Supernovae vs. the Cosmological Principle: A New Verdict ✛ ✎ ✦
Context
A cornerstone of modern cosmology is the cosmological principle, which asserts the large-scale homogeneity and isotropy of the Universe. Its robustness is supported by observations of the cosmic microwave background — the echo of the Big Bang — yet several anomalies, such as hemispherical power asymmetry, invite scrutiny of this paradigm. Type Ia supernovae, as standard candles, allow us to map the expansion of the Universe and test its isotropy: according to Hubble's law, discovered by Edwin Hubble, and the Friedmann equations, in an isotropic model the relationship between redshift and distance should be the same in all directions. Admixtures of dark matter and dark energy merely scale the expansion rate without breaking its symmetry. Detecting a preferred direction would therefore challenge the standard cosmological model.
Methods
The researchers analyzed the Pantheon+ compilation — a collection of 1701 supernovae from 20 different surveys — dividing it into six subsamples by origin (Low-z, PS1, SDSS, SNLS, DES, High-z). Two independent methods were applied. The first was dipole fitting: a correction with amplitude $A_D$ and a direction vector was introduced into the theoretical distance modulus, and Markov Chain Monte Carlo (MCMC) simulations were used to find best-fit parameters, including matter density $Ω_m$, the Hubble constant $H_0$, and the dipole direction. To break the degeneracy between $H_0$ and absolute luminosity, Cepheid calibrators from the SH0ES project, led by Adam Riess, were employed. The second method was hemisphere comparison: the sky was split by a random plane into two halves, $Ω_m$ was determined for each, and the anisotropy level AL was computed. This procedure was repeated for nearly 200,000 directions to find the maximum signal. Significance was assessed using isotropic simulations — 5000 mock catalogs that preserve the redshift distribution and covariance matrix of the real data but lack any intrinsic anisotropy.
Results
Dipole analysis of the full sample found no significant signal: the amplitude is constrained to $A_D$ < 0.764×10⁻³ (95% CL). However, the Low-z subsample (z < 0.093) showed a non-zero amplitude of $A_D$ = 0.952+0.454−0.403×10⁻³ with a direction of (l, b) = (149.77°, −12.20°), corresponding to a significance of ~2.27σ. An even stronger signal emerged from the combination of four surveys (IDs 5, 56, 63, 150): $A_D$ = 1.730+0.554−0.715×10⁻³ with a direction of (153.05°, −1.25°) and significance 2.79σ. Excluding Low-z from the full sample drastically shifts the dipole direction by 138°. The hemisphere comparison method for the full sample yielded a maximum anisotropy level of AL_max = 0.289±0.052 in the direction (127.97°, 17.90°) with a significance of just 1.56σ. Notably, this direction nearly coincides with that of the SNLS subsample (difference 4.45°), while the Low-z and High-z subsamples instead weaken the signal along this axis. Combining all three (Low-z, SNLS, High-z) reproduces the full-sample direction to within 1.5°, highlighting the dominant role of SNLS.
Implications
The results indicate that the apparent cosmic anisotropy is most likely an artifact of uneven sky coverage by different surveys, rather than a property of the expansion of the Universe. This is reminiscent of the situation with galactic coordinates: if you look at the night sky from a city, the Milky Way seems asymmetric, but the cause is the distribution of streetlights, not the structure of the Galaxy. Here, too, local inhomogeneities in the sample, especially the contribution of SNLS, mimic a preferred direction. Thus, dark energy and dark matter remain within the isotropic paradigm, and the cosmological principle withstands the test, as anticipated by pioneers of supernova studies like Saul Perlmutter.
Future development
The next generation of telescopes — such as the Vera C. Rubin Observatory (LSST) — will provide far more uniform supernova surveys, allowing either confirmation or definitive exclusion of weak anisotropic signals on large scales. Moreover, improvements in statistical analysis methods and systematic error control will reduce the impact of data inhomogeneity.
Impact
These findings will influence the interpretation of data on the expansion of the Universe and constrain models that predict anisotropy in dark energy.
Next steps
Detailed analysis of Pantheon+ subsamples, accounting for redshift effects and modeling of local structures (voids and superclusters), will help disentangle kinematic and gravitational contributions to anisotropy. Combining supernova data with the cosmic microwave background and the large-scale structure of galaxies will be key.
Key open problems
The article directly addresses the validity of the cosmological principle — one of the unresolved enigmas in physics. If the anisotropy proves real, it could point to new fields or modifications of Hubble's law on large scales, as well as a need to revise the Friedmann equations.
The knowledge machine's view
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
The study uses the Pantheon+ compilation (1701 Type Ia supernovae) to test the isotropy of cosmic expansion via dipole fitting and hemispherical comparison. The full sample shows no significant deviations, but a weak dipole signal (~2σ) is found in the Low-z subsample, which the authors attribute to uneven data coverage, particularly the dominance of SNLS.
Where its strength is
A key strength is the careful treatment of systematic effects related to sample inhomogeneity, and the demonstration of how these can mimic anisotropy. This underscores the importance of homogeneous surveys for testing the cosmological principle.
Where it could go next
- It might be interesting to compare the obtained constraints with the anisotropy analysis of cosmic acceleration in 2605.03004v1, which uses a different dataset and methods, to see how conclusions depend on the choice of tracers.based on: Ocean of Currents: Why the Universe's Acceleration Varies with Direction arXiv:2605.03004v1
- Related work 2607.01172v1 on systematic errors in distance calibration is nearby; it might be worth examining how accounting for such effects would impact the dipole signal in Low-z.based on: A Universe Without Funhouse Mirrors: How a Ruler Error Spawned Cosmic Monsters arXiv:2607.01172v1
- In light of the discussion on data inhomogeneity, the work 2606.24886 on cosmic topology could offer alternative statistics for testing isotropy that are less sensitive to uneven sky coverage.based on: Topology of the Universe: Listening to the Cosmic Drum arXiv:2606.24886
Worth trying
- Apply the methods used in 2605.03004v1 to the Pantheon+ data to compare constraints on dipole amplitude.
- Check how the systematic error correction proposed in 2607.01172v1 affects the significance of the dipole signal in the Low-z subsample.
- Use statistics robust to incomplete sky coverage from 2606.24886 to re-analyze anisotropy.
Why this may matter
Such tests help distinguish genuine cosmological signals from data artifacts, which is critical for confidence in the standard model and for interpreting future surveys like LSST.
Nearby works in our archive
- A Universe Without Funhouse Mirrors: How a Ruler Error Spawned Cosmic Monsterssimilarity 0.64
- Ocean of Currents: Why the Universe's Acceleration Varies with Directionsimilarity 0.64
- Gamma-Ray Bursts and Supernovae Find No Trickery in Dark Energysimilarity 0.61
- How Supernovas Ignite from Withinsimilarity 0.61
- The Universe Refuses to Agree on the Numberssimilarity 0.60
- Quantum Bounce: How the Universe Became Uniformsimilarity 0.60
- Topology of the Universe: Listening to the Cosmic Drumsimilarity 0.60
The closest work in the archive sits at 0.64 by meaning — the usual level for our corpus.
Pantheon+ can be compared to a patchwork quilt stitched from 20 different 'fabrics' — astronomical surveys. Each patch has its own texture and covers its own piece of the sky, and the brightest 'patches' (SNLS) set the tone for the whole pattern.
Key numbers
- Low-z dipole amplitude: 0.952×10⁻³
- Dipole signal significance: 2.27σ
- Anisotropy level ALmax: 0.289
- Number of supernovae in Pantheon+: 1701
- Direction difference after excluding Low-z: 138°