Cosmological foundations revisited with Pantheon+

TL;DR

This study reanalyzes the Pantheon+ supernova data to compare the standard $\\Lambda$CDM model with the timescape cosmology, introducing new statistical methods and emphasizing the importance of cosmology-independent data analysis.

Contribution

It presents a novel framework for model-independent analysis of supernova data and provides the first evidence that timescape cosmology may fit the data better than $\\Lambda$CDM.

Findings

Timescape cosmology may provide a better fit than $\\Lambda$CDM.

New statistics refine supernova light-curve analysis.

Emphasizes the need for cosmology-independent data reduction.

Abstract

We reanalyse the Pantheon+ supernova catalogue to compare a cosmology with non-FLRW evolution, the timescape cosmology, with the standard $\Lambda$CDM cosmology. To this end, we analyse the Pantheon+ for a geometric comparison between the two models. We construct a covariance matrix to be as independent of cosmology as possible, including independence from the FLRW geometry and peculiar velocity with respect to FLRW average evolution. This framework goes far beyond most other definitions of model independence. We introduce new statistics to refine Type Ia supernova (SNe Ia) light-curve analysis. In addition to conventional galaxy correlation functions used to define the scale of statistical homogeneity we introduce empirical statistics which enables refined analysis of the distribution biases of SNe Ia light-curve parameters $\beta c$ and $\alpha x_1$. For lower redshifts, the Bayesian analysis highlights important features attributable to the increased number of low-redshift supernovae, the artefacts of model-dependent light-curve fitting and the cosmic structure through which we observe supernovae. This indicates the need for cosmology-independent data reduction to conduct a stronger investigation of the emergence of statistical homogeneity and to compare alternative cosmologies in light of recent challenges to the standard model. Dark energy is generally invoked as a place-holder for new physics. For the first time, we find evidence that the timescape cosmology may provide a better overall fit than $\Lambda$CDM and that its phenomenology may help disentangle other astrophysical puzzles. Our from-first-principles reanalysis of Pantheon+ supports future deeper studies between the interplay of matter and nonlinear spacetime geometry in a data-driven setting.