I Distinguishing between the effects of dark matter and MOND

[Ranku]

How do we distinguish between the effect of dark matter and MOND with respect to flat rotation curves in galaxies? How would the shape of the rotation curve differ between the two?

 

[phinds]

How do we distinguish between the effect of dark matter and MOND with respect to flat rotation curves in galaxies? How would the shape of the rotation curve differ between the two?

If I understand it correctly, they don't vary much within a single galaxy. Where MOND fail is with multiple galaxies. For example, in the Bullet Cluster.

The Bullet Cluster provides the best current evidence for the nature of dark matter[4][8] and provides "evidence against some of the more popular versions of Modified Newtonian dynamics (MOND)" as applied to large galactic clusters

 

[strangerep]

How do we distinguish between the effect of dark matter and MOND with respect to flat rotation curves in galaxies? How would the shape of the rotation curve differ between the two?

The key difference is that dark matter is like a Rule of Lesbos -- you can bend and shape it to fit anything (and therefore it explains nothing).

With MOND, in contrast, one builds a mass model of a galaxy (using observations that are independent of principles involving Newtonian gravity), and solves the Poisson equation using that mass model as source. Then one applies the MONDian (universal) radial acceleration relation to predict a rotation curve. Such MOND-predicted rotation curves do very well against the actual data.

Btw, statements like "the Bullet Cluster falsifies MOND" are very much overblown, although the current situation remains fluid.

For a recent, but lengthy, summary of MOND-vs-DM phenomenology, see Stacy McGaugh's recent blog post: Checking in on Troubles with Dark Matter

 

[Ranku]

The key difference is that dark matter is like a Rule of Lesbos -- you can bend and shape it to fit anything (and therefore it explains nothing).

With MOND, in contrast, one builds a mass model of a galaxy (using observations that are independent of principles involving Newtonian gravity), and solves the Poisson equation using that mass model as source. Then one applies the MONDian (universal) radial acceleration relation to predict a rotation curve. Such MOND-predicted rotation curves do very well against the actual data.

Btw, statements like "the Bullet Cluster falsifies MOND" are very much overblown, although the current situation remains fluid.

For a recent, but lengthy, summary of MOND-vs-DM phenomenology, see Stacy McGaugh's recent blog post: Checking in on Troubles with Dark Matter

Suppose, for argument's sake, matter itself were to contain more mass than is observationally accounted for, would its gravitational effect be similar to dark matter or MOND?

 

[phinds]

Suppose, for argument's sake, matter itself were to contain more mass than is observationally accounted for, would its gravitational effect be similar to dark matter or MOND?

This is a pointless speculation.

 

[Vanadium 50]

The Bullet Cluster is not the...er...smoking gun that people say it is. It has two problems - one that it is moving too fast for a naet ΛCDM explanation, and the other is that there are galaxies where things happened in the other direction, e.g. Abell 520. Yes, MOND has trouble explaining it - but so does ΛCDM.

A fair statement is that MOND works well - probably better than ΛCDM - on galactic scales, but nowhere else. In principle with DM, one would look at rotationally supported galaxies, infer a0 as a measurement (and not a fundamental parameter of the theory) and there would be some distribution. In practice, there is a delta function - all a0's are the same. MOND says this is a law of nature, and ΛCDM says it's unimportant.

But the real point is that you won't separate the two models (families of models, really) by looking at rotation curves. That data could falsify MOND, but doesn't. You need to go to other scales, and that effectively means larger, not smaller. And that's problematic because less is known so the predictions are squishier.

​

​

 

[Ranku]

Suppose, for argument's sake, matter itself were to contain more mass than is observationally accounted for, would its gravitational effect be similar to dark matter or MOND?

Ok, let me re-phrase: Suppose there is more matter than is observationally accounted for, with distribution typical of matter, will its gravitational effect be similar to dark matter or MOND?

 

[Motore]

Suppose there is more matter than is observationally accounted for

The more is exactly what dark matter is (if you mean observationally accounted for as electromagnetically interacting).

 

[Ibix]

Suppose there is more matter than is observationally accounted for, with distribution typical of matter, will its gravitational effect be similar to dark matter or MOND?

You mean, if we just take the visible matter density and increase it by a uniform scale factor? Then no. If you are not adjusting the law of gravity but just adding mass then this is just what dark matter does. So your extra mass needs a distribution similar to the dark matter distribution, not the visible matter distribution.

 

[phinds]

There are also galaxies that simply cannot be explained at all by MOND, but only by dark matter. Here's an example of one where the rotation curve cannot be explained by MOND:
99-percent-dark-matter/

 

[Vanadium 50]

Alas, the Dragonfly galaxies have their own problems:

• The conclusion that thee is no DM depends on distance, and the distance measures are poor.
• The conclusion that these galaxies formed with DM and had it somehow stripped is not well-supported. If they never had any, how did they form? And if they had it removed, how did this happen without disrupting the stars?

One thing in astronomy I am not very unhappy about is the tendency to take a distribution, look at events on the 3 or 4 sigma tail, and cry "Inexplicable!" and go running to Science News.

If you want to attack MOND, don't look at questionably measured objects where it does best - look at well-measured objects where it does worst.

 

[Ken G]

The key difference is that dark matter is like a Rule of Lesbos -- you can bend and shape it to fit anything (and therefore it explains nothing).

