CMBR Evidence for Non-Baryonic Dark Matter

[Ranku]

It is said that CMBR indicates that dark matter is non-baryonic. How so?

 

[pinball1970]

It is said that CMBR indicates that dark matter is non-baryonic. How so?

A lot on the net about this if you Google, 'Dark matter non baryonic.'
Tricky for me to give you a pointer though as I am not qualified.
@phinds

 

[phinds]

Like @pinball1970, I am aware that the CMB indicates non-baryonic dark matter (because I recall having read that) but the details are beyond me.

 

[George Jones]

It is said that CMBR indicates that dark matter is non-baryonic. How so?

Try googling baryon loading CMB. The following seems to be a decent non-mathematical exposition:

https://plus.maths.org/content/cosmic-oracle-2

 

[Hyperfine]

Lyman Page's book "The Little Book of Cosmology" (Princeton University Press, 2020) might also be helpful.

Also, Bruce Partridge's discussion of the CMB in Helge Kragh and Malcolm S. Longair, Eds., "The Oxford Handbook of the History of Modern Cosmology", Chapter 8, Oxford University Press, 2019.

 

[kimbyd]

The basic reason is because before the CMB was emitted, the universe was a plasma. While the universe was a plasma, baryonic matter experienced pressure. This meant that on large scales, when normal matter fell into a gravitational potential well, the pressure would cause it to bounce back. Dark matter, on the other hand, doesn't experience pressure and so can't bounce.

This leads to a very clear signal in the power spectrum of the CMB, where the even-numbered peaks are suppressed because of the lack of bouncing (dark matter contributes to only the odd-numbered peaks on the CMB).

A number of years ago, Max Tegmark put together a series of videos which show how various parameters impact the CMB power spectrum, if you want a more detailed picture of this:
https://space.mit.edu/home/tegmark/cmb/movies.html

 

[ohwilleke]

It is said that CMBR indicates that dark matter is non-baryonic. How so?

It would be more accurate to say that the CMBR is consistent with a universe with non-baryonic dark matter, than that it "indicates" that this is the case.

This is because the CMBR peaks seen could also have another BSM source than non-baryonic dark matter (e.g. some modified gravity theories).

 

[kimbyd]

It would be more accurate to say that the CMBR is consistent with a universe with non-baryonic dark matter, than that it "indicates" that this is the case.

This is because the CMBR peaks seen could also have another BSM source than non-baryonic dark matter (e.g. some modified gravity theories).

I haven't seen any modified gravity theory which comes close to reproducing the CMBR power spectrum without dark matter. It seems quite unlikely to me to be possible given the nature of the physics involved.

 

[ohwilleke]

I haven't seen any modified gravity theory which comes close to reproducing the CMBR power spectrum without dark matter. It seems quite unlikely to me to be possible given the nature of the physics involved.

You would be wrong in thinking that, although its a quite reasonable thing to assume. See, e.g.:

Constantinos Skordis, Tom Złosnik, "A new relativistic theory for Modified Newtonian Dynamics" arXiv (June 30, 2020) (127 Phys. Rev. Lett. 161302 (2021)).

MOG 54(2) Int.J.Theor.Phys. 484-505 (2015).

A. Deur, "Effect of the field self-interaction of General Relativity on the Cosmic Microwave Background Anisotropies"
arXiv:2203.02350 (March 4, 2022) (39 Class. Quantum Grav. 135003 (2022)).

W.M. Stuckey, Timothy McDevitt, A.K. Sten, Michael Silberstein, "The Missing Mass Problem as a Manifestation of GR Contextuality" 27(14) International Journal of Modern Physics D 1847018 (2018). DOI: 10.1142/S0218271818470181.

Indeed, it is reasonable to hypothesize, based upon these examples, that almost any modified gravity theory that can reproduce flat rotation curves will, in its relativistic generalization, reproduce the CMB peaks.

 

[Hyperfine]

Did any of the papers listed immediately above provide a fit to the TE cross-correlation power spectrum? I am curious about their fits to the dip in that power spectrum in the region of multipole moments of 100.

