That rotation curves become flat at large radii is one of the most famous results in extragalactic astronomy. This had been established by Vera Rubin and her collaborators by the late 1970s. There were a few earlier anecdotal cases to this effect, but these seemed like mild curiosities until Rubin showed that the same thing was true over and over again for a hundred spiral galaxies. Flat rotation curves took on the air of a de facto natural law and precipitated the modern dark matter paradigm.
Optical and radio data
Rotation curves shouldn’t be flat. If what we saw was what we got, the rotation curve would reach a peak within the light distribution and decline further out. Perhaps an illustration is in order:
An obvious question is how far out rotation curves remain flat. In the rotation curves traced with optical observations by Rubin et al., the discrepancy was clear but modest – typically a factor of two in mass. It was possible to imagine that the mass-to-light ratios of stars increased with radius in a systematic way, bending the red line above to match the data out to the edge of the stars. This seemed unlikely, but neither did it seem like a huge ask.
Once one gets to the edge of the stellar distribution, most of the mass has been encompassed, and the rotation curve really should start to decline. Increasing the mass-to-light ratio of the stars ceases to be an option once we run out of stars*. Fortunately, the atomic gas typically extends to larger radii, so provides a tracer further out. Albert Bosma pursued this until there were again enough examples to establish that yes, flat rotation curves were the rule. They extended much further out, well beyond where the mass of the observed stars and gas could explain the data.
How much further out? It depends on the galaxy. A convenient metric is the scale length of the disk, which is a measure of the extent of the light distribution. Some galaxies are bigger than others. The peak of the contribution of the stars to the rotation curve occurs around 2.2 scale lengths. The rotation curve of NGC 6946 extends to about 7 scale lengths, far enough to make the discrepancy clear. For a long time, the record holder was NGC 2403, with a rotation curve that remains flat for 20 scale lengths.
Twenty scale lengths is a long way out. It is observations like this that demanded dark matter halos that are much larger than the galaxies they contain. They also posed a puzzle, since we were still nowhere near finding the edge of the mass distribution. Rotation curves seemed to persist in being flat indefinitely.
Results from gravitational lensing
Weak gravitational lensing provides a statistical technique to probe the gravitational potential of galaxies. Brouwer et al. did pioneering work with data from the KiDS survey, and found that the radial acceleration relation extended to much lower accelerations than probed by the types of kinematic data discussed above. That implies that rotation curves remain flat way far out. How far?
Postdoc Tobias Mistele worked out an elegant technique to improve the analysis of lensing data. His analysis corroborates the findings of Brouwer et al. It also provides the opportunity to push further out.
Weak gravitational lensing is a subtle effect – so subtle that one must coadd thousands of galaxies to get a signal. Beyond that, the limiting effect on the result is how isolated the galaxies are. Lensing is sensitive to all mass; if you go far enough out you start to run into other galaxies whose mass contributes to the signal. So one key is to identify isolated galaxies, and restrict the sample to them. KiDS is large enough to do this. Indeed, Mistele was able to show that while neighbors+ were a definite concern for elliptical galaxies, they were much less of a problem for spirals. Consequently, we can trace the implied rotation curve way far out.
How far out? In a new paper, Mistele shows that rotation curves continue way far out. Way way way far out. I mean, damn.
Optical rotation curves typically extend to the edge of the stellar disk. That’s about 8 kpc in the example of NGC 6946 given above. Radio observations of the atomic gas of that galaxy extend to 17 kpc. That fits within the first two tick marks on the graph with the lensing rotation curve.
UGC 6614 is a massive galaxy with a very extended low surface brightness disk. Its rotation curve is traced by radio data to over 60 kpc. It is one of the most extended individual rotation curves known. The statistical lensing data push this out by a factor of ten, and more, with no end in sight. The flat rotation curves found by Rubin and Bosma and everyone else appear to persist indefinitely.
So what does it mean? First, flat rotation curves really are a law of nature, in the same sense of Kepler’s laws of planetary motion. Galaxies don’t obey those planetary rules, they have their own set of rules. This is what nature does.
In terms of dark matter halos, the extent of isolated galaxy rotation curves is surprisingly large. Just as we come to the edge of the stellar disk, then the gas disk, we should eventually hit the edge of the dark matter halo. In principle we can imagine this to be arbitrarily large, but in practice there are other galaxies in the universe so this cannot go one forever.
In the context of LCDM, we now have a pretty good idea of how extended halos should be from abundance matching. A galaxy of the mass of UGC 6614 should live in a halo with a virial radius of about 300 kpc or less. There is some uncertainty in this, of course, but we really should have hit the edge with the lensing data. There should be some sign of it, but we see none.
One complication is the so-called 2-halo term. In addition to the primary dark matter halo that hosts a galaxy, when you get very far out, you run into other halos. Isolated galaxies are selected to avoid this to the extent possible, but eventually there will be some extra mass that causes extra lensing signal that would cause an overestimate of the rotation speed. I’ll forgo a detailed discussion of this for now (see Mistele et al. if you’re eager), but the bottom line is that it would require some unnatural fine-tuning for the 1+2 halo terms to add up to such flat rotation curves. There ought to be a perceptible feature in the transition from the primary halo to the surrounding environment. We don’t see that.
In the context of MOND, a flat rotation curve that persists indefinitely is completely natural. That’s what an isolated galaxy should do. Even in MOND there should be an environmental effect: the mass of everything else in the universe should impose an external field effect that eventually limits the extent of the rotation curve. How this transition happens depends on the density of other galaxies; by selecting isolated galaxies this effect is put off as much as possible. Hopefully it will be detected as the data improve from projects like Euclid.
The primary prediction of MOND is an indefinitely extended rotation curve; the external field effect is a subtle detail. Yet again, that is what we see: MOND gets it right without really trying, and in a way that makes little sense in terms of dark matter. Sometimes I wish MOND had never been invented so we could claim to have discovered something profoundly new, or at least discuss the empirical result without concern that the data would get confused with the theory. MOND predictions keep being corroborated, yet the community persists in ignoring its implications, even in terms of dark matter. It’s gotta be telling us something.
We have a press release about this result, so perhaps you will see it kicking around your news feed.
*We could, of course, invoke dark stars, but that’s just an invisible horse of a different color.
+There is a well known correlation between morphology and density such that elliptical galaxies tend to live in the densest environments. This means that they are more likely to have neighbors that interfere with the lensing measurement, so finding that identifying isolated ellipticals with a clean lensing signal is more challenging that finding isolated spirals comes as no surprise. Isolated ellipticals do exist so it is possible, but one has to be very restrictive with the sample.
“yet the community persists in ignoring its implications” The typical signature of cultist behavior.
They’re too much ideological and economic interests behind the “dark matter” thing to be discarded easily.
That they’re new, irreducible natural laws at galaxy complexity level is obviously a direct challenge to the ever present reductionist mindset in mainstream scientific thinking, that is even more unacceptable to particle physicists naive reductionism, their WIMPs “hypothesis” can’t be more weak.
Particle physicists assume that they are the only ones making “fundamental” research and only laws found at that level are fundamental, but as P. A. Anderson already said: “at
each level of complexity entirely new
properties appear, and the understand-
ing of the new behaviors requires re-
search which I think is as fundamental
in its nature as any other”.
Wow!
(As in better than that wow)
Maybe you should just ignore dark matter. There’s a lot more justification for that than ignoring MOND. MOND at least demonstrates that the problem lies with our gravitational model(s).
MOND doesn’t fix the problem though, it just patches the old models and it shares with them a central flaw – it only describes the gravitational effect but does not describe the causal physical mechanism. All physical effects have to have physical causes, don’t they?
“All physical effects have to have physical causes, don’t they?” Not necessarily if that “effect ” is a new fundamental, irreducible property. Like an independent axiom that is irreducible from other axioms.
Reductionism is intrinsically limited by assuming that almost anything should have “physical causes”, the reality of new irreducible properties/behaviors is ignored by this approach.
