http://arxiv.org/abs/2301.01024
We test the homogeneity of the Pantheon+ sample with respect to the intrinsic absolute luminosity $M=m_{Bi}-\mu_i$ of the type Ia supernovae (SnIa) in Cepheid hosts and in the Hubble flow. Here, $m_{Bi}$ is the corrected/standardized SnIa apparent magnitude and $\mu_i$ is the $i^{th}$ SnIa distance modulus obtained either from Cepheids (for SnIa in Cepheid hosts) or from the parametrized Hubble expansion rate $H(z)$ (for the rest of the SnIa). When $M$ is allowed to take a single value in the context of flat \lcdm cosmological background $H(z)$, we find the expected best fit values $M=-19.25\pm 0.03$, $\Omega_{0m}=0.33\pm 0.02$, $H_0=(73.4 \pm 1)$~km~s$^{-1}$~Mpc$^{-1}$ consistent with the original analysis of Brout et. al. When we introduce a new degree of freedom allowing $M$ to take two values, one ($M_<$) for nearby SnIa (distance $d_i<d_{crit}$, $\mu_i<5\;log_{10}(d_{crit}/Mpc)+25$) and one ($M_>$) for more distant SnIa, we find a $2-3\sigma$ tension between the two best fit values of $M_>=-19.215\pm 0.03$ and $M_<=-19.362\pm 0.05$ for $d_{crit}\simeq 20Mpc$. However, in contrast to the pure SH0ES data, this degree of freedom does not affect significantly the best fit values for the cosmological parameters $H_0$ and $\Omega_{0m}$ obtained from Pantheon+, for any value of $d_{crit}$, due to the dominant effects of the covariance matrix. When $M$ is allowed to take distinct values $M_i$ for each SnIa in Cepheid hosts we find using a KS test, that the $M_i$ of nearby SnIa ($d_i<20Mpc$) have less than 2.5\% probability to have been drawn from the same probability distribution as the $M_i$ of more distant SnIa ($d>20Mpc$). These results constitute hints of inhomogeneities in the Pantheon+ sample which could be due to large statistical fluctuations, unaccounted systematic effects or new physics.
L. Perivolaropoulos and F. Skara
Wed, 4 Jan 23
28/43
Comments: 9 pages, 3 figures, The numerical files for the reproduction of the figures can be found at this https URL