What is ironic about this characterization is that actually, a Lesbian rule is an extremely useful invention, and analogous applications in science are central to how it functions. (Consider Euclidean geometry, the classic example of a "straight rule" that turned out to not work in some applications.) The problem with a normal straight rule is that it is fixed in advance, before you encounter the application you need help with. We certainly find great use in this, but it only works when we can anticipate the problems we will face. For example, a fixed straight rule is great if you know you are going to build a rectangular structure, and you are fine with that constraint. Where a Lesbian rule comes in is when you need greater flexibility than rectangular structures, but you want something that is consistent from one stone to the next. So it's actually fine to use a Lesbian rule, so long as once you have crafted it, you are able to use it over and over. The problem comes in if you find you have to keep bending it anew, that is what would compromise its value. I would say we are still in the testing phase of dark matter, to see if the same Lesbian rule can be used over and over (for example, if the dark matter fraction in the universe is a fixed value across all ages), just not in a preordained way, or if we eventually give up on that approach.

With MOND, in contrast, one builds a mass model of a galaxy (using observations that are independent of principles involving Newtonian gravity), and solves the Poisson equation using that mass model as source. Then one applies the MONDian (universal) radial acceleration relation to predict a rotation curve. Such MOND-predicted rotation curves do very well against the actual data.

Except you are leaving out where those "universal" radial acceleration corrections come from! Of course, they come from the need to fit that data, so they are the quintessential example of a Lesbian rule also. Thus we see we are not debating between which one is a Lesbian rule, as they both are. We are trying to find out which one actually works like a Lesbian rule is supposed to work. Mainstream science has perhaps been too quick to side with one over the other, but at best we can agree the verdict is still out.

What is important to recognize is that DM and MOND are two different approaches to the problem, not that one or the other is inherently a better approach. With MOND, you are seeking a single law, but you will never be able to apply it to anything other than gravity. With DM, you can allow two different galaxies to have two different compositions, a flexibility that you are going to need if, in fact, the two galaxies actually do have two different compositions! (It's not like we don't already have galaxies with different compositions....) It's entire basis fundamentally rests on the well known fact that some matter interacts very little with light (neutrinos are already dark matter, for example), so we need to be ready if it turns out that lots of matter doesn't interact with light (and how else would we know?). So the flexibility of DM to handle situations that might actually in fact exist is certainly not a reason to discount its value. However, we do wish to do more with less if we can, so we would always go with MOND if we can interpret all behaviors in terms of a single correction to gravity. That is exactly what has so far not been demonstrated.

 

[strangerep]

So it's actually fine to use a Lesbian rule, so long as once you have crafted it, you are able to use it over and over. The problem comes in if you find you have to keep bending it anew, that is what would compromise its value.

Yes, I agree with your characterization.

I would say we are still in the testing phase of dark matter, to see if the same Lesbian rule can be used over and over (for example, if the dark matter fraction in the universe is a fixed value across all ages), just not in a preordained way, or if we eventually give up on that approach.

I was under the impression that the necessary form of DM halos are rather different for different galaxies, whereas a single MOND interpolation function needs to work widely (at least, within experimental error).

 

[Ranku]

There are also galaxies that simply cannot be explained at all by MOND, but only by dark matter. Here's an example of one where the rotation curve cannot be explained by MOND:
99-percent-dark-matter/

Does MOND do better in galaxies which otherwise requires less proportion of dark matter-to-matter?

 

[Ken G]

I was under the impression that the necessary form of DM halos are rather different for different galaxies, whereas a single MOND interpolation function needs to work widely (at least, within experimental error).

Yes, if dark matter is the cause, then different galaxy types have different amounts of dark matter acting in different ways. This would be analogous to how we need different metallicities in different galaxy types to understand how they behave. Not necessarily a bad thing, if it is indeed what is happening, but what's required is to have some other way of understanding why those differences exist. Like with metallicity, it conjures the fact that the action of stars raises metallicity over time, so we must correlate the different metallicities with different star formation histories. The added complexity becomes an opportunity to learn something else, not just an additional free parameter, but that's only useful if we can indeed learn that "something else."

 

[member 673059]

In wide binaries the question isn't between dark matter and MOND, since the effects of dark matter should be insignificant in wide binaries, but rather between Newtonian gravity/general relativity and MOND, because the modification to the acceleration parameter in MOND is still relevant. And so far it seems that MOND is more successful there.

 

[PeroK]

There's also the CMB evidence for dark matter:

https://www.physicsforums.com/threa...non-baryonic-dark-matter.1048465/post-6835462

 

[Vanadium 50]

that the necessary form of DM halos are rather different for different galaxies

Be careful. This is not a degree of freedom for DM. For whatever reason, DM arranges itself (probably via the formation of galaxies) to look like MOND, or if you prefer, Tully-Fisher. For whatever unknown reason, we see variations in the dark to luminous matter ratio, but always one that follows a particular relation.

If you ever want to start a fight at an astronomy conference say "MOND probably has nothing to do with gravity, but is an empirical fact that needs explanation." Then both sides will come after you.

 

[Ken G]

In fact I've seen it argued that even if dark matter is what is actually happening, the MOND paradigm is still a good quantitative approach to addressing it. So we might end up using MOND as a kind of convenient parameterization, even if we have good evidence of the existence of dark matter. (That could end up forging a kind of "peace" between the "sides"!)

 

[Vanadium 50]

peace

Peace, just like they have it in the Middle East!

I think the best you can hope for is Bierce's definition: "a period of cheating between two periods of fighting".

 

[Ranku]

we see variations in the dark to luminous matter ratio, but always one that follows a particular relation.

Could you elaborate?

 

[Vanadium 50]

Elaborate how?

