Chae Binaries Paper Published + Rebuttal of Banik + OSMU Talk

[strangerep]

For those who might not yet be aware...

Kyu-Hyun Chae's paper on wide binaries is now published: Chae: Wide Binaries Published Paper

There's also a Press Release from Sejong Univ.

The arxiv version (v4, Nov 2023) is here.

A formal Rebuttal of Banik's results, jointly authored by Hernandez and Chae, is also available.

Finally, here is a lengthy (over 2 hrs) OSMU Technical Seminar, mainly featuring Chae and Hernandez. Although quite tedious, it does cover a lot of explanation about how this kind of investigation is performed, and the nontrivial care with statistical methods that is required. An (imperfect, auto-generated) transcript is also available.

Chae notes near the end about how these results finally show that DM as an explanation for anomalous gravitational behaviour within galaxies can now safely be discarded. (Most of the DM in galaxies is supposedly in the surrounding halo, hence cannot be invoked to explain gravitational anomalies in wide binaries located at many places within the galaxy.)

[And now I wait to see whether Dr Becky will compose a new, scientifically honest, youtube video about Chae's work, and the rebuttal of Banik by Chae & Hernandez. Or will she just ignore them since that content doesn't conform to her preferred narrative? ]

 

[AndreasC]

I don't think this whole story will change the minds of many people. At this point everything has been blurred so much that everyone will probably just stick with what they already believed. Could this be improved with the next release of wide binary data from GAIA?

 

[martinbn]

I don't know anything about this, so I probably shouldn't comment, but one thing is unclear to me. The press release says that some binary stars show deviation from the Newtonian prediction. But what about the general relativistic prediction? Is it almost the same as the Newtonian in the case of wide binaries?

 

[AndreasC]

I don't know anything about this, so I probably shouldn't comment, but one thing is unclear to me. The press release says that some binary stars show deviation from the Newtonian prediction. But what about the general relativistic prediction? Is it almost the same as the Newtonian in the case of wide binaries?

Yeah pretty much.

 

[pinball1970]

For those who might not yet be aware...

Kyu-Hyun Chae's paper on wide binaries is now published: Chae: Wide Binaries Published Paper

There's also a Press Release from Sejong Univ.

The arxiv version (v4, Nov 2023) is here.

A formal Rebuttal of Banik's results, jointly authored by Hernandez and Chae, is also available.

Finally, here is a lengthy (over 2 hrs) OSMU Technical Seminar, mainly featuring Chae and Hernandez. Although quite tedious, it does cover a lot of explanation about how this kind of investigation is performed, and the nontrivial care with statistical methods that is required. An (imperfect, auto-generated) transcript is also available.

Chae notes near the end about how these results finally show that DM as an explanation for anomalous gravitational behaviour within galaxies can now safely be discarded. (Most of the DM in galaxies is supposedly in the surrounding halo, hence cannot be invoked to explain gravitational anomalies in wide binaries located at many places within the galaxy.)

[And now I wait to see whether Dr Becky will compose a new, scientifically honest, youtube video about Chae's work, and the rebuttal of Banik by Chae & Hernandez. Or will she just ignore them since that content doesn't conform to her preferred narrative? ]

I will keep an eye on Dr Becky!

There is a 2 hour discussion on this between Simon White (Max plank inst) and Stacy Mcgaugh (Case Western) (on YT so I will not post the link)

They discuss their relative positions and agree on the problems and ways to approach them for the fist 40 minutes or so then below.

Discussion is time stamped:

Evidence for DM

Case for MOND

Lamppost effect

DM sub halos

Missing Satellites

Bullet cluster

CMB

Wide Binaries and Gaia (only 5 minutes here unfortunately)

Other problems for LCDM

A lot to take in for me but some of you guys could fast forward to your place of interest and see what you think.

 

[ohwilleke]

Background for those who weren't following this debate much, because in the OP, strangerep just leapt in to the guts of the latest news.

Wide binaries are a pair of stars in a gravitationally bound system that are very distant from each other. Systems of more than two stars that are gravitationally bound in a system need to be excluded from the analysis, since if you miss a third-star in the system and look at a couple of them in a large numbered bound star system, your expectations of dynamics are greatly thrown off. One way to get results that seem non-Newtonian, but aren't, is to fail to exclude more than two star bound systems.