 

[ohwilleke]

Did any of the papers listed immediately above provide a fit to the TE cross-correlation power spectrum? I am curious about their fits to the dip in that power spectrum in the region of multipole moments of 100.

I haven't read the papers for a while and never with those specific details in mind. All of them are linked with open access full text papers available. Feel free to review them yourself. You can eyeball the charts above as well.

 

[Hyperfine]

Indeed, it is reasonable to hypothesize, based upon these examples, that almost any modified gravity theory that can reproduce flat rotation curves will, in its relativistic generalization, reproduce the CMB peaks.

And yet to summarily dismiss a question regarding fits to an observed CMB power spectrum.

In addition, how do these proposed theories treat observed galactic rotation curves that are Keplerian? For example AGC 114905? Reference: https://arxiv.org/abs/2112.00017

 

[ohwilleke]

And yet to summarily dismiss a question regarding fits to an observed CMB power spectrum.

In addition, how do these proposed theories treat observed galactic rotation curves that are Keplerian? For example AGC 114905? Reference: https://arxiv.org/abs/2112.00017

In that case of ACG 114905, the most likely cause is an angle of inclination measurement error in the initial assessment that it is Keplerian (an uncertainty that the initial paper describing ACG 114905 as a Keplerian galaxy itself identifies as a potential source of a serious problem with its assessment), citing Banik et. al. (2022) titled, “Overestimated inclinations of Milgromian disc galaxies: the case of galaxy AGC 114905“.

In other cases, an external field effect is the likely cause (at least in MOND). For example, in the case of the small velocity dispersion observed in the two group UDGs NGC 1052-DF2 and NGC 1052-DF4, inferring dynamical masses close to their stellar masses, taking the External Field Effect into account removes the tension between an apparent lack of MOND effects and the baryonic mass distribution in the galaxies. See Kroupa et al. (2018); Famaey et al. (2018); Haghi et al. (2019); and Müller et al. (2019).

One would also expect rare modified gravity prediction anomalies in systems that are far from equilibrium, although I'm not aware of any particular Keplerian galaxies in which this explanation is required.

 

[Vanadium 50]

I haven't seen any modified gravity theory which comes close to reproducing the CMBR power spectrum

Actually, the 3rd peak was predicted by McGaugh in 1990 or so in MOND before ΛCDM.

This is not the great victory that it sounds like. The statement in both models is that matter is "clumpier" than you would expect from visible matter alone. In MOND its because the phase space for gravitational capture is larger (i.e. faster moving objects are still bound, rather than escape) and in ΛCDM you have more matter and therefore more binding than visible matter indicates.

Since they are both saying the same thing, albeit for different reasons, you expect similar predictions.

Where I think this might get interesting is coupling this with BBN. Both models will likely reproduce the observed ⁴He fraction, as that is hard to move, what do they do with deuterium and lithium?

 

[ohwilleke]

Where I think this might get interesting is coupling this with BBN. Both models will likely reproduce the observed ⁴He fraction, as that is hard to move, what do they do with deuterium and lithium?

Given how soon BBN occurs after the Big Bang (up to about ca. 15 minutes), it could be that the background gravitational field almost everywhere is greater than the MOND constant a₀, in which case, it would have no effect.

 

[Vanadium 50]

That's not the effect. In MOND the density of ordinary (i.e. baryonic) matter is greater and that might imprint itself on the universe's chemistry.

 

[kimbyd]

You would be wrong in thinking that, although its a quite reasonable thing to assume. See, e.g.:

Constantinos Skordis, Tom Złosnik, "A new relativistic theory for Modified Newtonian Dynamics" arXiv (June 30, 2020) (127 Phys. Rev. Lett. 161302 (2021)).

This one uses a TeVeS theory where the scalar component is contrived to mimic dark matter (that is, its energy density scales close to $1/a^3$). I guess I'm not terribly surprised this could replicate the CMB data, as it's basically an alternative form of dark matter theory.

