from counts of late Universe sources such as radio galaxies and QSOs, which are assumed to be in “CMB frame”, largely agree that the CMB dipole direction is recovered, but not the magnitude. The implication is that observables in the late Universe are not in the same FLRW Universe. Independently, similar findings have emerged from studies of the apparent magnitudes of Type Ia supernovae (SN) and QSOs. In contrast, analysis of higher CMB multipoles confirms the CMB dipole magnitude. It should be stressed that although the statistics may be impressive, these results are based on partial sky coverage and this is an important systematic. 

Without doubt, the bread and butter of FLRW cosmology is the Hubble parameter H(z). In particular, Hubble tension casts a spotlight on H(0) = H(z=0). Here, we build on earlier observations for strongly lensed QSOs and Type Ia SN that H(0) values in the direction of the CMB dipole, loosely defined, are larger. Similar variations of H(0) across the sky have been reported for scaling relations in galaxy clusters. Note, within FLRW the value of H(0) is insensitive to the number of observables in any given direction, but of course the number of observables impacts the errors. Finally, a variation in H(0) across the sky recasts the Hubble tension discussion as a symptom of a deeper issue. 

Our findings are that QSOs and GRBs, on the assumption that they represent standardisable candles, return higher H(0) values in hemispheres aligned with the CMB dipole direction. Admittedly, in contrast to Type Ia SN, QSOs and GRBs are non-standard, but if they are merely good enough to track H(0), namely a universal constant in all FLRW cosmologies, then we arrive at results that contradict FLRW. The physics of strong lensing time delay, Type Ia SN, QSOs and GRBs are sufficiently different with different systematics. It is hence plausible that the Universe is anisotropic.