Dark Matter is real

[mathman]

Main Question or Discussion Point

Two recent studies have found galaxies with little or no apparent dark matter, indicating modifying gravity can't work.
https://iopscience.iop.org/article/10.3847/2041-8213/ab0e8chttps://iopscience.iop.org/article/10.3847/2041-8213/ab0d92/meta

 

[PeterDonis]

Two recent studies have found galaxies with little or no apparent dark matter, indicating modifying gravity can't work.

Just to be clear, what these galaxies have is a velocity dispersion that is what is expected from the visible matter alone using the standard theory of gravity, correct?

On the dark matter hypothesis, this is possible since it just means these galaxies have negligible amounts of dark matter. In other words, the dark matter model has a free parameter that can be used to predict the velocity dispersion of these galaxies as well as the velocity dispersion of all the other galaxies. Which still leaves an open question of why these galaxies have no dark matter--what makes these galaxies different from all the ones that have large dark matter halos?

But on the modified gravity hypothesis, these galaxies should not exist; every galaxy should have the modified velocity dispersion (the one that, in the dark matter model, is due to the dark matter halo) because in the modified gravity theory this dispersion is a consequence of the visible matter alone, plus the modified gravity equations, so there is no free parameter that can be adjusted. Proponents of modified gravity have given this as an advantage of the theory--fewer free parameters means a simpler theory--but this advantage becomes a fatal liability if new data falls outside the theory's predictions. As Einstein said, every theory should be as simple as possible, but not simpler.

 

[hyksos]

MOND (or "Modified Newtonian Dynamics") is a theory that suggests the canonical equations of gravity do not apply to the dynamics of stars orbiting galaxies , whenever accelerations become vanishingly small. The theory's motivation was to explain anomalous rotational velocities of stars in the outer arms of galaxies. The visible matter in the galaxies' cannot account for the high velocities, and perhaps this is due to the presence of a halo of Dark Matter centered around the galaxies' centers. Or alternatively, if MOND is correct, those velocities simply follow from fundamental physics, and no extraneous Dark Matter is needed.

The danger with modifications to fundamental physical law is that a single counter-example can falsify them. A darkless galaxy was first discovered by the Hubble Telescope and the 10-meter Keck in March of 2018. It was ignored as a statistical fluke. Then another darkless galaxy was found again in March 2019 : Galaxy NGC1052-DF4.

Both DF4 and its precursor are ultra-diffuse, and so contain 1000s of times less stars than the average galaxy. This is worse for MOND, since the accelerations would be even smaller than usual, and the discrepancies predicted by the theory would be relatively larger. This was not observed. Instead both darkless galaxies have rotational velocity curves consistent with canonical gravity. It is as if they simply lack the special Dark Matter that causes the anomalies in most galaxies.

Can MOND survive as a theory given these recent discoveries?

 

[zonde]

Proponents of modified gravity have given this as an advantage of the theory--fewer free parameters means a simpler theory--but this advantage becomes a fatal liability if new data falls outside the theory's predictions.

If observations falsify the theory it means it is indeed a good theory, just not the one that describes reality.

 

[Nik_2213]

Been skimming a conference proceedings on Local Bubble & Co.
One item made the point that supernova bubbles in a low mass galaxy etc can eject stuff faster than the local escape velocity, send it clean out into intergalactic space.
Okay, it may stay in the 'local group', but the shock-driven material and all it sweeps up is travelling too fast to fall back when it cools, as it would in eg Milky Way or Andromeda.

Could such ejecta carry 'Dark Matter' along ??

 

[Vanadium 50]

This is the same galaxy as last time. The degree of DM depends on its distance from us, and there is considerable disagreement on that.

Personally, I am always skeptical of "the single badly measured point that falsifies my opponent's theory thereby proving me right all along".

 

[Bandersnatch]

This is the same galaxy as last time.

There's two now. The second link in the OP is about the new one.

 

[Tghu Verd]

Could such ejecta carry 'Dark Matter' along ??

Possibly. We presume that DM only interacts via gravity (and a further general assumption is "at large scales"), so there would be an interaction with the ejecta, but whether that is sufficient to drag the DM with it - or the counterpoint that the DM inhibits the spread of the ejecta - would have to be an open question given that we still don't understand DM at anything but the "there seems to be something out there" level.

