LQG Legend Writes Paper Claiming GR Explains Dark Matter Phenomena

[PeterDonis]

I'm curious to know how they explain galaxy rotation curves without adding mass?

Adding dark matter to the model is adding mass. Dark matter has mass. It's just not visible mass. In other words, the dark matter hypothesis is that the actual mass of galaxies is larger than the mass we infer from what we can see.

 

[PeterDonis]

cosmology is not my area of expertise, but mathematics with an inclination towards physics

So what? This is an "A" level thread in the cosmology forum. That means you are expected to have background knowledge in the subject matter equivalent to a graduate student. If you don't have that background knowledge, you need to go get it if you want to be able to make useful posts in this thread. We are not going to clutter up an "A" level thread with explanations of things that anyone with the requisite background knowledge should already know. That includes the basic tenets of the $\Lambda C D M$ model and the dark matter hypothesis and the reasons for it. Anyone posting in this thread needs to already know all those things.

 

[astronomer]

As an astronomer, I have a few general comments on the current (interesting) discussion. From the astronomical point of view, it is a little bit surprising that a few, well established empirical facts, appear to be ignored. First: the flat rotation curve of disk galaxies is just one (and certainly not the most relevant) evidence that (in newtonian gravity) we "need" DM. Non/weakly rotating systems (such as elliptical galaxies and clusters of galaxies) also require DM (in newtonian gravity) to explain their internal dynamics (with converging results obtained from hydrostatic equilibrium of hot gas, gravitational lensing, stellar dynamics). No convincing answer is provided by GEM framework on this point: what GEM effect mimics DM in non rotating systems? Data just requires that the claimed GR effects should be independent of rotation. Second: some low-mass stellar systems (with lower stellar velocities than in disk galaxies) requires (in newtonian gravity) larger DM-to-baryon ratios: why GR effects increase at decreasing mass? Third: some low-mass stellar systems (with mass in the same range of systems in previous point) do not require DM at all: why in some system GR effects are big and in other system of similar mass are not required?

Astronomically, it would be very useful (and interesting) to have a reasonable and convincing global picture of the proposed scenario, before focusing on very specific aspects.

 

[Structure seeker]

@astronomer, I think you are overlooking some literature. Although this thread is more about Deur's theory which also possibly explains galaxy clusters much better, the other phenomena you present are well explained in the most actual MOND models.

Elliptical galaxies: http://www.scholarpedia.org/article/The_MOND_paradigm_of_modified_dynamics#Elliptical_galaxies including citations.

Gravitational lensing:
https://arxiv.org/abs/2007.00082
From the paper: "It (the proposed theory) must (i) return to GR (hence, Newtonian gravity) when r~ Φ ≫ a0 in quasistatic situations while (ii) reproducing the MOND law (1) when r~ Φ ≪ a0. It should also (iii) be in harmony with cosmological observations including the CMB and MPS, (iv) reproduce the observed gravitational lensing of isolated objects without DM halos, and (v) propagate tensor mode gravitational waves at the speed of light". And also versions of BIMOND cover this.

Ultra-diffuse galaxies (those that are usually explained as that they have no or little DM):
https://arxiv.org/abs/1901.02679
They are probably due to the MOND EFE (External Field Effect), although AFAIK there are some outliers with large error bars in the distance to the nearby massive galaxy causing the EFE.

Although cosmology is not my area of expertise, I follow the discussion on MOND or dark matter as a hobby.

 

[astronomer]

I agree with you about MOND. In fact (I did some work on MOND, some also in collaboration with some of the Authors in the papers you mention) one of the phenomenological advantages of MOND over the (current) GR proposals, is that MOND has an unifying concept that (in principle) is able to deal with the points I mentioned in #248, i.e., the existence of an universal acceleration scale a_0. However, I do not think MOND is the solution of DM problem, for other reasons, but here we are discussing GR.

In the astronomical community, what it seems missing from the GR attempts (at least, those of what I am aware of) is the they deal with specific cases, often presenting quite "ad hoc" solutions (I searched in the literature the ideas behind GR solutions, and the span of the proposed solutions is quite amazing, ranging from GEM - and what about nn rotating galaxies? to GR retarded potential effects due to unsteady gas accretion - and DM evidence in gas free systems? to binary black holes placed along the symmetry axis of disk galaxies - do we really believe that all galaxies do have a binary black hole on the rotation axis??? to effects of vacuum solution of GR - but also in newtonian gravity you can add an harmonic function to the potential produced by the sources, without affecting the Poisson equation, and yet producing enormous fields if you like, but no one is advocating this possibility). Nothing wrong with it, but of course we all know very well that in astronomy, many problems are of "general" nature, i.e., the global picture is as important as the solution of a single, specific problem. Therefore, it would be already a step forward if general consensus is reached (for example) about the viability of GEM to really substitute DM. As a side note, it is worth to recall that the burden of the proof rests on the proponent. For example, a very well known fact of disk dynamics is the instability of self-gravitating stellar disk made by almost circular orbits (as orbits are in disk galaxies), and it has been established beyond discussion that a DM halo can stabilize the disk (it should be recalled that the stability of disk galaxies was the first hint of the existence of DM halos in disk galaxies, before the observation of flat rotation curves). What about GR?