Sorry, but I don’t understand the question.
For years after Newton proposed his theory of gravitation and motion, many scientists could not accept it because it included (what we would now call) “action at a distance.” Like you, apparently, these scientists demanded a “cause” — something physically in contact with the planets, pushing them along in their orbits. (Angels maybe?)
Newton’s theory predicts elliptical orbits for the planets. That is all the “cause” you get and all you need. And no, the “cause” of the observed motions is not simply the force from the Sun. There are many possible non-Newtonian theories that include a central force but do not predict elliptical orbits (e.g. post-Newtonian theories). And many theories that do not include central forces as Newton understood them but which do predict elliptical orbits (e.g. Einstein’s!).
Likewise, Milgrom’s theory predicts flat rotation curves for galaxies. That’s your “cause” and that’s all you’re ever going to get.
As a summary of the mathematicist philosophy that’s pretty good. It gets right to the heart of the problem that has turned modern theoretical physics into a wasteland of mathematical fantasies. The two “standard models” of academic physics do not bear any resemblance to empirical reality – because they are full of the modern equivalent of angels and devils – dark matter, dark energy, quarks, gluons, etc..
So we don’t know the cause of the gravitational effect and therefore we can’t know the cause and anyway some math is the cause? That’s mathematicism or, if you like, instrumentalism alright, but those are philosophies not science.
Science is supposed to be the study of physical reality which excludes angels and devils and should exclude their modern undetectable variants because those things (ancient and modern) are not present in physical reality. Scientists do want to know how gravity works, mathematicists don’t care.
BTW, you do know that Newton himself was dissatisfied with the “action at a distance” account of gravity, don’t you? That’s what makes him still a great scientist. He didn’t hide behind his math. He was evaluating his model in terms of physical reality and found it wanting. He understood that his model’s failure to explain the cause of the gravitational effect was a shortcoming.
No such thoughts intrude upon the mathematicist belief system of course. And so the ongoing crisis in physics will continue until mathematicism is returned to the philosophy and mathematics departments for a full refund and the scientists, like Dr McGaugh, who actually study physical reality are put in charge of theoretical physics, not as it is now, the other way around.
Modern theoretical physics has devolved over the last 40+ years into the study, not of physical reality, but of a pair of “chosen models” which are treated as unassailable truths that can be tinkered with at the margins but whose central axioms and conceits cannot be questioned.
Forty years have been wasted chasing mathematical fantasies and all we have to show for it are two absurd standard models in which empirical reality is only described in terms of entities and events that make no appearance in physical reality. That’s forty years too long. Math is not physics.
It is fitting to be content with Merritt’s answer. The universe’s laws are very complicated. While locality in a set of laws can give certainty and confidence of having formulated proper laws, it is way too easy to theorize a local theory above a not yet well understood mathematical action at a distance – it’s something people can understand. Flogiston is a good example: some substance burns in whatever is burning. The mathematics of oxidation were instead way too abstract in the beginning, before chemical reactions began to be widely studied.
But I agree that things have to work in some way.
It didn’t matter that Newton didn’t have a cause – the basic mathematics was plenty at the time. And it went on like that, mathematics but often no picture. In the last century theories got so weird that some argued no picture existed. But in other areas, as we learn more, we find causes. Now the data has got so good that to make real progress this century, we need the underlying picture Einstein and Wheeler both said would be found in the future (several quotes from each of them on it). If there were courses on conceptual physics, we’d probably have found it by now, but as you say, mathematicism is everywhere.
I’ve found a way to shorten the evidence for one possibility, helical path refraction, down to a few lines. I hope this ultra-compressed version is of interest. Talking of causes, this would be one. (from Kerr, 2023)
———————————
In well known refraction, a light beam in a graded or layered medium has all points on its path linked via Snell’s law. For two points any distance apart, if 4 numbers are put into Snell’s law (two angles to the normal, two local speeds for light), the sides of the equation will always agree, and their agreement shows refraction is at work.
sin (theta1) / sin (theta2) = S1/S2
This can also be applied to orbits, with a slight variation on the same equation. It is applied to matter, assuming it is like light at a very small scale. The 4 terms are replaced by terms for the angle to the normal, and the local speed of light:
sin (arccos [v1/c]) / sin (arccos [v2/c]) = (1 – [2GM/r1c2])1/2 / (1 – [2GM/r2c2])1/2
Numbers are put in for any orbit: open, closed or radial, around any spherical mass M. v1 and v2 are speeds at radii r1 and r1. The two sides always agree, to ~ 16 decimal places. Again, their agreement shows refraction is at work.
Only on radial paths is this a near-proof, as there’s no need to bring in the dimensions, only a graded medium surrounding the mass, and matter on helical paths, so it’s straightforward. Any comment, particularly from Stacy, would be appreciated.
I think that is indeed a good approach – but it begs the question, what is the physical nature of the graded medium?
Space has mass of it’s own. And is distorted (“warped” “concentrated”) by matter.
It’s very small-scale waves in space itself. At a scale like the Planck scale, a mass, say a planet, is a lot of rotating disturbances in the circular dimensions. The mass ultimately consists only of vibration, and it gives off weaker vibrations that dissipate in the radial direction. That makes a graded refractive medium surrounding the mass, which moves away, thinning out, so the field keeps its shape, and from a distance could look like a solid object. The transmission speed of space increases radially, but is constant at any point in the field. Matter nearby, being like rotating light, gets refracted along helical paths. At large scales an excess of the medium builds up, and is what is taken to be DM.
In a different situation, the main drawback of PSG would be one has to assume this invisible medium exists – in a gravity theory it can seem ‘excess baggage’. But because of the present need for DM, similar assumptions are widely being made anyway, and it might be said that the medium in PSG has ‘more reason to be there’ than DM.
Indeed it is enough in order to be content, at the current state of potency of scientific analysis.
A graviton particle would be great, however the details of the MOND law cannot really be deduced from how much gravitons and so on (it would anyhow on first-order perturbations be a inverse square law). It’s a beautiful mess. We’ll have to settle on being content with ‘spooky action at a distance’ like Newton did for now. It’s clear that MOND is what the data tell, figuring out how (how and to what extent locality is preserved) will be up to the coming generations.
Huh, well I guess this is just a matter of taste. But I don’t think of Mond as a theory, but more as some sort of observational fit to the data. Kinda like Kepler’s laws, or the ideal gas law. And I do hope some day that we will find the underlying ‘thing’ that explains it. Though this may just be a pipe dream and may never come true. Still I find it fun to think about….
OK, let’s look at this from a dark matter viewpoint and see if we can prove a falsehood. The original idea was that each galaxy was surrounded by a (spherical?) dark matter halo, which might extend to, say, 20 scale lengths at most. Now we have a requirement for a ‘halo’ that extends for perhaps 100 or more scale lengths. So, we are looking at a ‘dark matter’ mass that might exceed 5 cubed times the previous estimate (say a factor of 100). At what size of halo does the total ‘dark matter’ imputed around galaxies exceed the total amount of ‘dark matter’ allowed in the universe in the ɅCDM model?
A flat RC implies an enclosed mass that increases linearly without limit – hence one reason why it can’t persist forever. But the community wrapped its head around that a long time ago, so as usual, good luck proving a negative.
It took thousands of years to recognize that the parallel axiom in Euclidean geometry was irreducible from the other axioms leading to the discovery of non Euclidean geometries by Gauss, Bolyai and Lobachevsky. It’s intrinsic human nature to look for underlying reasons, for “explanations”, but sometimes they’re not hidden reasons or hidden variables (remember Einstein and quantum mechanics), nature can’t care less about humans wishful thinking and if mathematics is any guide irreducibility is the norm not the exception, “understanding” is overrated, many times knowing is the best that we can do. Hopefully we don’t have to wait thousands of years for the “community” to acknowledge the presence of irreducible properties at any level of nature complexity hierarchy, including cosmic hierarchies.
A couple more questions please.