And perhaps the DM's escape velocity is different than the ejecta, so whether it falls back or keeps traveling may be different to baryonic matter.

Honestly, despite the experiments and measurements that suggest a Lambda-CDM model, it's all conjecture at the moment, we really don't know.

 

[Vanadium 50]

There's two now. The second link in the OP is about the new one.

But the new one is in the same association as the old one. I think everyone agrees DF2 and DF4 are at the same distance. I don't think there is agreement on what that distance is.

 

[Ken G]

The question may still be open, but it's not good news for MOND. It is certainly quite ironic-- the systems that end up proving the existence of dark matter may be the ones that don't have it! More importantly, understanding why they don't have it might give us the crucial clues we need as to what it is and how it behaves. I agree with the above point that it is only a win for dark matter if that hypothesis is able to explain the situations where it is not present, since it was invented to explain the situations where it is, and thus the latter cannot be used as a confirmation of the theory, but the former can be if there is some sensible reason for it.

 

[Orodruin]

More importantly, understanding why they don't have it might give us the crucial clues we need as to what it is and how it behaves.

I am not an expert in structure formation, but it seems to me that explaining the existence of ultra-diffuse galaxies without dark matter could be possible in several ways. Since it would be speculation (against forum rules) and I don't want to make a fool of myself (bad karma), I will keep those thoughts to myself.

 

[Vanadium 50]

I'll just say that stripping the dark matter from a galaxy without disrupting it is not a simple thing to do. It's not impossible, but it adds to the list of oddball features to these galaxies. It would be unusual for our first example, and more unusual still for the first two examples.

Many of the odd features go away if the galaxies in question are closer to us. In particular, the unusually bright globulars (which in fact are the mass tracers used in determining M/L) look much more typical if they are closer.

Everyone agrees DF2 and DF4 are associated (but probably not bound to each other). This means that many of the pecuilar features are shared, so it's not a particulalrly clarifying data point. Another example somewhere else in the sky would be very valuable An inability to find a second example would also be clarifying.

 

[Ken G]

I don't think it would be too speculative to note that we have observed cases of baryonic gas getting separated from dark matter, as in the Bullet cluster. There, two galaxy clusters are colliding, and the dark matter that controls the galaxy gravitational potential wells just goes right through. But the gas between the galaxies, which is most of the baryonic matter in the universe, collides in the center and gets left behind. Then all you'd need is for that dark-matterless gas to form stars in the distant future, and it could form diffuse galaxies without much dark matter. So that would seem to be one potential mechanism that we can see playing out, though I don't know if there is an expectation for what will ultimately happen to that baryonic gas. (And the Bullet cluster is already a kind of "smoking gun" for dark matter, as no MOND models can explain the gravitational lensing pattern in that cluster, but as with the dark matterless galaxies, MOND proponents can claim the cluster is a kind of fluke that is not being correctly analyzed somehow.)

 

[strangerep]

Many of the odd features go away if the galaxies in question are closer to us. In particular, the unusually bright globulars (which in fact are the mass tracers used in determining M/L) look much more typical if they are closer.

Thanks for mentioning this.

But, (pardon my density),... could you explain this in a bit more detail, pls?

 

[BWV]

Do dark matter-less galaxies imply anything about the nature of dark matter or lend more support to one hypothesis vs another? Seems puzzling that DM would be absent in some tiny percentage of galaxies.

 

[Vanadium 50]

Bullet cluster

Yes, but there's all sorts of disruption in the Bullet Cluster.

The people who model such things tell me that the way you strip dark matter from a galaxy without disrupting it is to have multiple small interactions, not one big one. Not impossible, but unusual.

could you explain this in a bit more detail, pls?

DF2 and DF4 are examples of low-surface-brightness galaxies. LSB galaxies are dim, gas-rich, and tend to be found disproportionately away from other galaxies. They live quiet lives, because without gravitational interactions with nearby galaxies, star formation is not triggered, so they stay gassy. Wikipedia claims they don't have supernovae, and this isn't entirely true: SN 2009z appears to be from an LSB galaxy. But the rate is way, way down.