 

[Structure seeker]

Hmm, I'm unsure whether we were discussing GR or more comparing it to $\Lambda CDM$ and Deur's theory and GEM.

 

[PeterDonis]

it has been established beyond discussion that a DM halo can stabilize the disk (it should be recalled that the stability of disk galaxies was the first hint of the existence of DM halos in disk galaxies, before the observation of flat rotation curves). What about GR?

Your terminology here might be somewhat confusing. Dark matter is a "GR" solution--one which contains matter (stress-energy) that is not visible and whose only effects are gravitational. It still uses the Einstein Field Equation of GR, with no modifications; the only "modification" from a model that only includes the matter visible to us is to also include matter that is not visible to us. The matter content (stress-energy tensor) is a free parameter in GR.

The Deur proposals which have been discussed in this thread are also (at least some of them) "GR" solutions, in that they (claim to) use the Einstein Field Equation of GR, with no modifications. They just (claim to) include effects which are (claimed to) not be included in the approximations used by "standard" models.

In other words, the issue between these two types of models is not who is using "GR". It is the simple fact that nobody can solve the Einstein Field Equation exactly for the given conditions, so everybody has to use approximations. And approximations inherently involve assumptions about what is significant and what is not. Deur's claims (at least in his proposed models that use the EFE without modification) amount to saying that there are effects that are significant but are left out of the "standard" approximations.

MOND is in a separate category here because it is not "GR"--it does not use the Einstein Field Equation without modification. But this thread is not really supposed to be about MOND but about Deur's proposals.

 

[astronomer]

Fully agree.

 

[Adrian59]

The $\Lambda CDM$ model is based on the Friedmann equations, which in turn are solutions to the Einstein Field Equation of GR.

None of this has anything to do with the $\Lambda CDM$ model. You are talking about models of individual galaxies that are developed to try and explain their observed rotation curves. $\Lambda CDM$ is a model of the universe as a whole.

This is the point I was making, maybe you put it better!

 

[timmdeeg]

The most obvious candidate for the discrepancy is that the self-interaction term of the GR Lagrangian includes a physical constant that should in principle be determinable from first principles in GR using only Newton's constant G and the geometry of the mass distribution. Deur doesn't actually calculate this constant from first principles in GR, however. Instead, he uses the same observations used to establish the MOND critical acceleration a₀ to calibrate this constant in the case of mass distributions with a geometry in line with an idealized spiral galaxy. He basically has faith that the physical constant in question could be calculated and would reproduce the observed value, without actually doing that involved calculation and no later paper from Deur has attempted to do that calculation.

As best I can discern, however, the paper reviewing Deur's work, basically does appear to use first principles to establish the value of this physical constant and in doing so finds that the effect is to small.

My analysis above of what I surmise is going on involves some guesswork and reading between the lines because the paper critical of Deur's approach basically reconstructs its own scalar approximation from scratch and compares its end results with Deur's, rather than going step by step through the analysis that Deur did in order to pinpoint where they believe he veered off course. It could be that this is not actually the issue, but it seems like the point of Deur's analysis most prone to a magnitude of conventional GR self-interaction effect outcome.

If I am right about what is going on, then the basic issue is that Deur is implicitly modifying GR by assuming a stronger coupling between gravitons of a given mass-energy than the coupling between gravitons and other fundamental particles from the Standard Model. (I'd be particularly curious if the strength of the attraction between gravitons was implicitly the square of the naive expectation, but I don't know how the observationally estimated parameter and the first principles parameter compare to each other.)

Could you please elaborate a bit about the "physical constant" included in the GR Lagrangian you are mentioning. Does this constant refer to a paper of Deur, or the recent paper of Barker et al?

If I remember correctly Deur mentioned coupling of gravitons in his earlier papers und uses the term field self-interaction later. Doesn' that mean the same?