(1) How actually does one determine the rotation velocity of (i) the orbiting stars, (ii) the gas beyond the stars?
(2) Spiral arms are often less tightly wound towards the edge of the disc. In the case of NGC 6614, the extreme ends of the two major arms are so far unwound that the stars associated with them – and any gaseous extrapolation of them – cannot be rotating round the nucleus at all; the arms are at too much of a tangent. Is all this taken into account, and could you perhaps adumbrate?
I wasn’t meaning to go into the question of ‘density waves’, whose significance for spiral arm formation is contested (Sellwood & Masters 2022), and I appreciate that stars do not entirely align with the arms. But surely that is where most stars occur. Stars are seen in the optical, and the most obvious feature of galaxies beyond the MW are the optically visible arms. I was asking about the trajectory of stars in the arms near the disc’s optical edge, where the arms, composed of stars and gas, drift tangentially away from the nucleus and cease to rotate around it. How is one to understand the ‘rotation’ curves of stars and gas from that point on?
Yes, spiral arms are a concentration of stars. That’s all I meant by the analogy to waves, I was not specifically advocating nor requiring that spiral arms be density waves.
No, stars and gas do not cease to rotate around the nucleus near the optical edge. I don’t understand where you get that idea.
I was trying to avoid making general assertions, my query being illustrated by the specific instance of NGC 6614, the galaxy in your figure and a larger image of which I linked to in case my description was not clear enough, as it appears not to have been.
Let me try another tack. Armed galaxies are for the most part spiraliform rather than a series of concentric circles, so the distance from r = 0 to any point along the arm will always be greater than the radial distance to that point. In some cases arms expand so much that at their extremes they seem almost to escape the gravitational field (e.g. NGC 6614, 2336, 2442, 6872). If there is any relation at all between stellar orbits and the arms, then the orbital velocities will be different than if the orbits were circular. Since it is not possible to image the motions of individual stars in faraway galaxies, are rotation curves based on the assumption that orbits are in fact circular?
Regarding very faraway galaxies, I note that according to one study (Genzel, Nature 2017) the rotation curves of high-redshift discs show an ultimate decrease in rotation velocity, ascribed partly to less DM around the outer disc and partly to greater velocity dispersion. ‘Our analysis leaves little space for dark matter in the outer disks (and inner halos) of massive, high-redshift star-forming galaxies’ – but by the same token suggests that MOND is not universal? This evolutionary aspect must have been researched a good deal more in the 7 years since, but as the study has attracted >300 citations, I’m a little loath to follow up.
Orbits are of course not perfectly circular, but the deviations therefrom are measured to be small in spiral galaxies. There is an effect called asymmetric drift, in which the eccentricities of stellar orbits grow over time – this is observed in detail in the Milky Way. We know how to correct for it in other galaxies; in most cases it is a small effect.
I discussed the claim of declining rotation curves that you cite in https://tritonstation.com/2017/03/19/declining-rotation-curves-at-high-redshift/ The short answer is that this claim is overblown. For all the metrics that I can measure in the same way for both samples, there is no apparent difference between local galaxies and those at high redshift.
If we had a single galaxy in a contracting universe, would we expect to see flat rotation curves? Once the contraction was more significant than the galaxy’s gravity, wouldn’t the rotation curve flatten?
That is a hypothetical I hadn’t considered. The universe being much much larger than a single galaxy, it shouldn’t matter to a galaxy’s rotation whether it is expanding or contracting. It the contraction gets to the point where it is more significant than the galaxy’s gravity, then we are close to a Big Crunch and have bigger problems than the shapes of rotation curves. I see no reason why this situation would lead to flat rotation curves.
If we take it that galactic mass is concentrated enough to be considered spherically distributed when we’re at the visible outskirts of the galaxy, then the flat rotation curves far, far away from the visible edge suggest that this is not due to the galactic mass, rather something cosmological and universal is going on. All that the galaxy does is provide a center of rotation. Of course, it could be MOND. But it is fun (and hard) to think of a plausible alternative mechanism.
If the analysis is using weak gravitational lensing (an effect of general relativity), how does it fit together with MOND? If lensing is used in the analysis, do we already assume something about gravity that would prevent us from making conclusions about whether or not gravity needs to be modified into the direction MOND?
Thx for clearing up my confusion 🙂
Good question. We basically assume that lensing gets the same boost as kinematics. That was a hang-up in writing a GR extension of MOND for a long time. TeVes (Bekenstein 2004) was the first theory that could do this. TeVeS wasn’t the final theory, but showed how it can be done. I consider this to be a requirement for theory, either as an extension of GR (lensing and kinematics must both be boosted in the low acceleration regime as in MOND) or for a dark matter theory (one has to construct a halo that knows about the baryon distribution and continues to large radii indefinitely with the density falling as 1/r^2).
I would imagine that MOND’s “why?” is just as interesting a question as what kind of particle could comprise dark matter. At the very least, it could be simply that the prior assumptions about the nature of spacetime are incomplete.
Indeed. Why does MOND happen? That is does happen is well established. How it can possibly be so is the head-scratcher. That’s what makes it such an important question.
I suppose such observation is also consistent with the minimal acceleration from quantized inertia?
Anything that reproduces MOND will also get this right. I have seen versions of quantized inertia that come close, but none that are quite right. Usually the issue is around a0; the behavior converges to MOND at low accelerations so is probably OK here.
Wow, this is great. Congrats to all the authors. So two questions (for which I’m too lazy to try and look up the answer for myself.) Your previous post on weak lensing data (from Brouwer et al) extended the acceleration range down to about 10^-12.5 m/s^2. What’s the new range? And second, there is some length/ acceleration scale for which Mond fails… galactic clusters. What is that acceleration? And is there any hope to get to that range in the lensing data? (Three questions; Our three weapons are fear, surprise, and ruthless efficiency, and fanatical devotion to the pope… :^)
Thanks.
The range of accelerations is the same, but the credible portion is larger. How much larger remains a matter of judgement. 10^-13, perhaps.
MOND fails in clusters because their accelerations are too high by a factor of ~2. This happens in the vicinity of a0; see https://arxiv.org/abs/2303.10175 I hope to write a post about all these things soon. Yes. Soon.
Tobias is looking into what can be done with lensing for clusters. Not clear yet what we can add, so definite maybe.
Re clusters: Ahh thanks, I was confused. Clusters have the same acceleration range a_0, it’s just that the effect is stronger… or that we are missing some of the mass. https://tritonstation.com/2024/02/06/clusters-of-galaxies-ruin-everything/
yes. Just so.
The ‘distinct RAR’ in clusters paper https://arxiv.org/abs/2402.12016 showed the large-scale pattern with more certainty – similarites and differences to individual galaxies. Is it possible that the equivalent of a0 in clusters is around 2e-9?
In galaxies, the transition (if one takes it that way) often starts around the same place, 2e-9, and ends at a0. Is it a valid approach to say both transitions may start at 2e-9, but in clusters it’s quick, while in galaxies it’s gradual?
Yes, insofar as one ends up with two parallel sequences like that. There is a lot more scatter in the cluster data, so one might also say the relation gets fuzzy at that point, with a band of higher acceleration points. Any way you look at it, it is hard to explain in any paradigm.
If there’s a difference to the transition, looking for reasons, it seems it can’t be the flat shape of galaxies, as MOND is found in ellipticals as well. Could it be to do with the distribution of matter in relation to the boundary – in clusters, I guess the boundary lands where there’s empty space. But in galaxies it lands where there are a lot of individual stars, so fields from near and far combine, with different strengths. So one possibility is that some kind of average is going down during the transition. Is that possible, or are there types of galaxies that have MOND, but which don’t fit with that, such as LSBs?