Naive, toy model MOND predicts that there should be a greater than Newtonian gravity pull between wide binaries (when they are far away enough from the galactic center), and GR predicts a nearly perfectly Newtonian pull between wide binaries, even if dark matter theories are correct (since the dark matter halo is huge enough to have no relatively different impact on the two wide binary stars which are much closer than a galaxy scale from each other).

There were hints years earlier that wide binary stars might not behave in Newtonian gravity, but the earlier data was too limited and low quality to be definitive about it.

Different investigators have looked at wide binary data from new astronomy observations (especially GAIA), looking at essentially the same new source data, and have analyzed it and have come to conclude different things.

Some see only Newtonian behavior. Some see clear non-Newtonian behavior. The confidence intervals for both sides are far too high for both of them to be right in both their conclusions and their claimed uncertainties. Both sides claim five sigma plus support for results that contradict each other. This is the cause of an academic debate among astronomers and astrophysicists over the meaning of this data set. The OP relays the latest volleys in that debate.

Like Stacy McGaugh, a leading MOND proponent and professor of astronomy and astrophysics in Maryland Ohio, I'm pretty agnostic about what the wide binary data really show at this point. The argument back and forth is very technical and depends upon a lot of data interpretation and statistical analysis.

Also GAIA is providing data in a liminal region between the clearly Newtonian and clearly deep MOND regime where the predicted effects of MOND are not very clear and depend upon which variant of MOND is used and the interpolation function used.

There is lots of room for subtle error or misanalysis.

To the extent that wide binaries are purely Newtonian, there are multiple modified/non-standard gravity approaches that can produce that result while replicating the core findings of MOND and vindicating a gravitational approach. So, a Newtonian result would not, as some people will surely claim (much as they wrongly did with the Bullet Cluster), disprove all forms of gravity based explanations of phenomena attributed to dark matter.

I agree, however, that it is very hard to find any dark matter particle explanation for non-Newtonian wide binary behavior. But, it might be possible but difficult if the wide binaries in the sample are out of equilibrium, and one could also always argue that experimental error or analysis error is to blame at this point.

 

[ohwilleke]

Case for Monday

I assume that this should be the case for MOND (modified Newtonian dynamics). The existence of Monday is, sadly, not in dispute, and everyone pretty much agrees that it was a bad idea.

 

[Vanadium 50]

professor of astronomy and astrophysics in Maryland,

Case Western. He left Maryland years ago.

My objections to using wide binaries as a test have been posted, and I will repeat them. I didn't think this was a good test when the data pointed in one direction and I don't think it is a good test when the data points in another.

Finally, problems with MOND are not evidence for ΛCDM, and problems with ΛCDM are not evidence for MONDl

 

[ohwilleke]

Case Western. He left Maryland years ago.

My bad.

Go Case Western Reserve University! (One of my good friends got her PhD in chemistry there, I have a copy of their handbook of chemistry and physics in a prominent place on a shelf in my living room and refer to it now and then, and it had a 3+2 engineering program with my undergraduate college).

Finally, problems with MOND are not evidence for ΛCDM, and problems with ΛCDM are not evidence for MONDl

Agreed. There is more than one dark matter particle theory, and there is more than one gravity based theory.

This said, ΛCDM is special in the sense of being the paradigm, and if dislodged as the paradigm, the reduces opposition from scientists reluctant to deviate from the paradigm to all other possibilities (IMHO this should have happened long ago).

Also, some kinds of proof, like evidence that wide binaries do not act in a Newtonian fashion, are generically a problem with a very large class of dark matter particle theories, and not just ΛCDM. It rules out pretty much all of the leading dark matter particle models with a dark matter halo mass distribution inferred from lensing and galaxy dynamics. Pretty much all current strong contenders for DM particle models would be ruled out. It it would take real contortions to work around the observation in a DM particle paradigm.

 

[strangerep]

I don't know anything about this, so I probably shouldn't comment, but one thing is unclear to me. The press release says that some binary stars show deviation from the Newtonian prediction. But what about the general relativistic prediction? Is it almost the same as the Newtonian in the case of wide binaries?

