Thread by @DScol

Lately, there have been different reports about the Hubble Tension from JWST, with different teams and methods. It’s been hard to follow. But initial sample sizes were small so in a new paper we combined, for the first time, all measurements to date, includes 5 JWST programs, and it makes a pretty clear picture. 🧵arxiv.org/abs/2408.11770

The first main question is “Do other techniques/instruments/teams JWST galaxy distances agree with past HST measurements?” It’s a resounding yes! 👏 All the offsets are on the 0.03 mag level and 1sigma from HST SH0ES, much smaller than Hubble tension scale of 0.18 mag. Conclusion, Hubble Tension is not a distance measuring problem.

The second question: “Could HST Cepheid measurements be non-linear, i.e., change with distance?”. This one is my favorite plot of paper - and is a resounding no. This is first time all the distance measurements have been compared to check linearity (or offset) and find linearity of 0.994+-0.01, non-linearity that would explain Hubble tension ruled out at 7 sigma level. Thats bigger than the Tension itself!

The third question is “Are different values of H0 from early JWST subsamples consistent? ” We made this graphic to show that you have to select some SN hosts and compare some of their stars (TRGB/Ceph/JAGB) to the SN to measure H0. However, the SNe have a lot of scatter and JWST samples are really small, so what H0 you get depends on the luck of which ones you pick, which is why you need a lot to reduce that uncertainty.

Since HST observed all the SN, we can ask this question as “what H0 would you expect if you only looked at this or that small subsample of SNe with JWST?”. It turns out, one would predict just what different groups are measuring with JWST - we can show the fluctuations are almost entirely what HST expected. This is just small sample differences.

If one combines all the JWST measurements from all the teams, and propagates the covariance and uncertainties correctly, you get the "reversion to the mean" of larger samples: H0=72.6+-2.0 km/s/Mpc from JWST. HST would have predicted H0=72.8 for the same sample, so again JWST and HST agree.

This is pretty amazing. The HST and JWST data are telling a consistent story, just that early JWST samples are too small to measure H0 well. I think that story got a little muddled by latest CCHP paper because some uncertainties were getting removed when they shouldn’t have been, but the important thing to me is that the actual measurements agree between telescopes & teams & methods.

Having lived through this Hubble tension stuff for some time, I’d say this is the most solid ground these measurements have ever been on. Please ask any questions!