There are funny things going on in the outer regions of clusters that are reasonably attributed to hydrostatic nonequilibrium. But I don’t see that as likely to fix everything for MOND. The deficit MOND faces in clusters occurs mostly in the inner regions, as if there is a core of dark material lurking there. This is also an issue for LCDM, where the baryon fraction increases with radius and doesn’t reach the cosmic value until way far out (around the virial radius or beyond). This is easier to excuse for LCDM where we already have dark matter, but it doesn’t make a whole lot of sense. This is another way in which clusters ruin everything https://tritonstation.com/2024/02/06/clusters-of-galaxies-ruin-everything/.
Does this concentration of apparent DM in the core provide a strong constraint on the properties of hypothetical warm DM as in hybrid models like νHDM ?
Yes, the concentration of DM in the cores of clusters is telling us something, but I wouldn’t say it is a strong constraint. It mostly says that whatever it is sank to the middle and squats there. Some theory papers suggest that may be a reasonable outcome for a big ball of neutrinos or sterile neutrinos provided they are massive enough. But I expect other things are possible.
This is more accurate: Fig. 2, left panel, of the ‘distinct RAR’ paper. I don’t know if it seems the same to you, but to me this is about the best clue, or set of clues, we have. The details of the transition may give away something about what’s going on in the outer regions of gravity fields. The differences and similarities between cluster scale and galaxy scale RAR, with a different acceleration scale, make MOND look more like an effect, less like a gravity theory. And what looks like an important point is that both transitions start at the same place, gbar ≈ 1e-8.8. But from there, for some reason the two versions are different: one transition goes down to 2e-9, the other to a0. That doesn’t look ‘irreducible’, I’d say there’s something going on.
it is perhaps not coincidental that the offset for clusters is shared by galaxies at the centers of clusters.
I know the data for clusters is a lot less clear, but assuming an acceleration scale of 2e-9, BCGs have the same offset between that and a0 as clusters do? If so, that doesn’t fit with my suggestion about it being to do with where the boundary lands. Not if the boundary in BCGs, which are generally giant ellipticals, lands among stars, as it probably does. Instead it starts to look related to the shape of the mass – it seems to me whatever’s going on, the clue is that there’s one pattern that kicks in right away, and another that has a gradual transition from the same point. Anyway, thank you.
Isn’t this weak lensing in contradiction with the more robust GAIA astrometry? (at least for the milky way)
https://arxiv.org/abs/2309.00048
IF the Milky Way rotation curve declines in a Keplerian fashion, then the Milky Way is different from every other galaxy in the universe. So yes, this aspect of the Gaia result you cite is in contradiction with more robust data of many varieties, including the lensing reported here.
It’s not that the Gaia data aren’t robust, it’s that they are much more involved to interpret. For starters, it is not correct to say there is a contradiction between our result and Gaia; there many aspects of the Gaia data that conform well to the radial acceleration relation. It is only the portion of the Gaia data at large radius where there are few stars where there is a potential discrepancy. I’ve discussed this extensively in a series of posts: https://rogue-scholar.org/posts/10.59350/yj9c7-kvs82 | https://rogue-scholar.org/posts/10.59350/nr1qw-99t29 | https://rogue-scholar.org/posts/10.59350/ndyxv-dng39 | https://rogue-scholar.org/posts/10.59350/1hskp-b5a62
We’ve reached the point with Gaia data where we have to worry about all the cross terms in the Jeans equation. The apparent problem arises where these go whacky – see Fig. 3 of https://arxiv.org/abs/2405.19028. So I suspect the apparent discrepancy will resolve when all this is taken into account. It is on my to-do list, but it will take some time to get to.
Hi @Tritonstation, thank you for this excellent answer, the paper you linked seems highly relevant and I will read those.
We are at a key turning point in the history of cosmology as we are entering the “decade of the surveys”
We should find a definite answer to this and other fundamental questions in the next 1-3 years, based on GAIA DR4. (including the observation of a breakdown of classical gravity in wide binaries)
However, few seem to know that apparently, it is possible to match GAIA astrometry accuracy with wide field spectrographs. I am a layman in how astrometry works but IMO a key element to constraint proper motions, beside the GAIA astro instrument, is to accuratelely determine the radial velocity, which is why GAIA comes with a spectrograph, the GVS.
The issues are that:
The GVS has suffered damage at the start of the mission, reducing its accuracy.
Secondly, despite being in space, the GVS can be quantitatively and qualitatively outperformed by ground based state of the art spectrographs.
Assuming the galactic halo and rotation curve can be better constrained by better radial velocities measurements (therefore not just via proper motion via the astro instrument), then it follows that we should already be able to perform a better analysis than https://arxiv.org/abs/2309.00048 and than https://arxiv.org/abs/2405.19028 because both only used GAIA (DR3) data.
Such an opportunity seems to have been currently overlooked and IMO should be your top priority as if your research group does not do it, I am afraid we will have to wait multiple years for someone else to leverage the better radial velocities datasets.
The currently (partially) released best survey for radial velocities is probably H3
https://arxiv.org/abs/1907.07684
https://arxiv.org/abs/2204.02989
H3 has high resolution but is however less wide than the upcomind sky surveys.
Regarding the upcoming surveys that might have their first releases in 1-2 years
the WEAVE wide spectroscopic survey seems to claims parity and in many aspects superiority and complementarity to GAIA
https://arxiv.org/abs/2212.03981
See figure 5
Moreover combined their astrometrics results should be much more accurate
The other best spectroscopic survey to come is 4MOST https://arxiv.org/abs/1903.02464
https://arxiv.org/abs/1903.02467
Both surveys aims an accuracy for radial velocities of 1-2 km/s which is far better than GAIA GVS? (wiki says the pre accident performance of GVS being Radial velocities are measured with a precision between 1 km/s (V=11.5) and 30 km/s (V=17.5). Don’t know if those are two distincts modes or if GAIA truly has 30km/s of uncertainty…
Another survey with close performance is the DESI milky way survey
https://arxiv.org/abs/2208.08514 (that has already started and since DESI DR1 is released, DESI milky way DR1 should already be available to insiders? if so they might already be able to rule out dark matter definitively, right now)
Another survey (smaller) is taipan https://arxiv.org/abs/1706.01246
Also I only mentioned radial velocities from spectrographs but maybe their state of the art measurements of other stellar parameters, like mass and chemical abundances, might help constraint the halo (so called 6D space)
Moreover the often forgotten, gold standard for stars parameters (including rotation and maybe velocities?) is asteroseismology, which is performed by other instruments, like Kepler, TESS and this field will soon be revolutionized by PLATO hence PLATO or the already available TESS might help constaint dark matter too.
TL;DR you might be able to definitely rule out dark matter (or MOND) by coupling the aforementioned GAIA DR3 analysises with better radial velocities already available via the H3 survey and Taipan survey and very soon available (publicly) and maybe already available privately next gen incoming surveys planned for the next 1-3 years, especially DESI MWS, WEAVES and 4MOST.
Conversely, the weak lensing analysis from your posts should be able to have order of magnitude more data via surveys incoming very soon, maybe the already released DESI DR1, Erosita DR1 and most certainly, the euclid first data release THIS DECEMBER.
I pray that cosmologists like you will investigate those key questions that might get definite answers solvable before the end of 2024.
Yes.
There are a lot of great surveys going on, and some provide complementary information to Gaia. I’ve already made use of some of that, from APOGEE. I look forward to the data continuing to improve, but caution that no part of this analysis is easy: it will take some time to digest all these results.
We’re certainly looking forward to Euclid results for gravitational lensing as well. How quickly the first data release becomes useful depends on what they release. So again, we’ll have to wait and see.
“This is another way in which clusters ruin everything”
It actually makes sense if we just acknowledge that clusters of galaxies are a higher, different complexity hierarchy than galaxies. This always happens when you try to apply a given theory to a higher complexity level. Reductionist mindsets are blind to this obvious fact.
Classical Reality -> cluster of quantum objects
Trees -> cluster of cells
Forest -> cluster of trees
Galaxy -> cluster of stars
…
The ideas, notions used in one hierarchical level can’t be used unchanged to the next higher or lower hierarchical level.
