Updated Pantheon+ and DES Supernova Data Leave Local Hubble Constant Near 73, Study Finds

A new arXiv preprint finds that recent updates to two major Type Ia supernova datasets barely move one of cosmology’s most contested numbers: the local measurement of the Hubble constant. In the analysis, the value stays near 73 kilometers per second per megaparsec, leaving a strong mismatch with the lower value inferred from the cosmic microwave background under the standard Lambda-CDM model of cosmology.

The paper, posted Sunday as arXiv:2607.24443v1, is titled “Updated 1.1% Precision Values of the Hubble Constant with Corrected Pantheon+ and Dark Energy Survey (DES)-DOVEKIE Type Ia Supernovae.” It was written by Shouvik Roy Choudhury of the Institute of Astronomy and Astrophysics at Academia Sinica in Taipei. The study updates a local, distance-ladder estimate of the Hubble constant using corrected Pantheon+ supernova data and, in this framework for the first time, DES-DOVEKIE supernova data.

Choudhury’s main updated result is H0 = 73.264 ± 0.806 km s−1 Mpc−1. The paper says that is a 6.85-sigma tension with a cosmic microwave background reference value of 67.24 ± 0.35 km s−1 Mpc−1 from combined Planck, ACT and SPT analyses within Lambda-CDM. By comparison, the paper cites the H0 Distance Network, or H0DN, community-consensus local benchmark at 73.499 ± 0.809 km s−1 Mpc−1, a 7.10-sigma difference from that same CMB reference.

The updated values around that benchmark are all similar. A second variant that replaces the full Hubble-flow treatment with corrected Pantheon+ data gives 73.139 ± 0.791 km s−1 Mpc−1, which the paper says is a 6.82-sigma tension. A DES-DOVEKIE-based run gives 73.267 ± 0.828 km s−1 Mpc−1, a 6.70-sigma tension. The core result is that the supernova-data revisions do not materially lower the local Hubble constant or erase the gap with the early-universe estimate.

What changed in the Pantheon+ update was relatively limited. According to the paper, a revised host-mass correction altered 114 of 1,701 apparent-magnitude rows in the public file. None of the 55 calibrator supernovae were affected, though 52 of 277 Hubble-flow rows were. That helps explain why the final H0 shift is small: the calibrators anchor the distance ladder used to convert supernova brightness into a measurement of today’s expansion rate.

For the DES-DOVEKIE analysis, Choudhury used 220 supernovae spanning about z = 0.025 to 0.151, where z, or redshift, is a measure of cosmic distance and expansion. Of those, 196 overlap with corrected Pantheon+ and were used to align the DES distances with the H0DN magnitude scale. The paper says extending the DES-DOVEKIE sample to roughly z ≈ 0.3 pushes H0 somewhat higher, but by less than the H0DN uncertainty.

The Hubble constant measures how fast the universe is expanding today. The “Hubble tension” refers to the long-running split between local measurements, which tend to cluster near 73, and early-universe estimates from the cosmic microwave background under standard Lambda-CDM, which come in closer to 67.

That makes this preprint a focused test of whether recent supernova recalibrations that have mattered in other cosmology debates also soften the Hubble tension. In this analysis, they do not. As Choudhury wrote in the paper, “Neither update resolves the Hubble tension.”