Wednesday, November 8, 2023

More On Wide Binaries, MOND and Deur

A new study concludes very emphatically that wide binary stars behave in a manner more like Newtonian dynamics and less like MOND. This seems pretty definitive, but we've seen contradictory conclusions on the wide binary data before, so I don't consider this to be the final word.

More importantly, it doesn't rule out all gravity based explanations of dark matter phenomena. In particular, Deur's gravitational self-interaction driven approach explains dark matter phenomena with a gravitational solution (with no dark matter or dark energy) over a very wide range of applicability and does not predict significantly non-Newtonian behavior in wide binaries

Deur's approach may or may not be consistent with consideration of non-perturbative standard general relativistic effects as claimed. But whether Deur is applying general relativity in an unconventional way that rigorously considers a feature of general relativity that most other researchers neglect, or is actually a subtle gravity modification theory that isn't quite identical to standard general relativity, it works. Deur's approach also roots MOND-like galaxy behavior, and really all dark matter and dark energy theories in a theory that has a fairly simple deep theoretical motivation rather than just being a phenomenological fit to some key data points. And, unlike LambdaCDM and other standard dark energy theories, Deur's approach does so without globally violating conservation of mass-energy.

Deur's approach explains dark matter phenomena not just where MOND works, but also in many circumstances where MOND does not work. 

For example, Deur's approach works in galaxy clusters, in the Bullet cluster, with respect to wide binary star systems, as an explanation for different degrees of dark matter phenomena in differently shaped elliptical galaxies, as an explanation for the two-dimensional arrangement of satellite galaxies around spiral galaxies, and as an explanation for dark energy phenomena as well. It even provides a potential explanation for the Hubble constant tension and can reproduce the cosmic microwave background observed by the Planck collaboration and the early galaxy formation observed by the Webb telescope.

If the conclusion of this paper holds up, it somewhat decisively tips the balance between Deur's approach and other gravitational explanations of dark matter phenomena.
We test Milgromian dynamics (MOND) using wide binary stars (WBs) with separations of 2−30 kAU. Locally, the WB orbital velocity in MOND should exceed the Newtonian prediction by ≈20% at asymptotically large separations given the Galactic external field effect (EFE). 
We investigate this with a detailed statistical analysis of Gaia DR3 data on 8611 WBs within 250 pc of the Sun. Orbits are integrated in a rigorously calculated gravitational field that directly includes the EFE. We also allow line of sight contamination and undetected close binary companions to the stars in each WB. We interpolate between the Newtonian and Milgromian predictions using the parameter αgrav, with 0 indicating Newtonian gravity and 1 indicating MOND. 
Directly comparing the best Newtonian and Milgromian models reveals that Newtonian dynamics is preferred at 19σ confidence. Using a complementary Markov Chain Monte Carlo analysis, we find that αgrav=−0.021+0.065−0.045, which is fully consistent with Newtonian gravity but excludes MOND at 16σ confidence. This is in line with the similar result of Pittordis and Sutherland using a somewhat different sample selection and less thoroughly explored population model. 
We show that although our best-fitting model does not fully reproduce the observations, an overwhelmingly strong preference for Newtonian gravity remains in a considerable range of variations to our analysis. 
Adapting the MOND interpolating function to explain this result would cause tension with rotation curve constraints. We discuss the broader implications of our results in light of other works, concluding that MOND must be substantially modified on small scales to account for local WBs.
Indranil Banik, "Strong constraints on the gravitational law from Gaia DR3 wide binaries" arXiv:2311.03436 (November 6, 2023).

In other dark matter news, fuzzy dark matter theories (a very light boson with wave-like behavior dark matter theory) are compared to cold dark matter theories (with which all sorts of empirical evidence is inconsistent).

As I've noted often, but articulated less often, models with very light dark matter particles (especially bosonic ones) look a lot like quantum gravity theories with a graviton that has zero mass but non-zero mass-energy and self-interaction. Many axion-like dark matter theories, like fuzzy dark matter theories, lean towards this description. So, we are gradually starting to see dark matter theories converge on a mechanism that bears great similarities to a gravity based explanation of dark matter phenomena.

Finally, this paper is interesting:
The immense diversity of the galaxy population in the universe is believed to stem from their disparate merging histories, stochastic star formations, and multi-scale influences of filamentary environments. Any single initial condition of the early universe was never expected to explain alone how the galaxies formed and evolved to end up possessing such various traits as they have at the present epoch. However, several observational studies have revealed that the key physical properties of the observed galaxies in the local universe appeared to be regulated by one single factor, the identity of which has been shrouded in mystery up to date. 
Here, we report on our success of identifying the single regulating factor as the degree of misalignments between the initial tidal field and protogalaxy inertia momentum tensors. The spin parameters, formation epochs, stellar-to-total mass ratios, stellar ages, sizes, colors, metallicities and specific heat energies of the galaxies from the IllustrisTNG suite of hydrodynamic simulations are all found to be almost linearly and strongly dependent on this initial condition, when the differences in galaxy total mass, environmental density and shear are controlled to vanish. The cosmological predispositions, if properly identified, turns out to be much more impactful on galaxy evolution than conventionally thought.
Jun-Sung Moon, Jounghun Lee, "Why Galaxies are Indeed Simpler than Expected" arXiv:2311.03632 (November 7, 2023).