So this history of DF2 needs to be unusual to have had interactions just so - you need enough perturbation to get the dark matter out, but not enough to trigger star formation. You can make the same argument for DF4, and I would agree that it too has to be unusual, but since it's in the same general area, it probably is not any more unusual.

OK, now what is the actual measurement? The claim of "no dark matter" is actually a claim the M/L is unusually low. It's much lower than the typical M/L for LSB galaxies, which tends to be above M/L for bright galaxies. (That's what LSB means after all). They measure M by looking at redshifts of nearby globular clusters, presumed to be in orbit around DF2. (Converting angle to distance, redshift to velocity, and distance and velocity to mass). They measure L by seeing how bright it is.

Both depend on the distance to the galaxy - the farther away the galaxy is, the farther away the globular is from the central galaxy (we see only the angular displacement) and the heavier the galaxy they orbit must be. However, it is not this simple, since you need to subtract off the Hubble flow, and this means the farther away the galaxy is, the slower the relative motion of the globulars, and therefore the lighter the galaxy they orbit must be. These partially cancel and you end up with a complicated but well-understood relation between the velocity of the globulars, the distance to DF2 and its mass. In the range of distances we are talking about, I believe the closer DF2 is to us, the heavier it is.

As an additional complication, there are only ten tracers - ten globulars thought to be associated with DF2. Everybody would like more, as ten is less constraining than "more than ten". But it is what it is.

Luminosity, of course, depends on the inverse square of distance.

OK, onto specifics. "DF2 has no DM" is, as I said, a statement that M/L is too low, which can be interpreted as either M is too low (no DM), or L is too high - i.e. DF2 is too bright. The associated globulars are also unusually bright. If the distance to DF2 were ~13 Mpc instead of ~19 Mpc away, the globulars would be of normal brightness, and M/L for DF2 would be a value more typical for LSB galaxies.

There is controversy in the literature about whether 19-ish or 13-ish is the right value.

I'll reiterate my position: DF2 has some oddball features that make it less than the smoking gun that some have claimed. DF4 is associated with DF2 so it ends up with the same oddball features. Before accepting any explaination for these features, I'd like to see an example in some other part of the sky.

 

[Ken G]

I don't really see how these observations can be made sense of either with MOND, or with dark matter. If we say MOND is right, and there is not dark matter, then these galaxies have low M/L due to their weird star formation history that somehow increases L efficiency relative to the Milky Way. But then we have the problem that these galaxies are low surface brightness and should have weak star formation, so they should have high M/L not low M/L.

But if we say dark matter exists, and these galaxies have low M/L due to the loss of dark matter, then we might be pleased to see the M/L corresponds to the baryonic Milky Way M/L, without dark matter. But that doesn't make sense either, because you still have an unusual type of galaxy that shouldn't have a normal baryonic M/L ratio, regardless of its dark matter history. Hence the bottom line is, it doesn't matter if you think there's MOND, or if you think there's dark matter, either way you still have a diffuse LSB galaxy that has no business having a normal baryonic-only Milky-Way-like M/L. MOND vs. dark matter is actually irrelevant to the real puzzle here, which is why don't diffuse LSB galaxies always have high M/L?

You could certainly say the ones that aren't high have had their dark matter stripped, but you still can't get them down to baryonic Milky-Way M/L, which is where these apparently are. It would be like claiming that whatever causes these galaxies to be diffuse and LSB has no effect on their baryonic M/L, which seems hard to swallow. That means something is rotten in Denmark.

 

[PeterDonis]

The claim of "no dark matter" is actually a claim the M/L is unusually low.

I'm not sure this is the best way to say it. The claim of dark matter is based on more mass being present, as seen in rotation curves, than the visible matter can account for using standard gravitational theory. The claim of no dark matter in these particular galaxies simply means that is not the case: the rotation curves can be accounted for using just the visible matter with standard gravitational theory. So M/L isn't "unusually low" except in comparison to all the other galaxies that are hypothesized to have dark matter; but that really means that all those other galaxies have M/L unusually high--too high to be accounted for by the visible matter alone. These galaxies, by contrast, have M/L "just right"--just right to be accounted for by the visible matter.