 

[Ranku]

we see variations in the dark to luminous matter ratio, but always one that follows a particular relation.

What do you mean by this sentence, exactly? How do the variations in ratio also follow a particular relation?

 

[Vanadium 50]

I mean what I wrote. The amount and distribution of DM in rotationally supported galaxies can vary, but it always varies to match MOND.

 

[ohwilleke]

How do we distinguish between the effect of dark matter and MOND with respect to flat rotation curves in galaxies? How would the shape of the rotation curve differ between the two?

Comparing dark matter and MOND is comparing apples and oranges.

One could compare a specific version of dark matter to MOND, or you could compare dark matter to modified gravity theories generally.

We know, and have known for decades, that pure toy-model MOND or a pure relativistic generalization of it is not correct, due mostly to its failures to fully describe galaxy cluster scale phenomena which are outside its range of applicability, but also that it is "close" to a correct theory and is perfect or nearly so in its domain of applicability of galaxy scale and smaller systems.

But there are also multiple proposed gravity based explanations for dark matter phenomena that work in domains of applicability where MOND fails.

One easy way to screen modified gravity theories is to determine if they reproduce MOND where MOND works.

Also, you really need two pieces of a dark matter theory: a theory of what properties the dark matter particles have (spin, mass, cross-sections of interaction with different kinds of particles, mean velocity, number of types), and a theory of how dark matter came to be distributed throughout the universe in the manner that it does.

Astrophysicists who support dark matter particle theories tend to ignore the second half of that problem, which make the theory capable of describing almost any system but also deprive it of most of its predictive value. You can imagine a distribution that explains any particular system, after the fact, but can't predict what systems will form or what properties they will have.

Key Relationships And Facts

In the case of rotation curves, one of the arguments for a modified gravity theory, although not an overwhelmingly decisively one, is that in a modified gravity theory, it is possible to determine the dynamics of a galaxy entirely from the baryonic (i.e. ordinary) matter distributions in the system. Every feature of the ordinary matter distribution should be reflected in the rotation curves. And, every galaxy with an identical ordinary matter distribution should have exactly the same rotation curves.

There are specific dark matter theories that can be basically ruled out by examination of the dark matter distribution that is inferred from galaxy rotation curves/dynamics and gravitational lensing.

Most importantly, you can rule out a dark matter theory in which dark matter is truly collisionless (i.e. interacts only via gravity), with particle masses sufficiently large to make quantum interactions between the particles negligible (pretty much any particle 20 keV or more of mass) in which dark matter would take an NFW distribution (named after the people who first calculated it) with a strong central cusp, modified by baryonic gravitational feedback which is too small to significantly modify this distribution. This is not what is observed. See, e.g., Jorge Sanchez Almeida, Angel R. Plastino, Ignacio Trujillo, "Can cuspy dark matter dominated halos hold cored stellar mass distributions?" arXiv:2307.01256 (July 3, 2023) (Accepted for publication in ApJ).

So, any viable dark matter particle theory must have: (1) dark matter-dark matter interactions a.k.a. self-interacting dark matter, (2) wave-like quantum effects of very light dark matter particles with an effect similar to dark matter-dark matter interactions, or (3) dark matter-ordinary matter interactions.

In practice, the correlation between distributions of ordinary matter and inferred distributions of dark matter (which is natural and necessary in modified gravity theories) is so tight that some sort of dark matter-ordinary matter interactions beyond mere gravitational interactions is very strongly favored over options (1) and (2) above.

In support of this conclusion, see, e.g., Paolo Salucci, "The distribution of dark matter in galaxies" (November 21, 2018) (60 pages, 28 Figures ~220 refs. Invited review for The Astronomy and Astrophysics Review); Antonino Del Popolo et al., "Correlations between the Dark Matter and Baryonic Properties of CLASH Galaxy Clusters" (August 6, 2018), https://arxiv.org/abs/2008.04052; Man Ho Chan, "Two mysterious universal dark matter-baryon relations in galaxies and galaxy clusters" arXiv:2212.01018 (December 2, 2022) (Accepted in Physics of the Dark Universe); Xuejian Shen, Thejs Brinckmann, David Rapetti, Mark Vogelsberger, Adam Mantz, Jesús Zavala, Steven W. Allen, "X-ray morphology of cluster-mass haloes in self-interacting dark matter" arXiv:2202.00038 (January 31, 2022, last revised November 1, 2022) (accepted by MNRAS); Aidan Zentner, Siddharth Dandavate, Oren Slone, Mariangela Lisanti, “A Critical Assessment of Solutions to the Galaxy Diversity Problem” arXiv:2202.00012 (January 31, 2022); Lorenzo Posti, S. Michael Fall “Dynamical evidence for a morphology-dependent relation between the stellar and halo masses of galaxies” arXiv:2102.11282 (February 22, 2021) (Accepted for publication in A&A); Camila A. Correa, Joop Schaye, "The dependence of the galaxy stellar-to-halo mass relation on galaxy morphology" arXiv:2010.01186 (October 2, 2020) (accepted for publication in MNRAS); Paolo Salucci, Nicola Turini, Chiara Di Paolo, "Paradigms and Scenarios for the Dark Matter Phenomenon" arXiv:2008.04052 (August 10, 2020); Paolo Salucci and Nicola Turini, “Evidences for Collisional Dark Matter In Galaxies?” (July 4, 2017); Edo van Uitert, et al., “Halo ellipticity of GAMA galaxy groups from KiDS weak lensing” (October 13, 2016); and Zhixing Li, Hong Guo, Yi Mao, “Theoretical Models of the Atomic Hydrogen Content in Dark Matter Halos” arXiv:2207.10414 (July 21, 2022)(distributions of hydrogen in interstellar space are also inconsistent with a dark matter particle that interacts only via gravity)

On the other hand, dark dark matter searches tightly constrain the cross-section of interaction between dark matter and ordinary matter for dark matter particles with masses on the order of the proton mass or greater, limiting any interaction between dark matter and ordinary matter to an interaction many orders of magnitude weaker than the weak force interactions between neutrinos and ordinary matter. (Direct dark matter searches don't themselves rule out dark matter particle candidates well in excess of 1000 GeV or more, but other considerations disfavor heavy dark matter candidates.)