Not by chance the notions, laws used to describe cells can’t be used to trees, forest or even less to planet ecological system level. Not by chance Newtonian/Einstein dynamics fails at galaxy complexity level and MOND fails at galaxy clusters level, or wide binary systems will not provide a definitive answer to MOND galaxy level dynamics, they are at different hierarchical levels, or even less likely to expect consistent results when using General Relativity at “universe” level complexity when it fails already at galaxy level complexity.
The growing list of contradictions generated by the the ad hoc introduction of dark matter(trying to save general relativity) is a textbook example of what happens when you try to extend a theory range of applicability to a higher complexity level, “epicycles” all over again.
Reality hierarchical structure is a direct manifestation of the emergence of new “fundamental” properties at every level, these new fundamental properties can only be discovered by direct observations/experiments. Reductionist mindsets have to deal with this sooner or later, or keep juggling with their “epicycles”.
Yes because the universe surely gives an ef about categories and their definitions some ape variant came up with. That a tree is something different than a forest is a matter of perception. No, observers are irrelevant. I hope we can skip the case of conscious observers (w/e ‘conscious’ even means). If observation oth is actually interaction (wth is then ‘observation’ still used and it gets morphed into ‘interaction’ only when speakers are pressed?) don’t we claim there’s no empty space per se? Like virtual particles and vacuum energy and all that. So how can a particle be ever unobserved? And if it’s interacting constantly then its wave function is in constant collapse, is it not? And how can we dream something that’s constantly collapsing, a process we know fuck all about, describes anything? I mean, if wave functions of everything are in constant collapse, does wave function even exist? Beyond some useless abstraction like … oh I don’t know, love?
Anyway, what happens when tree falls and gets transformed to something of a different caste? A lawyer is dispatched to inform it on new regime? Higher-up(s) wave(s) magic wand(s) and translate(s) laws that governed the tree to those governing whatever category you’ve reassigned that biomass to?
Vast meadow, rolling grassland for days. I plant two anemophilous trees, a female and a male. Because I’m obscenely rich and don’t have to work I do this 100 times all over the world. I set up a foundation that will look over all the pairs. No additional trees are to be planted near by. Minor interventions only, like watering in severe drought and such. Now, it is almost certain that at some point in considerable future at least one of those pairs will grow into something pretty much all will be able to call a forest. Phase transition?
As for energy of basically infinite density spread mushing about out there; do this. You (pl) take frame rates and time scale at which we’ve properly observed all that. Now set up multiple instances of the following experiment. A body of water, a stone (a random stone) suspended above at uniform height, on a string with a mechanism that can release the stone via remote command. At one site you put a camera into the water, at all others you place camera at varying distance (some will be on hilltops or in orbit). All cameras are obviously remotely operated and tied to same switching mechanism as releases. Even though experiments are at various altitudes you can figure out fairly well when stones are to hit the surface. Release the stones and start the cameras just prior to expected earliest splash. Cameras operate at capture rate which you’ve calculated by translating our observation of the universe in its frame of reference to regular down-to-Earth reference. Review as much footage as is the duration of our observation of the universe in relation to universe’s time scale. What can you tell me about fluid dynamics?
Everything is a matter of perception.
But inconsistencies always appear when you try to use your theories/notions at a different hierarchical level to the one where your theories/notions appear to be consistent with empirical evidence, and there’s no exception to this rule. The black matter blunder is just an small example of this.
Reality hierarchical structure is an objective fact, but once again reductionist mindsets are blind to this obvious fact, they keep generating inconsistencies(epicycles) by forcing their theories beyond their “natural” hierarchical level.
When science gets stuck, the puzzle in front of us is always initially a conceptual one. If we solve it we can get to the mathematics, sometimes we get some of that anyway, but the initial challenge is about what’s going on underneath what we observe. People then find all kinds of reasons for not confronting the conceptual puzzle in front of them. First they put in epicycles, as you say. They complicate their earlier picture, to cover new facts. What’s needed is a simplifying conceptual framework (elliptical orbits was a good one). Or they say, as people did a lot in the 20th century, that only mathematics and experimental results count. ‘Ideas’ don’t matter, so the vital conceptual puzzles (like interpreting QM) get left to the philosophers, and booted out of the field.
And to me assuming an irreducibility, before we even know very much, looks like it risks being another excuse for not taking on the conceptual puzzles. As it happens the two RARs in galaxies and clusters, a very new set of clues to add to a new puzzle, looks like a good example of a crossover between two scales. They’re similar enough to suggest that they can be described on the same basis, when we’ve learned more.
Let me point out that “scale”(related to distances)is not the same as complexity/hierarchical level, you can have very simple systems covering large scales while at the same time you can have complex/higher hierarchical levels at a lower distance scale.
It seems that theories always have a limited range of applicability, and that range of applicability is its “natural” hierarchical level.
Quantum mechanics limited to simple quantum systems.
General relativity limited to simple gravitational systems.
When you go beyond a theory natural hierarchical level you always will find inconsistencies because the theory assumptions never took in consideration the new irreducible properties presented in the new hierarchical level.
But physicists love their theories so much that they will postulate the “reality” of universe wide filling substances/fields, like when they postulated the reality of a universe wide filling luminiferous aether to justify the transmission of light in empty space, or now they postulate the reality of dark matter, or dark energy just to keep the current framework free of obvious contradictions.
Using a theory to predict the reality of a particle or phenomenon at their natural hierarchical level is what you expect, but then using it to “predict”/postulate a universe wide filling substance/field is really going beyond its natural predictive/explanatory power.
Any theory always have a limited predictive/explanatory power, and that is what naive reductionism always fails to acknowledge.
There’s no theory of everything precisely because Reality has a hierchichal structure.
I see what you’re saying, and I agree in some ways. Not because all theories are necessarily limited to a particular domain, but because people often apply theories beyond where they can realistically go, if you allow for unknown forces. We’re not very good at allowing for unknowns – assuming matter becomes infinitely compressed in a black hole is a good example. That’s applying GR well beyond its range – infinity is a long way away, and unknown forces will come into play before you get there.
But I don’t see why you assume limitations to theories must inevitably exist – when you put all the pieces together, including ones we haven’t yet found, we might be able to allow for the hierarchical structure, and connect things up anyway.
When I say that any theory always have a limited predictive/explanatory power I am referring to the fact that complexity always is a boundary to the predictive/explanatory power of any theory, large assemblies of discreet objects tend to present clustering or structures, like large assemblies of stars in galaxies or large assemblies of quantum objects in living beings, theories used to describe simple assemblies of discreet objects will fail on these large assemblies.
There are even results in formal mathematics showing that complexity is a source of incompleteness/irreducibility and that is fully in sync with the ideas first clearly exposed by PW Anderson in “More is different”.
It’s fine to say perhaps theories are always limited in this way. I think you’ve got to be careful if you start making statements about things that physics will never find in the next, say 5000 years. They did that in the early 20th century when they said c was a speed limit for everything. They should have said just light and matter, which was everything at the time anyway. But instead they implied that over the next 5000 years, nothing will be found that travels faster – they were not in a position to say that. And within 100 years, entanglement is more or less certain, three separate teams did loophole-free experiments in 2015. (In my picture properties of the dimensions allow that, which is not light or matter – some were loosely thought to exist already, but they’re irrelevant unless you think matter is vibrations in the dimensions.) So it’s best to say only what we’re in a position to say, and leave room for unknowns in the future.
But can the Milky Way just be ‘different’ because it is not isolated? The flat rotation curves to way far out apply only to isolated galaxies, right?
Right – the Milky Way would flunk our isolation criterion. Most galaxies have neighbors; only 16% of the KiDS sample is isolated by the criterion we apply, which is basically that there is no other galaxy > 10% of the stellar mass of the primary within 4 Mpc.