A. Deur, "Effect of the field self-interaction of General Relativity on the Cosmic Microwave Background Anisotropies"
arXiv:2203.02350 (March 4, 2022) (39 Class. Quantum Grav. 135003 (2022)).

This one isn't a modified gravity theory at all. It's a claim that actually, physicists have been doing General Relativity calculations incorrectly, and doing them the right way, according to this author, removes the need for both dark matter and dark energy. Suffice it to say I am very skeptical.

I don't have the expertise to be able to identify any problems with the theory, but it doesn't look like this person's papers have been interacted with much at all by the broader cosmology/GR community either, so it's hard to say what other theorists think of these pretty outlandish claims.

W.M. Stuckey, Timothy McDevitt, A.K. Sten, Michael Silberstein, "The Missing Mass Problem as a Manifestation of GR Contextuality" 27(14) International Journal of Modern Physics D 1847018 (2018). DOI: 10.1142/S0218271818470181.

Your link is broken for this one. Here is a link that I found:
https://arxiv.org/abs/1509.09288

I really don't like the theoretical approach they're taking. They're basically saying that they're not modifying gravity, and then proceed to modify gravity. It's just weird.

But the bigger problem is that their fit parameters to get that CMB graph don't match the cosmology from the nearby universe (basically they have a far, far larger Hubble tension than we currently see with $\Lambda$CDM).

 

[kimbyd]

Actually, the 3rd peak was predicted by McGaugh in 1990 or so in MOND before ΛCDM.

Perhaps in terms of the overall structure, but I don't believe it is able to reproduce the CMB in detail. Similar to the above theory which sorta works, but gets the parameters way off.

 

[ohwilleke]

As a preface to responses below, keep in mind that the four examples that I provided were offered up merely as a proof of concept.

It is in fact possible to describe the CMB peaks in a purely gravitational theory without dark matter. I've provided four examples of that being done. Necessarily, not all of those examples are the correct solution. But it can be done. It isn't even terribly hard to get very good matches to all of the CMB peaks, as the charts in my post illustrate.

Also, the gravitational approaches are in some cases matching the CMB peaks with fewer free parameters than the ΛCDM match.

This one uses a TeVeS theory where the scalar component is contrived to mimic dark matter (that is, its energy density scales close to $1/a^3$). I guess I'm not terribly surprised this could replicate the CMB data, as it's basically an alternative form of dark matter theory.

It isn't a TeVeS theory (that theory was by the late professor Bekenstein who was a colleague of Milgrom; the theory's name is a Hebrew pun). But, it is a different relativistic generalization of MOND. It is not an alternative form of dark matter theory.

I wonder if you may be commenting about MOG (in part) rather than Rel-MOND, but I can't tell. MOG, like TeVeS, is a theory with a tensor, a vector, and a scalar field.

This one isn't a modified gravity theory at all. It's a claim that actually, physicists have been doing General Relativity calculations incorrectly, and doing them the right way, according to this author, removes the need for both dark matter and dark energy. Suffice it to say I am very skeptical.

I don't have the expertise to be able to identify any problems with the theory, but it doesn't look like this person's papers have been interacted with much at all by the broader cosmology/GR community either, so it's hard to say what other theorists think of these pretty outlandish claims.

Whether it is truly incorporating an overlooked GR effect (specifically, the self-interaction of the gravitational field), or it is actually a modification of gravity, it doesn't matter much, because the fact is that it works exquisitely well to reproduce the observations.

Deur's approach replicates MOND in spiral galaxies, predicts an observed relation between apparent dark matter component and ellipticity in elliptical galaxies (absent in MOND and not predicted in other theories), fits cluster data (which MOND fails to do), explains the Bullet Cluster, predicts earlier galaxy formation than ΛCDM (consistent with the JWST results), is consistent with the EDGES 21cm data, and fits the CMB peaks, in an inherently relativistic model. It is also the only theory I've seen that has an explanation for dark energy phenomena that honors global as well as local mass-energy conservation.

It has fewer free parameters than GR (since it doesn't have a cosmological constant) or MOND (which adds the constant a₀) or ΛCDM (which adds new physical constants related to dark matter particle properties and amounts as well as a cosmological constant).