Similarly, dark matter made up of "hadrons" of ordinary quarks and gluons bound by the strong force, is pretty much completely ruled out. So, are "sterile neutrino" dark matter candidates that are more massive the the heaviest ordinary neutrinos but no heavier than "warm dark matter" (up to about 20 keV).

Another key fact, is that there is a strict relationship between amount of ordinary matter and inferred dark matter halo size that holds for all isolated galaxies. It does not hold for galaxy clusters, but there is a parallel scaling relationship that does hold for galaxy clusters that is equally tight. Dark matter particle theories, generically, predict a different mass scaling relationship in galaxy clusters than the one that is observed. See, e.g., Yong Tian, Han Cheng, Stacy S. McGaugh, Chung-Ming Ko, Yun-Hsin Hsu "Mass-Velocity Dispersion Relation in MaNGA Brightest Cluster Galaxies" arXiv:2108.08980 (August 20, 2021) (published in 24 The Astrophysical Journal Letters 917)

On balance, dark matter phenomena are more wave-like than particle-like which favors low mass dark matter particles or gravitational effects as an explanation. See, e.g., Alfred Amruth, "Einstein rings modulated by wavelike dark matter from anomalies in gravitationally lensed images" Nature Astronomy (April 20, 2023) https://doi.org/10.1038/s41550-023-01943-9 (Open access copy available at arxiv).

Hot v. Warm v. Cold Dark Matter And Its Implications

The estimated mean velocity of dark matter particles, which is "cold" to "warm", is something we can infer from the magnitude of large scale structure (galaxies, satellite galaxies, etc.).

Mean velocity of dark matter particles is correlated tightly with dark matter particle mass in "thermal freeze out" theories of dark matter creation in the early universe, and "thermal freeze out" dark matter candidates are all but ruled out.

This consideration, for example, strongly disfavors very heavy thermal freeze out dark matter candidates that can't be ruled out with direct dark matter detection experiments.

If you don't have a thermal freeze out dark matter candidate, however, you need a process to create and destroy dark matter particles in near perfect equilibrium that imparts the right mean velocity to these particles. No process that would do that is known. And, while we could miss a process like that with our current experiments to date for very low mass dark matter candidates, it would be much harder to miss a process like that for heavy dark matter candidates in excess of 1000 GeV.

Dark Matter Decay And Annihilation Signals

The observational constraints on theories with dark matter particles that decay to ordinary matter or photons are also exceedingly strict, so dark matter particle annihilation or decay has to be extremely rare. Mean dark matter particle lifetimes have to be on the order of magnitude of the age of the universe or longer.

Key Discriminants

One key way to distinguish between different dark matter and modified gravity theories is from observations of wide binary stars, i.e. pairs of stars that are very distant from each other but are still a gravitationally bound system, that are not unduly influenced by external gravitational fields (e.g. at the far fringes of galaxies or isolated in open space between galaxies).

In pure toy-model MOND, which generally works at these scale, wide binary stars should be more tightly gravitationally bound than Newtonian gravity would predict. In dark matter particle theories and some other modified gravity theories, this effect shouldn't exist. The data is inconclusive with different groups reaching opposite conclusions.

Another important discriminant between theories (dark matter and gravity-based alike) involves observations of the dynamics of stars well above or below the main disk of spiral galaxies. These observations are also inconclusive so far, and are hard to make, but preliminary results tend to show that MOND effects are predominantly in the radial direction of rotationally supported galaxies - a result that doesn't necessarily support all dark matter particle theories either. A discussion of those observations can be found here.

 

[Ken G]

That's quite a treatise on a complicated situation, thank you. I did have one specific question though, about this part:

But there are also multiple proposed gravity based explanations for dark matter phenomena that work in domains of applicability where MOND fails.

I'm not understanding how MOND can fail but another gravity based explanation could succeed in the domain you consider, because I thought "MOND" literally meant "modification to Newtonian dynamics," so anything that proposes a modification to Newton's universal law of gravity would count as MOND. It sounds like you have something more specific in mind when you refer to MOND, so what is the difference between MOND and a proposed gravity based explanation that replaces the need for copious amounts of dark matter?

 

[ohwilleke]

That's quite a treatise on a complicated situation, thank you. I did have one specific question though, about this part:

I'm not understanding how MOND can fail but another gravity based explanation could succeed in the domain you consider, because I thought "MOND" literally meant "modification to Newtonian dynamics," so anything that proposes a modification to Newton's universal law of gravity would count as MOND. It sounds like you have something more specific in mind when you refer to MOND, so what is the difference between MOND and a proposed gravity based explanation that replaces the need for copious amounts of dark matter?