The claimed Keplerian downturn in the Milky Way happens around 20 kpc, so we don’t need to appeal to the lensing data at all to say it would differ from other galaxies. The existing kinematic data for other spiral galaxies frequently trace out many tens of kpc; that already defines a norm that the Milky Way obeys until suddenly it seemingly doesn’t over a rather narrow range of radii. There are also measurements in the Milky Way itself – stars in the halo – that contradict the Gaia data. The issue isn’t whether the Gaia data are better – they are – but the assumptions that go into analyzing the Gaia data. Those are a more severe limiting factor and are what Koop et al. show is an issue at the relevant radii.
It used to be that neutrinoes “have no mass”, but now we know they do in fact have mass.
We also know they dont interact much with anything else.
So doesn’t it follow that most neutrinoes must still be around?
If so, they must more or less have distributed their mass uniformly throughout the universe?
Is everybody still ignoring the collective mass of all the neutrinos?
Yes. A remarkable turn of events.
Yes.
Yes.
Yes.
No. Neutrinos have to be considered when fitting the Planck CMB power spectrum, for example. This provides the strongest formal upper limit on the sum of neutrino masses. It also provides a potential test of LCDM, as I’ve previously pointed out: if a laboratory experiment (like KATRIN) were to measure a neutrino mass larger than the Planck limit, it would [in principle] falsify the LCDM structure formation paradigm.
I may come back to you re Genzel et al.
In the meantime, it occurs to me that one point in favour of rejecting the hypothesis that DM accounts for the rotation curves might be the observation that spirals become more tightly wound over time. Would you agree that a great mass of DM beyond the disc would counteract gravitational contraction of the arms, and has that been discussed in the literature? Sellwood & Masters 2022 have a section on rotation curves but do not address the point.
Do they? It is not obvious to me that spiral arms become more tightly wound over time, or how we could even make that determination. Spiral structure has been a rich field for a long time, in part for lacking clear answers. Sellwood is one of the great workers in this field, but I’ve never heard him suggest anything remotely like what you describe.
You seem to think of spiral arms as structural entities unto themselves like galactic I-beams. That’s simply not the right way to think of them. They can’t be pushed or pulled any more than you can push or pull a wave in the water. It will respond to perturbations by changing in form, but cannot be stretched out like a pinwheel made of soft material. Spirals aren’t material in that sense; they’re a local density enhancement where there happen to be more stars at present.
Pringle & Dobbs 2019:
‘In density wave theory, α [the pitch angle] stays constant in time, being simply a property of the underlying galaxy. In other theories (e.g. tidal interaction, and self-gravity), it is expected that the arms wind up in time … For these theories, it would be expected that a sample of galaxies observed at random times should show a uniform distribution of cot 𝛼 [proportional to] t. We show that a recent set of measurements of spiral pitch angles (Yu & Ho) is broadly consistent with this expectation.’
Sellwood & Masters 2022:
‘The superposition of several coexisting modes causes the spiral appearance to change rapidly and the arms to appear to wind up over time. We argue that other theories have weaknesses ….’
Sure, there are some theories that predict that winding tightens arms. But does this happen in reality? I don’t think we know this observationally. I have no particular horse in this race, but have talked to Jerry Sellwood about it often enough to be skeptical of lots of the theories that are out there.
This is an interesting topic, with lots of theories. Tiret & Combes have an interesting discussion of spiral structure (etc.) in MOND (specifically AQUAL). Indeed, that spiral waves persist far out into the HI beyond the edge of the stellar disk is an argument in favor of MOND, as one needs some disk self-gravity for that to happen yet DM halos should dominate at those radii. The way out of that has always been “we don’t understand spiral structure well enough.”
Though certainly interesting, the subject of spiral structure is of little relevance to this post.
If one runs with the rather obvious implication of spiral structures that the arms – two arms, generally, projecting from opposite sides of the nucleus equator, and particularly obvious when one looks at the dust lanes traced by IR – actually issue from the nucleus, then the subject is potentially very relevant.
As I have half remarked already, the essence of these structures is that the arms expand outwards. Close to the nucleus, stellar orbits are close to co-rotational; further out the structure is progressively more expansive. If the apparent trajectory of the arms more-or-less reflects the actual vectors, then necessarily the rotation velocities further out will be higher than if the orbits were circular.
Beyond the MW, velocities are measured from the varying Doppler shift of the rotating stars en masse rather than for individual stars. Hence, so far as I can see, circular orbits on average are assumed rather than observed.
I would be interested to see someone consider whether there might be a quantifiable relationship between the stellar part of the RC profiles and the spiral morphology. Perhaps this possibility should be ruled out before one commits oneself to the idea that the Newtonian/Einsteinian theory of gravity ceases to be valid beyond a certain point.
No, that is not how it works. Orbits do not become more expansive further out; spiral arms do not result from stars emanating from the nucleus. Near-circular orbits are measured by Gaia using proper motions as well as Doppler radial velocities. You are pushing a picture that is simply wrong.
Yes, there is a quantifiable relationship between the stellar part of the RC and the distribution of the stars, and not just in spiral arms. That’s Renzo’s rule. This relation persists into the regime where Newtonian/Einsteinian gravity clearly fails, so one needs something extra, be it dark matter or a modified force law.
You are making arguments that were examined and discarded as obviously unworkable forty years ago. I won’t teach a whole course on galaxy dynamics on this blog; if you don’t believe me then go read Binney & Tremaine’s textbook.
This is an indirect statistical method with quite a few steps from the lensing signal to the implied flat rotation curves. It is certainly intriguing but I find that plotting it and stating it as if it were a directly measured rotation curve is a bit disingenuous.
Hi Stacy.
Thank you for this post on “rotation curves”.
I tend to agree with steppenwolf433 in that the Mistele et al paper is principally about the spherically-symmetric gravitational field as derived from weak lensing data. The orbital speeds then apply to objects in circular orbits at any orientation, not necessarily in the plane of the galactic disk. It is then a jump to implying the rotation curve of disk galaxies might extend out to 1 Mpc. Indeed your previous post 26-Apr-2024 “The MHONGOOSE survey…” suggests that galaxies have hardish edges and don’t continuously decline into the intergalactic medium, so going out to 1 Mpc may be a stretch too far.
Nevertheless, the result is a good plus for MOND and something new for LCDM to find a fudge for.
Separately, Table 1 in the Mistele et al paper only gives a few velocities and states that the full table is available in the machine readable version. Do you know how one can access that? Thank you.
Didn’t see this comment until after I replied to the one it replies to, which I mention because I expect WordPress will make the order of replies look funny.
First, the data are on the journal site: https://iopscience.iop.org/article/10.3847/2041-8213/ad54b0. At the bottom of each displayed table there is a link to the full data table, either in ascii or machine readable format.
Second, the “edge” you refer to applies to the cold gas distribution. There can be other things beyond that; a common thought is that the gas continues in ionized form.
Third, the gravitational potential extends well beyond the truncation of the mass distribution, wherever that may be. We see no hint of it in the data, so one needs either a very extended DM halo or a theory like MOND that makes the rotation curve flat no matter where the mass ends.
Fourth, the spherical approximation is pretty good when one gets over 100 kpc out. Everything looks like a point mass from far enough away.
I agree with the concerns you and steppenwolf raise – you sound like me in our collaboration meetings. But this is what the data show. Note that we corroborate Brouwer et al on the point that inferred rotation curves are flat way far out: see their Fig. 3. We’ve been able to improve on their analysis, pushing the credible region still further out, but they did all the heavy lifting and the basic result is already in their paper. We thought they kind of buried the lede (despite one of us encouraging them pre-publication to emphasize this remarkable aspect of the data), and that people are only having this reaction now seems to confirm that it warranted saying again.
The new thing we have done is use these data for the Tully-Fisher relation, which Brouwer et al. did not consider. I hope to get to that in another post, but if you’re eager, it is in the paper along with the data tables.
Hi Stacy.
Thank you for the reply – much appreciated – I have successfully picked up the full circular speed data.