It doesn't add any new forces or new particles to GR plus the SM (apart from a possible massless spin-2 plain vanilla graviton with a coupling constant functionally related the Newton's constant G), although it may actually be a subtle modification of GR rather than actually being an overlooked GR effect (I also lack the expertise to know). It doesn't add a scalar or a vector field to the tensor field of GR.

His approach only modifies GR as conventionally applied in astrophysics in distributions of mass-energy that aren't spherically symmetric in very weak gravitational fields since gravitational field self-interactions are inherently second order effects. So, it doesn't disturb any of the strong field conclusions of GR in contexts where GR is well proven like black holes, neutron stars, and binary compact object systems.

Deur make predictions with a better fit to the data and range of applicability than either MOND (which fails in clusters) or ΛCDM (which fails at galaxy and galaxy cluster scales and has cosmological scale issues as well). Needless to say, GR as conventionally applied without dark matter is also a horrible fit to the observed reality in a wide variety of circumstances (even with a cosmological constant).

Unlike MOND, Deur's approach isn't simply a toy model phenomenological fit to the data. Deur is working from very vanilla axioms and the GR Lagrangian.

So, if Deur's approach turns out that it is actually a GR modification, rather than correctly being described as an overlooked GR effect, who cares?

He's published seven papers in peer reviewed journals in the field since 2009 (he is primarily a QCD physicist and has published many other papers in that field, and is basically working from the gravity as QCD squared paradigm). Some of those papers have co-authors.

I agree that it would be better if his work was vetted by more physicists in the field. But at a minimum, no one has published a rebuttal to any of the published peer reviewed papers and his approach deserves more attention.

Your link is broken for this one. Here is a link that I found:
https://arxiv.org/abs/1509.09288

I really don't like the theoretical approach they're taking. They're basically saying that they're not modifying gravity, and then proceed to modify gravity. It's just weird.

But the bigger problem is that their fit parameters to get that CMB graph don't match the cosmology from the nearby universe (basically they have a far, far larger Hubble tension than we currently see with ΛCDM).

Apologies for the bad link. Probably fat fingers on my part. Thanks for finding a good link.

The authors have discussed their work in some threads at PF.

Conceptually, it is more similar to Deur than to other gravity based theories to explain dark matter phenomena. Unlike many efforts to explain dark matter phenomena in rotationally support galaxies with GR, the authors aren't relying of gravitomagnetic (GEM) effects and instead are relying on different non-linear effects in GR.

 

[ohwilleke]

Stacy McGaugh has posted a lengthy and well supported blog entry discussing the issues of what the CMB says about dark matter v. MOND, and about the history of the development of LambdaCDM v. what we knew about CMB.

 

[timmdeeg]

So, if Deur's approach turns out that it is actually a GR modification, rather than correctly being described as an overlooked GR effect, who cares?

Deur's approach isn't a GR modification. We show here that within GR’s formalism, these terms can be regrouped in an overall term D(z) factoring the right hand side of Eq. (9). Instead Deur drops the cosmological principle, see his modified Friedmann equations (9) and (14), representing an inhomogeneous and anisotropic universe.

In Deur's universe the SN Ia data are due to its anisotropy and not to accelerated expansion.

So from this it seems that field self-interaction is an ingredient within GR. But I agree, "who cares?" should empiric data you mentioned support Deur.

 

[PeterDonis]

Deur's approach isn't a GR modification. We show here that within GR’s formalism, these terms can be regrouped in an overall term D(z) factoring the right hand side of Eq. (9).

One thing that's missing from this, though, is an actual derivation of those modified equations from the field equation. I don't see that anywhere; Deur just asserts equations (10), (11), (12), and (13) without argument or derivation.

 

[timmdeeg]

One thing that's missing from this, though, is an actual derivation of those modified equations from the field equation. I don't see that anywhere; Deur just asserts equations (10), (11), (12), and (13) without argument or derivation.