MOND is a specific theory devised by Prof. Mordehai Milgrom in 1983. It proposes that gravity behaves in a way consistent with general relativity below a threshold gravitational acceleration magnitude a₀ and is stronger according to a simple formula that creates flat rotation curves for weaker gravitational accelerations, subject to an "external field effect". See generally, this Scholarpedia article by Professor Milgrom about MOND.

The term does not have the broader meaning that you mistakenly ascribed to it.

There are many other gravity based efforts to address phenomena attributed to dark matter (maybe a dozen of which have received serious attention).

 

[phinds]

The term does not have the broader meaning that you mistakenly ascribed to it.

None-the-less, the generally held belief is that it does. I'm sure I have read numerous times here on PF that "MOND" is a collection of theories.

 

[ohwilleke]

None-the-less, the generally held belief is that it does. I'm sure I have read numerous times here on PF that "MOND" is a collection of theories.

There are a collection of variations on MOND theories. But those various generalizations of Milgrom's 1983 theory are themselves distinct from non-MOND modified gravity theories.

All MOND theories (there are at least nine or ten variations of MOND) are elaborations of Milgrom's 1983 theory which has a few moving parts (like how it should be generalized relativistically, what sort of interpolation function between the Newtonian and non-Newtonian regimes should be used, and whether it should modify gravity or inertia).

Some of the more prominent non-MOND gravity based approaches are J. Moffat's modified gravity theories such as MOG which have scalar, vector, and tensor elements, conformal gravity theories, Verlinde's emergent entropic gravity theories, f(R) theories, and Deur's approach to non-perturbative gravitational field self-interactions that derive from or only subtly deviate from general relativity in weak gravitational fields. One of the newest which I haven't really considered to carefully yet, is described in Gianni Pascoli, "A comparative study of MOND and MOG theories versus the κ-model: An application to galaxy clusters" arXiv:2307.01555 (July 4, 2023). Another recent one is Kimet Jusufi, Genly Leon, Alfredo D. Millano, "Dark Universe Phenomenology from Yukawa Potential?" arXiv:2304.11492 (April 22, 2023).

 

[Vanadium 50]

Sadly, on the internet one can "win" an argument simply by showing more stamina and to drive out competing arguments in an avalanche of words.

"MOND" as generally used corresponds to a whole family of theories. Either side of F=ma can be modified. You might get more force at low accelerations, or you might get more acceleration for the same force (which might or might not be restricted to gravity). We could call these models MOND. MONG and MONI. But we don't.

Then there is the matter of tha tyransition region between MONDy and Newtonian behavior. There is no real evidence for Milgrom's form - even he says as much. A theory that predicts one (and I am unaware of any, but creating one is trivial) could in principle be distinguished from others by behvior at the transition, but it is not an area where we have a lot of data.

Then there's the treatment of how we treat external fields and the related question of superposition. And then there's the whole question of relativistic extensions. And, and, and...

 

[ohwilleke]

"MOND" as generally used corresponds to a whole family of theories. Either side of F=ma can be modified. You might get more force at low accelerations, or you might get more acceleration for the same force (which might or might not be restricted to gravity). We could call these models MOND. MONG and MONI. But we don't.

While there are a number of MOND theories, that acronym is not generally used to refer to the whole universe of gravity based explanations of dark matter phenomena or modifications of gravity. It is one specific approach centered around a transition at the physical constant a₀.

It is not used to describe the work of Moffat or Verlinde or Deur or any other number of theories that don't have the basic outlines of Milgrom's work built into them directly. I have never seen those theories included within the term MOND in the literature.

Admittedly, there is a lot of the literature that is oblivious to the fact that there are gravitational approaches to describing dark matter phenomena other than the one that Milgrom came up with first which is the most studied one. But, that is a very different thing than saying that specific non-Milgromian gravity modification theories are part of MOND.

For example, the paper Federico Lelli, et al., "Cold gas disks in main-sequence galaxies at cosmic noon: Low turbulence, flat rotation curves, and disk-halo degeneracy" arXiv:2302.00030 (January 31, 2023) (Accepted for publication in Astronomy and Astrophysics) states:

Milgromian dynamics (MOND) can successfully fit the rotation curves with the same acceleration scale a₀ measured at z≃0.

Similarly, MOND and Moffat's theories are distinguished from each other (even though both explain DM-phenomena with gravitational modifications) in Yongda Zhu, et al., "How Close Dark Matter Halos and MOND Are to Each Other: Three-Dimensional Tests Based on Gaia DR2" arXiv:2211.13153 (November 23, 2022) (accepted for publication in MNRAS).

MOND is distinguished from Verlinde's theories here.

"Modified gravity" is a general and non-theory specific term, but MOND is not.

 

[Ken G]

Still, now that I understand you are taking a strict meaning of MOND, I can better understand how you are summarizing the landscape. In part from what I've seen, and in part from your summary, it seems that we are as far away from a new theory of gravity to add to relativistic and quantum dynamics as we are from a new set of dark matter particles to add to the particle zoo. One can see why there is general favoritism to simply finding the "dark matter particle", since then we just have the attributes and interactions of that particle to understand, much like the discovery of the neutrino, and we can otherwise maintain the existing dynamical machinery. If it's MOND, then the particles stay the same, but the complete dynamical machinery must change! That's a lot of overhead. I have faith in the process of science, that this monumental task will eventually be undertaken successfully, but I haven't much faith it will happen any time soon, so I doubt I will see it. If it takes a century, none of us will!

