Sabine Hossenfelder’s new video on the science paper in question.
Here is the relevant excerpt.
*"The Nobel Prize winning observation heavily relied on regularities of [Type 1a] supernovae. They assumed that these stellar explosions always happen in a very similar way no matter what galaxies the supernova’s in. Astrophysicists like to refer to them as standard candles. The new paper which just appeared says that, well, that just isn’t so.
They looked at the newest collections of data of supernovae from two different experiments. Together, these are more than 3,000. This is much more data than Perlmutter, Schmidt and Reese had in the 1990s when they did their analysis. Back then, they had fewer than 100.
The authors of the new paper took the data apart into groups by two different criteria. The first is how far the light traveled from the galaxy to us. This is the same as the Nobel Prize winners did. The further the supernova, the dimmer it appears. And since the universe expands, the wavelength of the light gets stretched as it travels. This is called the red shift. It’s the relation between the dimming and the red shift that tells us how the universe expands.
The new thing is that they also look at how old the galaxy was when the supernova happened. This is a different type of information. Basically, the overall light emission from the entire galaxy has a different spectral shape depending on how old the galaxy is. The kicker is that they find these different properties are correlated. The younger the galaxy, the dimmer the supernovae, independent of the red shift. The problem is now that naturally the further back in time you look, the younger the average galaxy. But since the age is correlated with the supernova properties, this looks exactly like the key evidence for the accelerated expansion of the universe.
Basically, they say that it’s wrong to conclude that the expansion of the universe is accelerating. It’s just that younger galaxies have on average dimmer supernovae. They don’t say why this might be the case. They simply say this is what the data really say. And this isn’t one of those maybe results with low statistical significance. This is a whopping 5.5 sigma correlation which they call a serious systematic bias.
They also redo the calculation for the expansion of the universe and find that it currently isn’t accelerating. It’s actually slightly decelerating though it seems to have accelerated in the past. If it goes on to decelerate then this might mean the universe will actually recolapse. And yes, this finding is neatly compatible with other recent results from DESI and the dark energy survey. They also found that it looks like this somewhat alleviates the Hubble tension, though it doesn’t make it go away.
The paper has been peer-reviewed and it’s published in a decent journal, but this correlation in the red shift data isn’t entirely new and its relevance has been debated among astrophysicists for a couple of years now. Most believe that the effect is there but much smaller than what this group says."*
Thank you,
Walter.