Of course it's true that all of this depends on the estimate of L being right for these galaxies, which, as you note, is still open to question.

 

[PeterDonis]

If we say MOND is right, and there is not dark matter, then these galaxies have low M/L due to their weird star formation history that somehow increases L efficiency relative to the Milky Way.

What does this mean? I don't know of any free parameter in MOND that corresponds to "star formation history that somehow increases L efficiency relative to the Milky Way".

you still have an unusual type of galaxy that shouldn't have a normal baryonic M/L ratio

Why not?

 

[Ken G]

What does this mean? I don't know of any free parameter in MOND that corresponds to "star formation history that somehow increases L efficiency relative to the Milky Way".

It's not an element of the MOND theory, it is a logical ramification of applying MOND to the Milky Way galaxy and the low M/L galaxies at the same time. If all is baryon, then the lower M/L for those special galaxies must mean they are efficient (per baryon) at making L relative to the Milky Way. That is precisely the opposite of what we would expect for diffuse LSB galaxies which should have weak star formation, so does not seem a likely explanation of the situation.

Why not?

Same reason-- it's an LSB galaxy, so should not have the same L efficiency per baryon as the Milky Way. The point being, raising the Milky Way M/L with lots of dark matter doesn't explain why an LSB galaxy would have the same M/L as would the Milky Way if you restrict to baryon M. If one holds to the dark matter explanation, then the Milky Way has a deeper gravity well which should create a more rapid star formation rate and lower the M/L compared to a galaxy with no dark matter.

 

[PeterDonis]

If all is baryon, then the lower M/L for those special galaxies must mean they are efficient (per baryon) at making L relative to the Milky Way.

So again, what free parameter would you vary to change the "efficiency" at making L? You can't just wave your hands and say it can change; what free parameter allows it to change?

it's an LSB galaxy, so should not have the same L efficiency per baryon as the Milky Way

Same question as above.

 

[Ken G]

The star formation rate per baryon is the light-creating efficiency per baryon. That's the parameter in question. So the issue is, what is it about a galaxy that affects the star-forming efficiency? There are a lot of things, but one would certainly expect dark matter to deepen the gravity well and increase that efficiency, which can also be thought of as the timescale for baryons in the ISM of the galaxy to get included into stars. Wouldn't that timescale be expected to be longer in a diffuse LSB galaxy with no dark matter, than in a galaxy with dense spiral arms?

 

[PeterDonis]

The star formation rate per baryon is the light-creating efficiency per baryon.

Ok, so the free parameter is basically the fraction of baryonic matter that is in stars (as opposed to gas clouds, dust clouds, etc.). But while that would be expected to affect the overall visible brightness of the galaxy, it would not necessarily affect the rotation curve (since the non-visible baryonic matter would be expected to be distributed similarly to the visible baryonic matter). And it's the rotation curve, not the overall brightness, that is well matched to the visible matter in these new galaxies, whereas it does not match the visible matter well in galaxies which are believed to have large dark matter halos. And dark matter can explain that difference because it does not have the same interactions as baryonic matter, so its distribution can be very different. Whereas MOND says that all galaxies should have rotation curves that do not match the visible matter if you use standard gravity theory, because the whole point is that it's a different gravity theory, that changes the predicted rotation curve from the visible matter.

 

[Ken G]

Yes that's an interesting point, that we also have information from the rotation curve. I'm just saying that if we take dark matter to be the explanation, at first it seems satisfying that the M/L in these galaxies is similar to the baryonic M/L for the Milky Way. But on further thought, that actually sounds like a problem, since I would expect galaxies like this to be less efficient at forming stars (they are more diffuse and we are saying they don't have the deep dark matter well). If that's true, then we'd actually expect a higher M/L than for the baryons in the Milky Way, not the same M/L. It's just curious.

 

[ohwilleke]

Two recent studies have found galaxies with little or no apparent dark matter, indicating modifying gravity can't work.
https://iopscience.iop.org/article/10.3847/2041-8213/ab0e8chttps://iopscience.iop.org/article/10.3847/2041-8213/ab0d92/meta

Actually not.