On the other hand, one of the things that does seem to offer promise to be resolved in the foreseeable future is "precision cosmology." Some claim it's already here, but I am not among them, not as long as their remains tension in the determination of the Hubble parameter. But if I'm an optimist, and I think the next round of cosmological observations will line everything up nicely as things appear to be turning out, then we are going to be looking at dark matter and dark energy as the key elements of cosmological dynamics. As long as that continues to be true, then dark matter is pretty much here to stay, regardless of any successes of MOND. As I said above, it may turn out that MOND is only used as a convenience, a way to parametrize the effects of dark matter in galaxies because it is easier to use than some whole new paradigm for equipping galaxies with dark matter and its interactions. Or it might turn out that dark matter is only used as a convenience in cosmology, a way to parametrize some unknown dynamical effect on the expansion of the universe. An uneasy peace, perhaps, and maybe only a period of "cheating" as alluded to above, but that does seem to be where we are, and I'm not sure I see that situation changing any time soon. But who knows?

Then of course, there is also the very nonnegligible chance that we are completely missing something, and it is neither MOND nor dark matter. That will require thinking outside the box, maybe one of the alternatives that has been mentioned or something completely new. The one thing that has always been true of scientific thinking is it tends to overestimate who close it is to reaching the complete story!

 

[ohwilleke]

Still, now that I understand you are taking a strict meaning of MOND, I can better understand how you are summarizing the landscape. In part from what I've seen, and in part from your summary, it seems that we are as far away from a new theory of gravity to add to relativistic and quantum dynamics as we are from a new set of dark matter particles to add to the particle zoo. One can see why there is general favoritism to simply finding the "dark matter particle", since then we just have the attributes and interactions of that particle to understand, much like the discovery of the neutrino, and we can otherwise maintain the existing dynamical machinery. If it's MOND, then the particles stay the same, but the complete dynamical machinery must change! That's a lot of overhead. I have faith in the process of science, that this monumental task will eventually be undertaken successfully, but I haven't much faith it will happen any time soon, so I doubt I will see it. If it takes a century, none of us will!

On the other hand, one of the things that does seem to offer promise to be resolved in the foreseeable future is "precision cosmology." Some claim it's already here, but I am not among them, not as long as their remains tension in the determination of the Hubble parameter. But if I'm an optimist, and I think the next round of cosmological observations will line everything up nicely as things appear to be turning out, then we are going to be looking at dark matter and dark energy as the key elements of cosmological dynamics. As long as that continues to be true, then dark matter is pretty much here to stay, regardless of any successes of MOND. As I said above, it may turn out that MOND is only used as a convenience, a way to parametrize the effects of dark matter in galaxies because it is easier to use than some whole new paradigm for equipping galaxies with dark matter and its interactions. Or it might turn out that dark matter is only used as a convenience in cosmology, a way to parametrize some unknown dynamical effect on the expansion of the universe. An uneasy peace, perhaps, and maybe only a period of "cheating" as alluded to above, but that does seem to be where we are, and I'm not sure I see that situation changing any time soon. But who knows?

I would say that the balance has already tipped. Lambda-CDM which is the paradigm is dying a death of several dozens serious cuts, if not the thousand of the Chinese proverb.

Lots of the barriers to acceptance of modified gravity approaches have been overcome now that some modified gravity theories can explain galaxy clusters and the Bullet Cluster, and can reproduce the Cosmic Microwave Background. The failure to dark matter particle theories to predict the early formation of galaxies observed by the James Webb Space Telescope (JWST) has also helped tip the balance.

Another factor tipping the balance is the reality that you need not just a new particle but also a new force somewhat similar to a gravity modification to describe what is observed in a dark matter particle paradigm.

One of the particularly notable efforts to root MOND's phenomenological fits into a deeper theory with a full range of applicability and not just a phenomenological fit with a limited domain of applicability, has been the work of Alexandre Deur who has devised a gravitational theory to achieve that. Deur claims (and others dispute) that it is merely a correct implementation of GR that implements it in a way the considers non-perturbative effects usually ignored. But even if Deur's wrong and his equations really do modify GR, they modify GR in a way that addresses the CMB, galaxy formation timing, clusters, the bullet cluster, the dependence of inferred dark matter amounts on galaxy shape, etc. in a way that has fewer free parameters than GR with dark matter, a single field, no dark energy, conservation of mass-energy at a global level, and a capacity to make predictions that unduly free DM theories struggle to. Deur proposes that this comes from a second order gravitational field self-interaction effect that falls off more slowly than the first order Newtonian one does with distance in mass distributions which are more disk-like than spherically symmetric or are two point systems - using the geometry of the mass distribution to bridge the issues that MOND struggles with, in close analogy to QCD (the quantum theory of the strong force that binds quarks and gluons together). In strong gravitational fields where GR deviations from Newtonian gravity are clear, these second order effects are dwarfed by the first order effects and imperceptible. But in very weak fields at very great distances, the slower falloff of the second order effects relative to the first order Newtonian-like effects in weak field GR eventually become stronger than the first order effects and become noticeable.

There is really no DM particle theory out there which fits such a broad range of data.

Neither self-interacting dark matter, nor warm dark matter (which involves collisionless DM at the keV order of magnitude mass threshold where it starts to be wave-like), which were once quite popular, solve some of cold dark matter's most serious problems. See, e.g., Mark R. Lovell, et al., "Local Group star formation in warm and self-interacting dark matter cosmologies" arXiv:2002.11129 (Feb. 25, 2020) (accepted by MNRAS); Isabel M.E. Santos-Santos, et al., "Baryonic clues to the puzzling diversity of dwarf galaxy rotation curves" (November 20, 2019) (accepted for publication in MNRAS) (also noting the failure to SIDM to reproduce what is observed); and Alyson Brooks, "Re-Examining Astrophysical Constraints on the Dark Matter Model" (July 28, 2014) (published in Annalen der Physik) (noting that SIDM and warm dark matter has most of the same problems as cold dark matter theories, as also shown in this paper by the same author together with a co-author which was published in MINRAS).

