Adrien - Thursday, June 11, 2026

💥 What if we're wrong about the supernovae used to measure the Universe?

Type Ia supernovae, those stellar explosions used as landmarks to measure the Universe, might not be as reliable as we think—and we will soon know for sure.

A team of researchers has developed a new method that, thanks to artificial intelligence and future data from the Vera C. Rubin Observatory, promises to refine our understanding of these "standard candles." Their approach, named CIGaRS, could upend our view of cosmic expansion and the enigmatic dark energy.


Illustration of a white dwarf near a main-sequence star.

This method relies on images and mathematical calculations. It allows the determination of the age and composition of exploding stars—valuable information for estimating their distance.

It all began with the discovery of dark energy in 1998, thanks to type Ia supernovae. These explosions, caused by white dwarfs cannibalizing a neighboring star, appeared so uniform that their brightness served as a standard yardstick for measuring cosmic distances. But astronomers realized that their glow varies slightly depending on their galactic environment, complicating matters. The new method integrates all these factors—dust, galaxy ages, explosion frequency—into a single coherent model.


Dark energy remains one of the greatest mysteries of modern cosmology. It is thought to make up 68% of the Universe's content and began to dominate about 4 billion years ago, accelerating its expansion. It's like a swing that slows down then suddenly speeds up without being pushed: that's what dark energy does. To understand this phenomenon, extremely precise distance measurements are needed, and that's where the new CIGaRS method comes in.

This approach allows distances to galaxies to be estimated with high accuracy without resorting to long and costly spectroscopic observations. It will be essential for analyzing the data flood from the Legacy Survey of Space and Time (LSST) at the Rubin Observatory. The researchers claim that their model avoids selection and modeling biases.

The results of this study, published in the journal Nature Astronomy, pave the way for a more precise cosmology. By fine-tuning the calibration of these standard candles, scientists hope to better understand dark energy and the fate of the Universe. And all of this thanks to a simple idea: simulating the Universe in a computer to reveal its mechanisms.
Ce site fait l'objet d'une déclaration à la CNIL
sous le numéro de dossier 1037632
Informations légales