 

[Adrian59]

This is just absurd. Lambda-DCM is described by a GR metric and nothing else - the FLRW metric family. All its physical content is contained in this full GR model. This is a theory of cosmology. There is no Lambd-CDM model of galaxy. That is a separate problem, in which dark matter is typically used, but it is not using the Lambda-CDM cosmological model.

I can see where the problem lies. Possibly, I should have made my entries clearer, but I was using the 'ΛCDM' term in this thread to mean the use of dark matter in galaxy dynamics, since that was what this thread was discussing. I should have specified dark matter and not ΛCDM (c.f. #254). Apologies for the confusion.

ΛCDM; that is a model of the universe, not of galaxies.

It is interesting that both the quotes above (from PAllen: 'There is no lamba-ΛCDM model of galaxies', and PeterDonis: 'ΛCDM; that is a model of the universe, not of galaxies', appear to want to separate galaxy dynamics from full ΛCDM. I thought one of the strengths of ΛCDM was that it was a full description of the cosmos. Have I got this wrong? MOND and even Deur's position is of a unifying description of the cosmos.

 

[PeterDonis]

I thought one of the strengths of ΛCDM was that it was a full description of the cosmos. Have I got this wrong?

Yes. Nobody has "a full description of the cosmos" in the sense you mean--a single unified detailed model that accounts for everything at all scales.

$\Lambda C D M$ uses a unified theory, General Relativity, that covers phenomena at all scales. But $\Lambda C D M$ itself is a particular model built using that theory to cover a particular domain, the evolution of the universe as a whole. It certainly does not also include detailed models of individual galaxies; like all models of the universe as a whole, it treats the matter in the universe as a continuous fluid and ignores the clumping at smaller scales.

Similarly, models of galaxies that attempt to explain their rotation curves are models of galaxies. They are not models of either the universe as a whole or systems on smaller scales, such as individual stars or solar systems. This applies to MOND and Deur's models just as much as to dark matter models.

MOND and even Deur's position is of a unifying description of the cosmos.

No, they aren't. As above, nobody has a single unified detailed model that accounts for everything at all scales. MOND's and Deur's models of galaxies do not also include the evolution of the universe as a whole. They might claim that they could, using similar general principles, construct separate models of the universe as a whole, but those would still be separate models.

 

[timmdeeg]

No, they aren't. As above, nobody has a single unified detailed model that accounts for everything at all scales. MOND's and Deur's models of galaxies do not also include the evolution of the universe as a whole. They might claim that they could, using similar general principles, construct separate models of the universe as a whole, but those would still be separate models.

Deur claims that field self-interaction "increases" gravity within and as a consequence "decreases" gravity outside mass distributions and hence mimics Dark Matter and $\Lambda$ as well. In his model of the universe the SN Ia data aren't interpreted as accelerated expansion but as to be due to the universe' anisotropy, in other words, the cosmological principle doesn't hold according to Deur. Consequently Deur's modified Friedmann equations don't contain $\Lambda$ but an anisotropy term instead.
I wonder however if Deur's anisotropic universe is a "separate model" (in your sense) or if follows from his claimed field self-interaction. It would fit to this picture that according to some articles which investigate certain models of an anistropic universe the necessity to assume $\Lambda$ can be dropped. I haven't these articles at hand though.

Whereas MOND "just" claims to mimic DM.

 

[PeterDonis]

Deur claims that field self-interaction "increases" gravity within and as a consequence "decreases" gravity outside mass distributions

Yes, but in my description, this corresponds to the theory (GR), not the specific model. He still does not have a single unified model that captures both the evolution of the universe as a whole and the dynamics of individual galaxies. Nobody does.

 

[Adrian59]

Yes. Nobody has "a full description of the cosmos" in the sense you mean--a single unified detailed model that accounts for everything at all scales.

$\Lambda C D M$ uses a unified theory, General Relativity, that covers phenomena at all scales. But $\Lambda C D M$ itself is a particular model built using that theory to cover a particular domain, the evolution of the universe as a whole. It certainly does not also include detailed models of individual galaxies; like all models of the universe as a whole, it treats the matter in the universe as a continuous fluid and ignores the clumping at smaller scales.

Similarly, models of galaxies that attempt to explain their rotation curves are models of galaxies. They are not models of either the universe as a whole or systems on smaller scales, such as individual stars or solar systems. This applies to MOND and Deur's models just as much as to dark matter models.