George Herold said...

Nice, Thanks. Do you have a paper or name (that I can search for) to find quantum gravity theories; with zero rest mass, but non-zero mass-energy and self-interaction?

neo said...

arXiv:2310.12114 (astro-ph)
[Submitted on 18 Oct 2023]
Dissipationless collapse and the dynamical mass-ellipticity relation of elliptical galaxies in Newtonian gravity and MOND
Pierfrancesco Di Cintio
Download PDF

Context. Deur (2014) and Winters et al. (2023) proposed an empirical relation between the dark to total mass ratio and ellipticity in elliptical galaxies from their observed total dynamical mass-to-light ratio data M/L = (14.1 +/- 5.4){\epsilon}. In other words, the larger is the content of dark matter in the galaxy, the more the stellar component would be flattened. Such observational claim, if true, appears to be in stark contrast with the common intuition of the formation of galaxies inside dark halos with reasonably spherical symmetry. Aims. Comparing the processes of dissipationless galaxy formation in different theories of gravity, and emergence of the galaxy scaling relations therein is an important frame where, in principle one could discriminate them. Methods. By means of collisionless N-body simulations in modified Newtonian dynamics (MOND) and Newtonian gravity with and without active dark matter halos, with both spherical and clumpy initial structure, I study the trends of intrinsic and projected ellipticities, Sérsic index and anisotropy with the total dynamical to stellar mass ratio. Results. It is shown that, the end products of both cold spherical collapses and mergers of smaller clumps depart more and more from the spherical symmetry for increasing values of the total dynamical mass to stellar mass, at least in a range of halo masses. The equivalent Newtonian systems of the end products of MOND collapses show a similar behaviour. The M/L relation obtained from the numerical experiments in both gravities is however rather different from that reported by Deur and coauthors.

Comments: 10 pages, 8 figures, submitted, comments welcome
Subjects: Astrophysics of Galaxies (astro-ph.GA)
Cite as: arXiv:2310.12114 [astro-ph.GA]

George Herold said...

Thanks Andrew, I'm an ignorant solid state experimentalist. But MoND seems to be a hint to some type of quantum gravity(QG). When I look into QG theories I only see stuff about the high energy limits. Do you know of anyone thinking about the low energy limits? (Gravitons in the lowest energy state of the universe, interacting with each other and giving us something that looks like MoND.)

Ryan said...


Mitchell said...

Why would MONDish binaries contradict Deur?

andrew said...


Heuristically, because Deur is describing the self-interaction of big gravitational fields, and the gravitational fields are two stars in isolation aren't big enough for the self-interactions of those fields to be signifiant.

Deur reproduces MOND in galaxies because the gravitational fields of gillions of stars in galaxies are vastly more significant in such large systems and acquire MOND-scale strength.

Another heuristic observation is that the way that the somewhat non-intuitive way that self-interaction effects scale has similarity to how event horizons of black holes scale with mass, which isn't as one might expect, simply a function a mass within a certain volume of event horizon.

You can look at the math in the link in this post to my previous blog post doing that math.

MOND uses a simpler phenomenological formula to fit the data which fits pretty much all scales of galaxies. But MOND is too strong an effect in low mass handful of stars or less scale systems, and it is too weak an effect in very high mass galaxy cluster systems that also have a different geometry.

andrew said...

See also

Guy said...

I read the Banik23 paper and it builds a towering tower of models and assumptions. To the extent that a non-specialist can tell they all make superficial sense, and the paper was preregistered, so there wasn't a lot of model selection going on. He throws some accurate darts at some other Gaia DR3 papers. I think the folks finding positive MOND data are going to need to put on their big-statistician pants and pull out their slide rules.

Mitchell said...

Stacy McGaugh says he's still in the middle about wide binaries, because Chae claims that when he repeats the analysis of Banik et al, the evidence for MOND increases...?

Maarten Havinga said...

Hi Andrew,
Nice to see you blogging about this interesting turning point in scientific history!