I notice that all the circular velocities in Figure 2 from Mistele et al show a decline beyond 500 kpc; this is clearer when the distances are plotted on a linear rather than logarithmic scale. To my eye the declines run parallel to the NFW curves, but with higher speeds. I appreciate that the error bars at these distances are large and that the apparent declines may not be significant statistically. Nevertheless it seems likely that the circular speeds do not remain flat indefinitely.
The data for late type galaxies is consistent with mildly declining rotation curves if one goes far enough out. They’re also consistent with remaining flat, given the uncertainties.
I caution against drawing any conclusions about paralleling NFW. When we get far enough out, the 2-halo term (all the DM besides that of the central galaxy) kicks in, and LCDM no longer predicts a pure NFW profile.
Surely the tyranny of flatness must end at some point, even in MOND, but it is not clearly detected in these data, hence our use of “indefinite.”
Right, the tyranny of flatness only stops due to EFE in Mond, and so by selecting isolated galaxies you-all found it flat out to what we can measure. It’s sweet.
It is certainly true that this is an indirect statistical method, something I find generally unsatisfactory. One does have to average over many galaxies, as I said. Nevertheless, this is what the data imply for the average gravitational potential. I don’t see how that is disingenuous – should we *not* say what the data imply? Moreover, this is an overview in blog form. If you want all the scientific caveats where we sweat the many details, read Mistele et al.
Yep. The WordPress nesting makes the order of replies look funny, since the reply this is a reply to was written before the one that appears above it.
This situation reminds me of the early days of the cusp-core debate. When we first pointed out the problem, we analyzed rotation curves. Theorists complained that what they predicted was the inner density profile. That is tested directly by the measured rotation curve, but if what they want to see is the density profile, we thought fine, we can get that from the data – hence de Blok et al. 2001. Then, after showing them what was requested, the complaint became that what we were measuring was rotation curves to we should stick to that.
The lensing measurements measure a quantity called the excess surface density. That provides a measure of the gravitational potential, but not one that most of us find intuitive. Both Brouwer et al. and Mistele et al. translate the excess surface densities into radial accelerations, which is fairly straightforward. I’m used to looking at the radial acceleration relation, so I knew immediately upon seeing Brouwer et al’s result that it implied flat rotation curves way far out. Most people don’t have that intuition, but do have familiarity with the circular velocity curve as a tracer of the potential.
Rotation curves appear to stay flat. That is not a surprising result from an empirical, historical perspective. It was surprising in the 1970s, it was expected thereafter. The weird thing is that it apparently continues as far as we can trace it. How far? At least 300 kpc, and it seems to persist to 1 Mpc. I get that this seems outrageous, so we worried a lot about it, but ultimately there is nothing in the data to say otherwise.
I am an engineer and was a student of material science so your article is my introduction to this thanks!
My understanding is that the calculated rotation curve is the velocity that would keep these stars and gas in orbit around the galaxy. Since the outer stars and gas are moving faster than this velocity, might they simply not be bound to the galaxy and be moving away from it?
This is the inferred circular velocity of the gravitational potential, so test particles like stars on zero eccentricity orbits would be moving this fast. The escape velocity depends on how far the mass distribution extends, which we don’t know.
In the Milky Way, there are a handful of high velocity stars known that are likely unbound. Those don’t stick around, departing on a timescale of a few hundred million years. That’s a long time to you and me, but very short compared to a Hubble time. So anything that doesn’t stick around tends to leave relatively quickly. The stars and gas we see is, in bulk, not doing that, or we wouldn’t see it.
A few million years seems to be enough for the formation of galaxies, if we accept the current mainstream thinking, but obviously we should doubt/question everything:
JADES-GS-z14-0, formed just 290 million years after the Big Bang.
https://www.astronomy.com/science/webb-discovers-the-earliest-known-galaxy-for-now/
Ah thanks so to follow this out Jeremy, stars in galaxies may be able to form at similar time scales to how long it takes them to leave the galaxy. Or at least stars can be added to galaxies at similar time scales that they leave galaxies.
With a search I found that research of astrophysicist Romeel Dave found that 90% of matter is in the intergalactic medium.
https://www.quora.com/Is-there-matter-between-galaxies
Suppose that as the galaxy moves through space, it collects matter, and suppose a higher percentage of matter is collected on the outer part of the galaxy than the inner part or more on one side than the other. This would destabilize the center of mass of the galaxy and could then result in the outermost stars in some places having enough velocity to escape. This would be a constant and dynamic process and so the galaxy would replace stars lost with growing ones further inside.
Certainly there is lots of baryonic matter in the intergalactic medium, though it is spread so thin that it a very hard vacuum. Lots has been written about both accretion onto galaxies from the IGM and the expulsion of baryons back out into it. Gas flows driven by things like supernovae are usually invoked for the latter. Getting rid of stars is harder; the don’t respond to winds of low density material, and once bound to a gravitational potential well, they are extremely hard to excavate. One needs to transfer a lot of kinetic energy to them somehow. Close triple interactions can do that, but are very rare because space is big and you need to get three objects very close together at the same time.
arXiv:2406.01705 (hep-ph)[Submitted on 3 Jun 2024]Dark MatterMarco Cirelli, Alessandro Strumia, Jure ZupanWe review observational, experimental and theoretical results related to Dark Matter.
any one reading this ? 550 pages
yes, but not quickly. I am likely to write [another] 500 pages on these subjects before I read all of these 500.
550 pages of dense material on dark matter.
so do you have any thoughts on the content of that paper arXiv:2406.01705?
Well, no, I don’t like to comment on things until I get a chance to read them. That’s going to take a while, so I will comment on the abstract, which is a nice one-liner:
We review observational, experimental and theoretical results related to Dark Matter.
The first thing to realize is that when people say “dark matter” a better term would be “mass discrepancy” or “acceleration discrepancy.” The former is a more general term; the latter was suggested by Bekenstein to more accurately describe what is going on: we observe a discrepancy from known laws at low accelerations.
By calling it dark matter, we presume the answer. It took me a long time to realize this, and “dark matter” is an easy thing to say, so I understand why people do it, but it is a kind of category mistake. The evidence only indicates a discrepancy, it doesn’t provide a guarantee that the reason for the discrepancy is dark matter within the broader category of either dark matter or a modification of the dynamical laws.
They list three categories of results that they review. Among these, the only one that provides evidence is from observational astronomy. The experimental results, while vast and worthy of review, so far provide only non-detections and upper limits. That’s not a form of evidence beyond telling us what the answer is not. Theoretical results presumably have to do with ideas about what the dark matter might be; here I expect (though I have not checked) this is mostly and perhaps entirely in the realm of new particle physics because that is the category they appear to have selected.
As I’ve discussed many times here, there are many aspects of the observational evidence that disfavor the usual particle dark matter interpretation. If that is the case, then particle physics simply isn’t relevant.
I browsed the table of contents and skimmed chap. 5 pretty good. (I know nothing) figure 5.5 on page 132 tells the story. And seems to me, not much space left for DM to exist in.
Once again, Dr. McGaugh, thank you for this blog and your efforts to make difficult science accessible to non-experts.
Stacy
could you rule out dark matter halo and support MOND at extremely large significant by 5+ sigma in your paper
are there any dark matter distribution and theories that support flat curve for millions of light years
I’m not aware of any theory that predicts a dark matter distribution that gives a flat rotation curve so far out. However, the problem with falsifying dark matter is figuring out what it predicts (and getting people to concur). So on the one hand, what you ask has already been accomplished. On the other hand, it is an impossible ask: the dark matter is invisible, so one is free to distribute it however one needs to explain any observation. That in itself is bad, and is to me a bigger problem than any particular observation, but not everyone feels that way.
Question – if the flat rotation curve out to many, many galactic (visible) radii is due to dark matter, then does the angular momentum of the galaxy become “unreasonable” in some way? I’m assuming that the dark matter is also in orbits like those of the visible matter. Maybe if the angular momentum is large enough relativistic frame dragging becomes significant? Or some such thing?