Axion-like particle (ALP) dark matter and similar ultra-low mass fuzzy dark matter theories, mostly proposing bosonic dark matter particles with masses on the order of 10^-23 eV plus or minus a couple of orders of magnitude (which approach the same ballpark as the mass-energy of a graviton of typical frequencies) seem to do better so far, but are relatively new and less well vetted at this point. CERN recaps some of the basics of these theories.

 

[Ken G]

That's quite an interesting take, you are either way ahead of the curve on this, or buying off on speculative results that are not as well vetted as you claim. I don't know which, but either is still interesting, and more promising than the perspective I expressed above!

 

[ohwilleke]

That's quite an interesting take, you are either way ahead of the curve on this, or buying off on speculative results that are not as well vetted as you claim. I don't know which, but either is still interesting, and more promising than the perspective I expressed above!

We certainly don't have a final answer yet.

But optimism that we will get there is warranted, because of the torrent of new astronomy data that we are collecting, and our increasingly computational capacity to make sense of it.

Also, importantly (because there is no saving the satisfied scientist), in astronomy, it is widely understood that we don't have a model the neatly and cleanly describes all observations the way that the Standard Model of Particle Physics (deservedly) is recognized for doing in high energy physics. Everyone agrees that we need more data and that it is challenging to make the existing data (e.g. the value of Hubble's constant) mutually consistent.

There are hundreds of new astronomy papers each week, most of which report on new data, while there are only dozens of new high energy physics experimental data papers each week, dominated by four or five major colliders. The astronomy community is also more fractured than the high energy physics community, because new data is coming from more comparatively small research groups than in HEP, so it isn't quite as vulnerable to groupthink.

There is room for challenges to the Lambda CDM paradigm to develop, and this is happening on multiple fronts. Dark matter particle theories are still in the majority, but are also fractured themselves into many competing variant theories.

The down side of the current situation, however, is that because there are so many papers out there, everyone is looking only at their own little corner of their own research interests. Few scientists are looking at the big picture. So, quite a bit of research is done on proposals that existing work in some other subfield using different methodologies already strongly disfavors. It may take a generation or so for the constraints of existing research to really cross-pollinate across the larger discipline of astrophysics.

 

[Ranku]

I mean what I wrote. The amount and distribution of DM in rotationally supported galaxies can vary, but it always varies to match MOND.

Ok, got it now.

 

[Ken G]

I hope I see significant advances in this big picture. I guess the question is, when is the appropriate time for someone to expand on what you wrote above and deliver the "current state" of that picture? If one does it now, it will probably have to end with the need for more data, but if one waits too long, the reaction will be "yeah everyone knows that except the hangers-on."

 

[Vanadium 50]

I think you need more information, and good information - not just nibbling at the edge cases.

Suppose one of the supercomputing modeling types were to say that her DM modes reproduce MONDy behavior. Too much DM and you get pressure-supported galaxies not rotationally-supported. Too little and you never trigger star formation. Oh. and by the way, Tully-Fisher pops out too. That would be another,

Or suppose an LHC experiment - or LZ - detects DM. That would be another thing.

Guessing "when" sort of requires guessing "what".

 

[Ken G]

Yes that's another way to frame it, we could ask the question, what is the discovery that would settle the matter?

If someone detects dark matter decay of some kind, then the MOND world can just say, all that has happened is some new particle has been discovered, not the first time that's happened. It doesn't resolve anything unless the amount of it can be connected with the right amount required, and how does one do that when all one sees is decay without knowing the decay rate?

And if someone shows that a particle can be postulated that would do what dark matter needs to do, but there is no evidence that such a particle actually exists, then we have Dirac's neutrino suggestion without the Cowan-Reines experiment. The inability to manipulate astrophysical sources in the laboratory will likely continue to present significant challenges to resolving the issue.

After all, Aristarchus proposed the heliocentric model before 200 BC, and it wasn't until Galileo,18 centuries later, that scientists found a way to confirm it!

 

[ohwilleke]

Yes that's another way to frame it, we could ask the question, what is the discovery that would settle the matter?

No one discovery will "settle the matter" and even several significant discoveries taken together will only greatly winnow the field of possibilities.

But various kinds of discoveries would rule out lots of theories and narrow the parameter space dramatically for other kinds of theories (as would simply having wider diffusion of existing experimental results that aren't universally known in a systemic way).

Often different tests rule out different parts of the parameter space of particular theories.

Examples of Existing Often Piecemeal Constraints

For example, in the case of primordial black hole dark matter candidates several different constrains are pieced together to get our current parameter space, (1) Hawking radiation and the age of the universe is used to theoretically rule out very small primordial black holes, (2) the absence of gravitational lensing in seemingly empty space constrains how much dark matter can be in the form of large primordial black holes, and (3) the dynamics of rocky objects in the solar system is an important tool to constrain how much dark matter can be in the form of asteroid mass primordial black holes.

For collisionless dark matter candidates, heavy dark matter particles are ruled out by (1) the fact that galaxies don't universally have cuspy inferred dark matter halos and (2) wave-like dynamics in inferred dark matter.

The existence of observed levels of galaxy structure rules out collisionless thermal freeze out dark matter particles of less than about ca. 10 eV (which would be "hot dark matter").