I agree that we haven't got a single equation to describe all scales, but I am less sure that ΛCDM is not a theory of galaxies as well. Probably I need to be more specific. I suppose what I am really referring to is what is known as the 'Standard Model of Cosmology' (SMC) which as its name implies is a complete model, and as such should have all scales covered. I often somewhat lazily use the term ΛCDM as synonymous with SMC. To me the SMC contains the big bang, inflation, baryogenesis, nucleosynthesis, dark matter and dark energy. Strictly speaking ΛCDM is dark energy (Λ) and cold dark matter (CDM). Of course behind all this is GR and the Friedmann equations on the large scale.

However, galaxy rotation curves are described by CDM in the Navarro-Frenk-White (NFW) profile which to me is still part of SMC and ΛCDM. My understanding is that the requirement for cold dark matter is especially necessary for the galactic scale.

 

[PeterDonis]

I often somewhat lazily use the term ΛCDM as synonymous with SMC.

Ok, this makes it clearer what you meant, but you should understand that it is not standard usage. In fact the term "Standard Model of Cosmology" is not standard usage either as far as I know. But it is true that the field of cosmology contains models at all scales and that those models are supposed to be consistent with each other. It is also true that dark matter is believed by most cosmologists to be necessary in models at multiple scales (at the very least, the universe as a whole, and at the scale of galaxies).

 

[Adrian59]

This paper is claimed to be thoroughly refuted by https://arxiv.org/abs/2303.06115, so the debate goes on.

Having been active in this thread, and as such I have read the references offered by other contributors, including this one from Lasenby et al. I thought that there was an issue with this paper that I mentioned in #211. I asked for clarification on this but none has been forthcoming. Has anyone re-read this reference and able to comment. To aid any comment I will copy my original question below:

However, maybe someone can clear up an issue that I've found in the Lasenby et al paper, already referenced. The authors examine NGC 1560 saying, 'we will restrict our attention here to the model having the MN density profile (16), which can be treated almost entirely analytically and suffices to demonstrate the shortcomings of the overall approach'. Using this Miyamoto–Nagai (MN) density profile, they get equation (21) which is soluble by a numerical method. They comment that one can get a simpler expression by approximating and solving this expression (22). Both solutions are plotted in figure 2.

What they do not plot in this figure are the experimentally observed values though the authors do say 'Although the rotation curves obtained using either (21) or the analytic approximation (22) appear to fit the rotation curve data for NGC 1560 shown ... in a pleasing way'. Although, had they plotted the experimentally observed velocity values, one would see that these derived curves from the gravito-magnetic approach are compatible with these experimentally derived values. So it is difficult to square this with an alleged failure of this novel approach.

The authors, then, plot the curve without a gravito-magnetic correction and get a standard rising and decreasing curve, the one usually shown as evidence of the need for dark matter, on a separate graph (figure 3), and plot the same combined line from figure 2. They comment that this standard 'curve peaks at velocities around 420 km s−1 (readopting SI units for the moment); this is much higher than one would expect for what is meant to be a dwarf galaxy'.

But, where do you get this plot from since the experimentally observed values are no where near this, but quite accurately match the values obtained with the gravito-magnetic approach?

 

[astronomer]

I think GEM solutions are quite clearly ruled out. The burden of proof is on the proponent, not on people producing counter arguments, and the counter arguments seem quite strong. In any case the single argument that I think should close the debate about GEM (and we can notice that no answer is provided) is why DM is required by newtonian gravity also in NON rotating system. Unless GEM answer in a clear way to this point, the whole discussion looks quite strange

 

[kodama]

I think GEM solutions are quite clearly ruled out. The burden of proof is on the proponent, not on people producing counter arguments, and the counter arguments seem quite strong. In any case the single argument that I think should close the debate about GEM (and we can notice that no answer is provided) is why DM is required by newtonian gravity also in NON rotating system. Unless GEM answer in a clear way to this point, the whole discussion looks quite strange

why DM is required by newtonian gravity also in NON rotating system. Unless GEM answer in a clear way to this point, the whole discussion looks quite strange

does this also apply for gravitational self interaction ?

 

[astronomer]

I have no idea about gravitational self interaction, however I think it could be of some help for sake of clarity to reach a definite (hopefully) conclusion at least on some of the proposed possibilities to substitute DM with some GR unexpected, weak field effect. It seems quite obvious that "solutions" SPECIFICALLY based on rotation are ruled out on empirical grounds, they cannot be fixed by sophisticated math or physical ideas. If we agree on this point, we can move forward and ask for example why systems of similar mass/structure can require in newtonian gravity largely different amounts of DM. What is the convincing, clear, and robust answer to this point provided by gravitational self interaction? As an astronomer, I would like to see at least an attempt to deal with this low-level question, before embarking in complicated discussions about unexpected GR phenomena....