In fact, systems like DF2 and DF4 are natural and predicted results of MOND as originally formulated in 1983 via its external field effect which essentially means that MOND effects do not arise when the system in buried in an external gravitational field stronger than the MOND cutoff strength when combined with the local gravitational field strength. DM in contrast, needs DM to form galaxies so these systems shouldn't exist in that theory.

This post (by one of the leading MOND researchers) suggests that the correctly calculated MOND prediction is 14+/- 4 and that the measured value is 8.4 with a 90% confidence interval upper limit of 10. So, it does not disprove MOND, the paper's calculation simply failed to consider the external field effect (a part of the original MOND theory since 1983 that is not widely known). A paper by MOND's inventor further spells out this scenario and others where the external field effect is absent or only partial, for example, in this published paper from the year 2000. The limited data points used in the calculation (ten) also suggests that the measured value is likely to be an underestimate as it was in FORNAX. From the comments to the previous link by its author:

On closer reading, I notice in the details of their methods section that the rms velocity dispersion is 14.3 km/s. It is only after the exclusion of one outlier that the velocity dispersion becomes unusually low. As a statistical exercise rejecting outliers is often OK, but with only 10 objects to start it is worrisome to throw any away. And the outlier is then unbound, making one wonder why it is there at all.

Consider: if they had simply reported the rms velocity dispersion, and done the MOND calculation correctly, they would have found excellent agreement. This certainly could be portrayed as a great success for MOND. Instead, tossing out just one globular cluster makes it look like a falsification. Just one datum, and a choice of how to do the statistics. Not a wrong choice necessarily, but a human choice… not some kind of statistical requirement.

One of the authors of the original DF2 papers addresses a variety of concerns (of the kind that quite honestly should have been addressed at a pre-print/peer review stage rather than post-publication) (hat tip Backreaction).

In particular, he justifies at great length his velocity dispersion calculation, although the paper really fails seriously in failing to address just how problematic and assumption prone it really is and the reasoning behind the choices made. The uncertainty due to fundamental assumption issues is greatly understated.

He acknowledges that he screwed up the MOND calculation and shifts attention from that mistake to a different dwarf galaxy (Dragonfly 44) where MOND might be off without conclusively showing that this is the case. The original DF2 article author states:

The whole MOND / alternative gravity discussion in the paper rests on a misunderstanding on my part.

He acknowledges the need for more and better data to get a more accurate measurement, some of which can be done quite easily (and really should have been done prior to publication in Nature).

He unconvincingly argues that "lacking" and "without" have different meanings while backpedaling on the "no" dark matter claim, although this criticism isn't honestly such a big deal since other language in the abstract does clarify the point (and indeed highlights that the dark matter a priori prediction was off by a factor of 100 v. a factor of about 0.4 at most for the correctly done MOND prediction).

These immediate responses to the claims of MOND falsification were elaborated in a more formal format in a 2018 pre-print which has been submitted to MNRAS for publication.

Crater II Compared

The same external field effect had already been predicted and observed in a galaxy known as Crater II.

Crater II is an unusual object among the dwarf satellite galaxies of the Local Group in that it has a very large size for its small luminosity. This provides a strong test of MOND, as Crater II should be in the deep MOND regime (gin≈34km2s−2kpc−1≪a0=3700km2s−2kpc−1). Despite its great distance (≈120 kpc) from the Milky Way, the external field of the host (gex≈282km2s−2kpc−1) comfortably exceeds the internal field. Consequently, Crater II should be subject to the external field effect, a feature unique to MOND. This leads to the prediction of a very low velocity dispersion: σefe=2.1+0.9−0.6kms−1.

Stacy S. McGaugh, "MOND Prediction for the Velocity Dispersion of the `Feeble Giant' Crater II" (November 3, 2016).

This is a big deal because under a wide range of dark matter hypotheses, the velocity dispersion could have been no lower than 5 km/s and was expected to be more like 11 km/s to 24 km/s.

The actual velocity dispersion of Crater II was measured with the latest and greatest telescopes in a result first announced six and a half weeks after this prediction was made on December 19, 2016. What did they find?

A velocity dispersion of 2.4 km/s to 3.0 km/s.