The orbits of the dark matter particles can not be like those of the visible matter as this would make spiral disks unstable. The orbits of DM particles are generally thought to be very radial (highly eccentric) and approximately randomly oriented. The DM halo doesn’t have to have much angular momentum, so I don’t think this is a worry.
Stacy
if the wide binaries test provides for MOND at 5+ sigma significance plus your results show here “Rotation curves: still flat after a million light-years” plus prior like RAR and BTFR be enough to kill dark matter for individual galaxy ?
That is the question I keep posing: what is the standard by which dark matter would be falsified? I’m not the person who needs to answer this.
That is the wrong question, even if simply rhetorical. There is no scientific evidence for the existence of dark matter beyond its role as rolling fudge factor necessary to maintain the belief that our gravitational models, developed in the context of the Solar System, are applicable to systems whose existence, scale, and complexity was completely unknown to Newton and Einstein at the time they developed their models.
The argument for dark matter is circular:
We have a universal law of gravity.
Dark matter is necessary to fit large scale systems to that law of gravity.
Therefore dark matter must be there because we have a universal law of gravity.
From a scientific perspective there is no evidence for dark matter beyond that circular reasoning. You cannot falsify a belief in something that has no objective existence in empirical reality. You cannot falsify a person’s belief in angels beyond stating that there is no empirical evidence for their existence in physical reality. Same goes for dark matter.
Science is not supposed to be the study of belief systems. That is the realm of philosophers and theologians. What separated science from those disciplines is that it restricts itself to the study of empirical reality. That restriction is what made science successful. Math is an indispensable tool in that endeavor but it is only one tool.
Unfortunately, modern theoretical physics has devolved into the study of mathematical models and the philosophy of mathematicism has taken root as the default operating paradigm. Math dictates to physics how it should interpret its findings and sends physical researchers on fruitless, decades-long searches for invisible entities required to justify mathematicism’s unshakeable belief in its flailing 100 year old models.
Until physical researchers wrest control of the scientific agenda from the theorists the science of the academy will remain in a crisis of its own making. Theory cannot drive physical research; physical research has to drive theory.
Audience. Audience Audience Audience.
It does not matter what I think the standard should be. It does not matter what you think the standard should be. The issue is what it would take to get scientists who work on dark matter to reconsider. For me, speaking as someone who once believed in dark matter as much as anyone, it was the series of fine-tuning problems that arose trying to use dark matter to explain phenomenology that I later learned had been predicted by MOND.
Others scientist need to get there on their own; none are going to accept dark matter is wrong just because we say so. Hence my question: to provoke them to consider a situation that they mostly have not, and are reluctant to even contemplate.
You are correct that there is too much theory-directed data interpretation going on. This is one reason I left physics: the string theorists were taking over, but they were predicting untestable things: that, to me, is not physics. So yes, theory needs to sit its ass down and listen to the data, not the other way around. But just saying that should happen doesn’t make it so.
I don’t think anybody should accept that dark matter is wrong simply because I say so but shouldn’t there be a requirement that those who insist dark matter is right offer some scientific justification for that claim beyond “it makes our model work”? The dark matter hypothesis should be justified by more than just mathematical convenience at this point, shouldn’t it?
Your point that saying theory should follow the data won’t make it happen is true if the discussion is limited to isolated individuals like you and me – but if it becomes a widely discussed issue?
Yes, they do. They don’t, by and large, so yes, we need to bring it up. But be aware that we’re communicating from two very different places in terms of unspoken assumptions. That makes communication challenging.
what is your preferred explanation for galaxies clusters – dark matter or another modify gravity for galaxies clusters
I don’t have a preferred explanation for galaxy clusters. Some of the evidence points to dark matter, some to MOND. The sum makes no sense, so I don’t know what to make of it.
It’s important to acknowledge what we don’t understand. What isn’t OK is to be sure we do know more than we really do. Scientists overestimate how much we know all the time, and clusters of galaxies are a prime example.
does lambda cdm and standard dark matter and Newton gravity explains clusters of galaxies without any fudging ?
Lambda and CDM are themselves fudges: https://tritonstation.com/2017/03/06/lcdm-has-met-the-enemy-and-it-is-itself/ And no, LCDM does not explain all of the data without fudging, but it depends on which aspect of the data you’re considering – see Clusters ruin everything.
I found out that I used the word ‘mathematicism’ wrong. I didn’t mean Max Tegmark’s view, I just meant overemphasizing the mathematical side of physics, and underemphasizing the conceptual side.
Congratulations Stacy to you and colleagues for the important discovery about flat rotation curves extending much further out.
It adds to something I was thinking about – any explanation must include as little as possible of a conspiracy to make FRCs. DM halos in ΛCDM have this problem in spades. In my theory there are different versions of an explanation, and a lot less of that problem (they lead to MOND which means FRCs). But choosing between them is affected by that issue, and it’s more of one now.
Now all we have to do is VERIFY and VALIDATE the instrumentation methodology for spectral velocity measurements; then we can proceed with the velocity flatness assertions. REDSHIFT is caused by various factors, ie., not Doppler shifts. This has led to vast errors and wrong theories; measurement MUST BE verified. So far, astronomers and astrophysicists have NOT verified that spectral velocities correlate with astrometric velocity measurements.
Lensing isn’t a Doppler effect, so the agreement between it and methods kinematics is itself a form of corroboration.
Not sure what’s happening but your recent responses are cluttered with formatting code and appear truncated.
Am traveling so been trying to reply on my phone. Apparently that’s bad medicine.
weak lensing depends on redshift see references such as
https://ui.adsabs.harvard.edu/abs/2017ApJ…850…24T/abstract
Lensing depends on the distance to the lens and to the source behind it. That is adjudicated by redshift in a cosmological context. So again, that lensing gives the same result as kinematics corroborates both the Doppler interpretation of the kinematic data and that redshift represents cosmological separation. None of this would work if either of those were broken.
You’ve probably seen this, but just in case, Sabine Hossenfelder has admitted it’s very strong evidence, and it has had 300k views in 5 days
Although she’s lightweight in some ways, and admits she has swung backwards and forwards, she’s respected, and it’s good when the truth about the evidence seeps out in any form really, especially after, or during, a long frustrating time trying to wake people up to it.
One should change one’s mind with the evidence.
when we say the universe is expanding, I think it is not expanding within gravitationally bound systems like the solar system or the Milky Way. The transition between the edge of a bound system to the expanding universe I never quite understood. And now a million light years from the center of a galaxy is part of the gravitationally bound system – that is what flat rotation curves imply to me. So where are the interstices where the universe is expanding?
That is, two test masses placed at rest with respect to the distant galaxies will fall/orbit within the limits of a gravitationally bound system, but will drift apart while remaking at rest where the expansion of the universe dominates. What happens in the in between space if any and where is this space?
Or is the universal expansion universally present, however minutely affecting the orbit of the earth?
Good question. I’ve wondered the same thing about the interface between a bound system and the cosmic expansion.
All we infer from the lensing data is that isolated galaxies seem to exert an influence that persists indefinitely. The situation is different for most galaxies that have neighbors. For example, it appears that one needs to consider the cosmological constant in the dynamics between the Milky Way and Andromeda.
All I would say that we really know is that it appears that the cosmic background is expanding and that the best approximation we have for the dynamics of objects within this expanding universe is MOND.
Something that has been on my mind for a long time, but I can never seem to really comprehend whether the question is useful or not is this. Let’s assume we have created the notion of 3 spatial dimensions and 1 time dimension out of local necessity – as an extension of more primitive ideas – and that this notion has helped us interpret our local environment. Are we so sure that this same notion should be applied to nonlocal observations, or for observations where the distinction between time and distance is of little value?
At question is whether we can truly say what the local acceleration of a very distant star actually is from only the distant measurement of the spectrum, or from lensing. I mean, of course we can say it, but does it lead to something useful?
Certainly many of the attempts to find a deeper theory of everything invoke higher numbers of dimensions. Whether these matter to the issues at hand is hard to say – it is usually assumed not. But that so many things are broken in weird ways might be taken as a clue.