Fit the host-galaxy mass-step location and test alternative host-correlation models

Determine the host-galaxy stellar-mass step location for each Type Ia supernova sample and establish whether a continuous relation between host-galaxy stellar mass and standardized supernova luminosity, or no mass-step correction, provides a more reliable cosmological model than a fixed step at 10^10 solar masses.

Background

The baseline cosmological analysis fixes the host-galaxy mass-step location at 1010 solar masses. The authors find that shifting the step to higher values improves the fit quality, suggesting that the optimal location may be closer to the sample’s median host-galaxy stellar mass.

A robust determination is not performed because the current SNANA pipeline does not provide robust fitting of the mass-step location, and the analysis also does not test continuous mass-dependent corrections or the absence of a mass-step. These unresolved modeling choices matter because host-galaxy correlations affect standardized Type Ia supernova distances and therefore cosmological inference.

References

Ideally, the mass-step location should be fit for each SN Ia sample, but the current \verb|SNANA| pipeline does not provide a robust fitting of the mass step location. When we fit for a mass-step using our analysis pipelines on simulated data, we see a bias in recovered cosmological parameters, at least in part due to the built-in assumption that the dust properties of galaxies change at $10{10} M_{\odot}$ in the dust model for SN-Unite \citep{popovic2023dust2dust}. We similarly do not consider alternative forms of the mass-step, such as fitting a line as a function of the host-galaxy stellar mass, or not assuming a mass-step at all, and leave such investigations to future work.

— Supernovae Unite: Host-Galaxy Mass Measurements of Type Ia Supernovae and Their Impact on Cosmology  (2609.05321 - Lee et al., 4 Sep 2026) in Section 4.2, “Sensitivity to Analysis Choices”