I'd like to start a discussion on the effects of self-interaction for wide binaries. Suppose a graviton is coming from a point-like mass with an inclination of 30 degrees to the milky way disc plane. The most nearby other massive object is some 300000 AU (5 ly) further away (the avg distance between stars). The one after that 600000, and so on. If we want this graviton to be contained in the 2D structure of gravity around the disc, we can calculate the sum of deflection angles. For light these are 4MG/rc2 which is really tiny. The sum after several objects tends to a logarithm, and in the milky way with radius 50000 ly we can expect there to be around 10000 stars encountered in a straight line from the middle of the milky way. The sum of 1+1/2+...+1/10000 is around 10. In order for Deur's approach to work, which I certainly hope, the first deflection angle should therefore be on the order of 3 degrees, rather more since also larger angles should be captured. Let's put it on 7 degrees.
Doesn't this mean that for a wide binary some 15000 AU from each other that the deflection angle of one star on the other should be a factor 300000/15000 = 20 larger? Which is 140 degrees?

Maarten Havinga said...

Even if we must lower it to ~20 degrees, should that not mean that the "flux tube" effect between the wide binary enlarges the effective gravity? I'm not the expert, nor did I read Deur's papers closely, so correct me if I'm wrong.

andrew said...

I appreciate the thoughtful comments from all of you, and have gotten a little break from saving widows and orphans, which had been taking up most of my time, but now I'm bogged down doing all of the administrative garbage involved in running a law firm that I ignored while doing the worthwhile stuff. Life is hell. See you soon.

Maarten Havinga said...

Hi Andrew,
If your statement on saving widows and orphans is literally true, I'm very happy you prioritize them above physics! And I would gladly delete my comments if that would help them. In fact, for my comments above I think I'm already satisfied with the answer you gave me on physics forums (nickname structure seeker) that the flux tube effect is too small for wide binaries to be noticable straight from Deur's formula's.
Which is logical given that the acceleration from their mutual (unmodified) Newtonian gravity should be close to a_0.

My only question is more centered on the effect of planetary discs of the WB. Is it possible these discs also draw some extra gravity of the stars towards their plane? At the cost of gravity perpendicular to it? I ask because looking closely at figure 10 in Banik et al, for 2-3 kAU there seems to be a bump around v~1 and the maximum of WBs is found around v~0.4 (20% gravity drawn towards the planetary disc?) instead of the expected 0.5. At higher distances the effect is less noticable (the early maximum stays around slightly longer). Could this be due to WBs with aligned planetary discs? Thay would rhyme more with the planet nine MOND explanation - it seems like MOND is at work also on the outer solar system.

andrew said...

"If your statement on saving widows and orphans is literally true, I'm very happy you prioritize them above physics!"

It is. While somewhat tongue in cheek, this is basically what a lot of my recent cases in my day job as a lawyer involve.

"Is it possible these discs also draw some extra gravity of the stars towards their plane? At the cost of gravity perpendicular to it?"

Yes, but the magnitude is negligible, since the star makes up such a huge share of the total mass of the system, making the star-planet system virtually spherically symmetric, and because the aggregate mass of the system is so small.

Maarten Havinga said...

Hi Andrew,
Let's take a wide binary with average star masses (both 25% of the sun's mass) and 2 kAU separation. The gravitational acceleration between this binary is then ~1.6 a_0. Suppose one or both of these stars has a (not very rare) hot jupiter (jupiter has around 1/262 the mass of the star) at the avg distance between a hot jupiter and the star, namely 1/16 AU. Are you in this situation also sure that the system is close enough to spherically symmetric?
I still think such situations can lead to the hump in figure 10 of Banik et al around 2-3 kAU and v~1 for stellar discs that are aligned by accident and the early maximum for non-aligned stellar discs. For the 3-5 kAU graph it might explain the broadened peak as first peak non-aligned with much disc mass near the star and second 'peak' simply almost spherically symmetric. At least there's more to that figure than simply GR, IMO.

andrew said...

Even if you didn't have spherical symmetry, the aggregate mass of the system would still be too small.

And even Jupiter is ignorable relative to the mass of a star.

Maarten Havinga said...

Hi Andrew,
Thanks for your answer; but I'm kinda left unsatisfied and doubting which papers to follow.

Judging from your answer, no single neutrally selected set of WB data should have more gravity than Newtonian. However, I can't accuse Chae of cherry picking (in a new article he rebuts the accusation of Banik that his results came from unreliable data). So I can't accept that wide binaries follow GR to high precision. Yet I'm a fan of the idea of gravity SI to explain MOND's grand predictive success.

What about protoplanetary discs? They have a much smaller mass ratio than spiral arms in a galaxy, but just wondering since they are 2D.

Maarten Havinga said...

I think after everything that the EFE in gravity SI should be there, with gravity boosted in any direction. Perhaps it appears when incorporating the spin of the gravitons?

andrew said...

"I'm kinda left unsatisfied and doubting which papers to follow."

Me too.

"What about protoplanetary discs? They have a much smaller mass ratio than spiral arms in a galaxy, but just wondering since they are 2D."

I don't know. I haven't run the numbers.

Maarten Havinga said...

OK thanks!