Revision 1

Superseded · first revision · record

Title: Are evolving dark energy hints driven by supernova systematics?
Plain: Exploding stars called type Ia supernovae serve as distance markers, and the dark energy hint depends on which supernova catalog is used. Small calibration errors between nearby and distant supernovae could fake the signal.
Precise: The DESI DR2 preference for $w0-wa$ ranges from about $2.8\text{ to }4.2\sigma$ across Pantheon+, Union3 and DES-Y5, and the 2025 DES reanalysis (Dovekie) lowered the DES value. Determine whether a distance-modulus offset of order 0.04 mag between low-redshift ($z < 0.1$) and high-redshift supernovae, from photometric calibration, selection, or host-galaxy dust and age evolution, accounts for the signal. Answer: identification of such a systematic or a demonstration that all such offsets are below $0.01\,\mathrm{mag}$.
Settled by: A single-survey supernova sample spanning $0 < z < 1$ on one photometric system, cross-checked against BAO-only and CMB-only fits.
Kind: well-posed
Answer type: identification
References: (empty)
Literature status: contested
Status note: The 2025 DES-Dovekie recalibration reduced the preference from 4.2 to about $3.2\sigma$ with DESI DR2 and CMB.
Posed since: 2025