Isn't there literature describing how equation (5) based on assuming the cosmological principle evolves to a modified equation if that principle is dropped? To me equation (10) seemed a bit like hand-waving. I wonder if the property "anisotropy" can be described by a single number with a correction factor as shown in equation (10), but thought that's still text book physics.

 

[PeterDonis]

Isn't there literature describing how equation (5) based on assuming the cosmological principle evolves to a modified equation if that principle is dropped?

I believe there are papers treating models that are not isotropic or homogeneous, but I don't have any specific references.

A key issue with any such model is that, without homogeneity and isotropy, it's no longer clear how to even construct a coordinate chart on the spacetime, since there are no longer a set of observers or spacelike slices picked out by any symmetries of the spacetime.

 

[timmdeeg]

One thing that's missing from this, though, is an actual derivation of those modified equations from the field equation. I don't see that anywhere; Deur just asserts equations (10), (11), (12), and (13) without argument or derivation.

Possibly, if Deur says "... , these terms can be regrouped in an overall term D(z) factoring the right hand side of Eq. (9)." doesn't mean more that D(z) stands for a deviation from homogeneity and isotropy without referencing to a specific model. As such it is not a derivation.

The key point seems that any such deviation from homogeneity and isotropy influences the luminosity distance and hence the interpretation of the SN Ia data:

https://arxiv.org/pdf/1412.8371.pdf

The Supernovae observations are good tests about the structure of the space-time on different scales. This is a very important point, in fact some years ago Zel’dovich [35] studied the importance of the effects of the inhomogeneities on light propagation and also in the next years [36–42]. To check this model we calculate the luminosity distance in order to compare theoretical approach with experimental data. We explain the acceleration of the Universe without invoking the presence of a cosmological constant or dark energy.
...
We are sure that the voids in the Universe dominate, while matter is distributed in a filamentary structure. Therefore photons must travel through the voids and the presence of inhomogeneities can alter the observable with respect to the corresponding FLRW model of Universe, homogeneous and isotropic.
The key point is that in this model we have two contributions to the Hubble diagram of the Supernovae Ia: inhomogeneity to the large scale geometry and 9 anisotropy can generate dynamically effects that may remove the need for the postulate of dark energy.

So "may remove the need for the postulate of dark energy" shows that Deur isn't alone.
Apart from Deur's work it seems the cosmic distance ladder isn't carved in stone. Perhaps the Hubble tension points to that.

 

[Adrian59]

It is said that CMBR indicates that dark matter is non-baryonic. How so?

I am not sure what kind of detail to your question you were expecting? The currently accepted explanation is that given by:

The basic reason is because before the CMB was emitted, the universe was a plasma. While the universe was a plasma, baryonic matter experienced pressure. This meant that on large scales, when normal matter fell into a gravitational potential well, the pressure would cause it to bounce back. Dark matter, on the other hand, doesn't experience pressure and so can't bounce.

This leads to a very clear signal in the power spectrum of the CMB, where the even-numbered peaks are suppressed because of the lack of bouncing (dark matter contributes to only the odd-numbered peaks on the CMB).

Essentially the signal in the CMBR that gives rise to estimations of dark matter are the variations or anisotropies that are plotted in a now well-known graph shown by Ohwilleke in #9. Both the dark matter/ dark energy model and the modified gravity model can more or less reproduce this graph and the issues are in the fine details.

 

[horacio torres]

The basic reason is because before the CMB was emitted, the universe was a plasma. While the universe was a plasma, baryonic matter experienced pressure. This meant that on large scales, when normal matter fell into a gravitational potential well, the pressure would cause it to bounce back. Dark matter, on the other hand, doesn't experience pressure and so can't bounce.

This leads to a very clear signal in the power spectrum of the CMB, where the even-numbered peaks are suppressed because of the lack of bouncing (dark matter contributes to only the odd-numbered peaks on the CMB).

A number of years ago, Max Tegmark put together a series of videos which show how various parameters impact the CMB power spectrum, if you want a more detailed picture of this:
https://space.mit.edu/home/tegmark/cmb/movies.html

Dark Mather is a real headache to understand

Horacio