Stretch to stretch, dust to dust: lower-value local $H_{0}$ measurement from two-population modelling of type Ia supernovae

TL;DR

This work revisits the local Hubble constant measurement by applying a two-population Bayesian hierarchical approach to Type Ia supernovae, leveraging the bimodal SALT2 stretch distribution to separate environmental and intrinsic effects. By linking calibration Cepheid distances to the corresponding supernova population in the Hubble flow and allowing population-specific extinction and luminosity properties, the authors find a Milky Way-like extinction coefficient for the young population ($R_{ m B}\approx 4$) and a lower $R_{ m B}$ for the old population ($ar{R}_{ m B}\approx 3$), with a small intrinsic luminosity offset between populations. The analysis yields $H_{0} \\approx 70.6-71.5$ km s$^{-1}$ Mpc$^{-1}$, depending on modeling choices, reducing the tension with the Planck CMB value by about 30–50% and suggesting that environment-sensitive extinction modeling can reconcile local and early-Universe measurements within a flat $\\Lambda$CDM framework. The results also highlight tensions with the Pantheon+ extinction model and emphasize the importance of calibrator–flow consistency and potential sample-selection effects in precision cosmology with SN Ia.

Abstract

We revisit the local Hubble constant measurement from type Ia supernovae calibrated with Cepheids (SH0ES) by remodelling the supernova data using two supernova populations emerging from the observed bimodal distribution of the SALT2 stretch parameter. Our analysis accounts for population differences in both intrinsic properties (related to possible initial conditions, including supernova progenitor channels) and host-galaxy extinction (expected from well-known environmental differences associated observationally with the two populations). Based on a two-population Bayesian hierarchical modelling of the SALT2 light-curve parameters from the Pantheon+ compilation, we simultaneously constrain intrinsic and extrinsic properties of both supernova populations, match probabilistically the calibration supernovae with the corresponding population in the Hubble flow, and derive the Hubble constant. The difference between the two supernova populations is primarily driven by their mean absolute magnitudes and total-to-selective extinction coefficients. This is related but not equivalent to the traditional mass-step correction (including its broadening for reddened supernovae). The mean extinction coefficient of the supernova population used to propagate distances from the calibration galaxies to the Hubble flow is found to be consistent with the Milky Way-like interstellar dust model with R_B~4 and substantially higher than the extinction model assumed in the SH0ES measurement. Allowing for possible differences between reddening in the calibration galaxies and the corresponding population in the Hubble flow, we obtain H_0=70.59+/-1.15 km/s/Mpc. For the most conservative choice assuming equal prior distributions, we find H_0=71.45+/-1.03 km/s/Mpc. Our reanalysis of type Ia supernovae results in a reduction of the discrepancy with the Planck H_0 by at least 30 and up to 50 per cent (3.5-2.2sigma).

Figures (9)

• Figure 1: Distribution of supernova stretch parameter in the Hubble flow (left) and in the calibration galaxies (right). The stretch distribution in the Hubble flow exhibits a bimodality which is well reproduced by a two-population Gaussian mixture model (the dashed line). The calibration supernovae are consistent with originating from the population associated with the high-stretch peak (young population). The right panel demonstrates this property by comparing the calibration sample to the stretch distribution for type Ia supernovae found in the same comoving volume as the calibration galaxies, and to the corresponding two-population Gaussian mixture model. For the sake of better visual comparison, we keep the same normalisation of the calibration sample and the young population from the calibration volume (right panel). The Hubble constant measurement presented in this work assumes that only the young supernova population (associated with the high-stretch peak of the stretch parameter distribution) is involved in propagating Cepheid distances to the Hubble flow.
• Figure 2: Constraints on hyperparameters of the two-population mixture model from the analysis of type Ia supernovae in the Hubble flow from the Pantheon+ compilation. The red and blue colours denote respectively the old and young populations (associated with the low- and high-stretch peak of the stretch distribution, respectively). The contours show $1\sigma$ and $2\sigma$ credible regions containing 68 and 95 per cent of two-dimensional marginalised probability distributions.
• Figure 3: Comparison between the best-fit two-population model and the supernova data in the Hubble flow. The panels show supernova absolute magnitudes obtained from the standard linear corrections in stretch and supernova colour, as a function of the colour parameter (left) and the stretch parameter (right). For a more direct comparison with the noisy data, the dashed lines show the best-fit model predictions which take into account noise due to measurement uncertainties (omitted for the solid lines). The data on the left panel are restricted to supernovae with $x_{1}>0.5$ or $x_{1}<-0.8$ representing high-purity samples of the underlying young and old populations. The apparent difference between the slopes of the colour correction at $c>0$ in the two populations reflects the difference in the derived mean extinction coefficients $R_{\rm B}$. The right y-axis of the right panel shows the fraction of supernova host galaxies with $\log_{10}(M_{\star}/M_{\odot})>10$. This demonstrates a close connection (although not equivalence) between the two-population model and the traditional mass-step correction Briday2022Wiseman2023. The old population is dominated by high stellar-mass hosts, whereas the young population is widely distributed over host-galaxy stellar masses.
• Figure 4: Constraints on the Hubble constant and selected parameters in model B assuming independent scales of host-galaxy interstellar reddening $\tau$ and scatter $\sigma_{M_{\rm B}}$ in the young supernova population of the Hubble flow (HF) and the calibration (cal) galaxies, and a fixed, Milky-Way like distribution of $R_{\rm B}$ in the calibration galaxies. The best-fit model demonstrates a reduction of the Hubble constant due to a larger reddening scale in the calibration sample than in the Hubble flow ($\tau_{\rm cal}>\tau_{\rm y,HF}$). The contours show $1\sigma$ and $2\sigma$ credible regions containing 68 and 95 per cent of 2-dimensional marginalised probability distributions.
• Figure 5: Comparison between the distribution of supernova colour parameters in the calibration sample and the best-fit model with the reddening scale $\tau$ measured either from the Hubble flow supernovae (model A) or from the calibration supernovae (model B). The best-fit reddening scale in the calibration sample is larger than in the Hubble flow and this preference is driven by a higher fraction of reddened supernovae ($c\sim 0.1$) than in the corresponding young population from the Hubble flow.