In most of galactic dynamics, the fields are so weak, and the motion so slow, that Newtonian gravitation is usually considered sufficiently accurate. GR effects are far too weak.

 

[strangerep]

I don't think this whole story will change the minds of many people.

Yeah, and the data extraction/analysis is too difficult/tedious for most people.

At this point everything has been blurred so much that everyone will probably just stick with what they already believed. Could this be improved with the next release of wide binary data from GAIA?

I anticipate that it will improve the accuracy of sky-projected velocities of stars, among other things...

 

[strangerep]

My objections to using wide binaries as a test have been posted, and I will repeat them. I didn't think this was a good test when the data pointed in one direction and I don't think it is a good test when the data points in another.

There a 2 parts to this, iiuc. (1) your objections related to the EFE, and (2) objections related to the contradictory statistical data analysis+results of different researchers.

For (2), that's why close analysis of the methods, and sometimes-heated debate amongst the protagonists are necessary. (Have you studied the detailed rebuttal by Chae+Hernandez against Banik's methods?)

For (1), the intragalactic EFE relevant to wide binaries was closely investigated by Chae + Milgrom by numerically solving AQUAL equations for external fields aligned at various angles relative to the plane of a wide binary. This makes the WB test rather better than one might initially think (imho).

 

[strangerep]

There is a 2 hour discussion on this between Simon White (Max plank inst) and Stacy Mcgaugh (Case Western) (on YT so I will not post the link) [...]

For the benefit of others, this discussion is best heard and then supplemented by reading Stacy McGaugh's blog where he provides extra detail about the factual data aspects that they disputed in the discussion.

 

[mitchell porter]

On the specific topic of the wide binaries, McGaugh's most nuanced judgement is found in this blog post:

https://tritonstation.com/2023/11/2...-made-about-the-situation-with-wide-binaries/

In the public debate (edit: I mean the ongoing debate about wide binaries, not the recent video where McGaugh and White discuss dark matter vs modified gravity in general), we hear the most about Chae, who claims to not only have knockdown evidence of non-Newtonian behavior, but that it even favors a particular MOND theory (AQUAL); and about Banik et al, who claim that there's no significant deviation from Newtonian behavior.

McGaugh says:

I am at present inclined to trust more the super-clean sample of Hernandez [et al] – the high quality binaries where there is a chance that we’re actually measuring what we want to measure. There are only a few hundred such binaries, so the statistical confidence is modest (2.6σ).

So McGaugh's tentative opinion, is that wide binaries do provide evidence for MOND, but not as sharp as Chae says.

 

[martinbn]

In most of galactic dynamics, the fields are so weak, and the motion so slow, that Newtonian gravitation is usually considered sufficiently accurate. GR effects are far too weak.

To a lay person like me it is not obvious. After all even Mercury and the Sun need GR. I suppose wide binaries are wide enough to be Newtonian.

 

[Vanadium 50]

To a lay person like me it is not obvious. After all even Mercury and the Sun need GR. I suppose wide binaries are wide enough to be Newtonian.

Mercury's orbit is changed by less that 0.5%.

The GR component is 7.5% of that.

The nearest star is 300,000 times further away.

This is not a quick or simple measurement.

 

[martinbn]

Mercury's orbit is changed by less that 0.5%.

The GR component is 7.5% of that.

The nearest star is 300,000 times further away.

This is not a quick or simple measurement.

And those stars are several millions times as heavy as Mercury, so it isn't obvious to someone like me. But my question was if their orbiting is Newtonian in the sense that the GR prediction is close enough to the Newtonian. Because in the news letters GR wasn't mentioned and to a lay person it is not obvious if wide binaries are Newtonian. Now that was answered. (I am curious why, but that's way off topic)

 

[Vanadium 50]

And those stars are several millions times as heavy

But increasing the mass of your test mass is a factor of 2 (at most) and not a factor of a million.

We can't expect a total non-expert to have formed good intuition on graduate-level physics. You have to know something and start somewhere.

 

[martinbn]

We can't expect a total non-expert to have formed good intuition on graduate-level physics. You have to know something and start somewhere.

True, but the press release was not written for experts.