Dark matter candidates that interact via the weak force with the same "weak force charge" as all other particles that interact via the weak force is ruled out (1) from about 1 GeV to 1000 GeV by direct dark matter detection experiments (which themselves combined several different kinds of parameter space exclusions from different kinds of direct dark matter particle detection experiments), and (2) for all masses below about 45 GeV (half of the Z boson mass) by W and Z boson decays. In a somewhat smaller mass range near the electroweak scale (tens and low hundreds of GeVs), dark matter particles with "milli" and "micro" weak force charges are also pretty much ruled out.

Dark matter particle candidates that interact with the Higgs boson with a Yukawa coupling proportional to rest mass (like the Standard Model interactions of Standard Model particles with the Higgs boson) are strongly disfavored in a mass range of about 2 GeV to 62 GeV because such a particle would throw off the predicted decay fractions of Standard Model particles that are reasonably close to the predicted values by easily detectible amounts.

Dark matter candidates in the meV to 10 eV mass range or so, are disfavored by astronomy based N_eff measurements (i.e. estimates of the effective number of neutrino species).

The Bullet Cluster did not, as some people claim, rule out all modified gravity theories, but it is a big problem for many simple, single field, spherically symmetric modified gravity theories like bare bones toy-model MOND theories where the effect is unrelated to the geometry of the matter distribution. It also poses different kinds of problems for dark matter particle theories because galaxy cluster collisions like the Bullet Cluster are far too common in the observable universe relative to LambdaCDM cosmology predictions of their relative speeds and frequencies.

Potential New Discoveries

As noted above, more precise data on wide binary star dynamics can provide a key generic constraint on lots of possible dark matter and modified gravity theories.

An observation of a "no dark matter" low surface brightness dwarf galaxy isolated from other galaxies, if made, would be an important constraint on many possible theories (both dark matter particle theories and modified gravity theories). Dark matter particle theories and MOND-like theories, however, both have plausible explanations for "no dark matter galaxies" near other more massive galaxies (tidal stripping for dark matter particle theories, and the external field effect in MOND).

Better modeling of the gravitational feedback of ordinary matter on dark matter distributions (which is just on the brink of being precise enough to be useful) is another method by which a lot of theories could be favored or ruled out, and by which parameter spaces could be greatly constrained for the remaining theories. This is one of the main fudge factors in existing supercomputer modeling of dark matter theories.

Larger data sets of the alignment of satellite galaxies in the plane of their primary galaxies for spiral galaxies would be useful in discriminating between theories that are spherical symmetric and those that can have asymmetrical effects.

Currently, the observed scatter in the Tully-Fischer relationship has a magnitude consistent with being entirely due to measurement error. As the measurements used to make those comparisons grows more precise with more data from better telescopes, the extent to which there is or is not intrinsic scatter as opposed to mere measurement error is important. Less scatter favors modified gravity theories. More scatter favors dark matter particle theories.

More data from the James Webb Space Telescope and instruments like EDGES (21 cm wavelength observations) of the very early universe is going to tightly constrain structure formation models which will require a major purge of dark matter particle theories.

If someone detects dark matter decay of some kind, then the MOND world can just say, all that has happened is some new particle has been discovered, not the first time that's happened.

While not strictly speaking impossible, this possible experimental breakthrough is way down my list of likely possibilities in the foreseeable future.

Dark matter particle candidates that have any kind of significant Standard Model particle interactions (which it would have to in order to have detectible decay products or to be found in a collider) are pretty much ruled out up to the mid-hundreds of GeVs to the low single digit TeV mass scale depending upon the candidate from collider factors, cosmic ray observations, and direct dark matter detection experiments. But, as noted above, a variety of factors disfavor dark matter particle candidates heavier than the ranges that are already ruled out by these means.

There are a variety of reasons (beyond the scope of this thread) to think that new undiscovered particles are unlikely to be found in the next generation of colliders or astronomy "telescopes" (using the term broadly).

The True Nightmare Scenario

One very real possibility, more likely than a dark matter particle candidate showing up at a particle collider, is that our observations will become over constrained.

In other word, we might end up with data that rule out every dark matter particle theory and every modified gravity theory and every hybrid of the two that we can imagine that work with no limitations on their domain of applicability.

If this happened, we'd have to figure out which constraint we thought was a real constraint actually has some kind of loophole.

 

[Ken G]

Maybe that last possibility is not really a nightmare after all. The piecemeal approach to winnowing the theories that you describe above is certainly good science, but is a kind of nightmare scenario of its own, because how many great discoveries in the history of science occurred that way? We could say that the quest to know the neutrino mass, or find the Higgs particle, were winnowing processes, but in both cases we had good reason to believe those particles would be the solutions and indeed they were, pinning down their masses and patting ourselves on the back for anticipating that success is not really the kind of breakthrough discovery we are talking about.

Instead, the most significant breakthroughs tend to happen when we realize that something we thought we could take as secure turned out to be the thing that was wrong. In other words, there was some kind of blockage that, once removed, created an opportunity for a much simpler solution to the problem! I'm sure you can think of examples of "outside the box breakthroughs" as easily as I can, but I would throw out some of the biggest in the history of astronomy (the heliocentric model, unblocked by realizing that stars should not show parallax because they are ridiculously far away, the Big Bang model, unblocked by realizing that the universe could have an origin that was outside of our physics, the age of the Sun, unblocked by realizing matter could contain a vast source of fusion energy, etc.). Maybe your nightmare scenario is just what we need right now!