Non-baryonic cold dark matter ("CDM"), the observational evidence
- Thread starter DeiRenDopa
- Start date
In a nutshell, various physicists have written down every New Fundamental Force Law they can think of, and none of them describe the data. Simply saying "Gravity is stronger than GMm/r^2 at large r" doesn't work. Saying "Gravity is stronger than GMm/r^2 at small velocities" doesn't work. Saying "Gravity is stronger than GMm/r^2 at small GMm/r^2" doesn't work (that's MOND; it's the one that came closest to fitting rotation curves). And so on. You're welcome to think up alternatives, but the list of tried-and-failed equations (or even forms of equations, or trends, or scaling laws) is pretty comprehensive.
Meanwhile, one new type of matter ("something cold and noninteracting") fits all of the data across the board. It didn't have to; there were a dozen major chances for the data to support or falsify this, and it's come down on "support" every time.
geni
Anti-homeopathy illuminati member
- Joined
- Oct 14, 2003
- Messages
- 28,209
No that is modification of a known fundimental force. What about a new one?
However it would require a rewrite of the standard model. So far things that have requied that have tended not to work out too well.
sol invictus
Philosopher
- Joined
- Oct 21, 2007
- Messages
- 8,613
There are extremely sensitive tests of so-called "fifth forces" in laboratory experiments on earth. It's very, very difficult to invent one that's strong enough to affect galactic rotation but isn't ruled out by those experiments.
Moreover, the bullet cluster observation pretty much kills the whole idea - that was a direct observation of the dark matter halo of two galaxy clusters.
Actually we know the SM is not complete, and many extensions of it (proposed for totally different reasons) contain a natural DM candidate.
You're just arguing over semantics. If the real force on a star is GMm*(1/r^2 + const*v^2/r^3), is that a "modification to gravity" or the addition of a new (v^2/r^3) force? That might make a difference later when you want an underlying explanation, but it's an unimportant distinction when you're trying to calculate how things move around a galaxy---and that's all you need to do in order to compare the equation to the observations.
Last edited:
Wangler
Master Poster
- Joined
- Feb 20, 2008
- Messages
- 2,228
I thought that not only did MOND come closest to fitting rotation curves, it showed a remarkable ability to do so, across a broad range of galactic sizes.
That's what I meant by "came closest". It does a good job of fitting rotation curves, but a terrible job of fitting any of the other data (lensing, the Bullet Cluster, cosmology, etc.). Chris Stubbs at Harvard is analyzing whether MOND fits non-rotational components of galaxy dynamics (disk oscillations) and it sounds like the answer there is going to be another "No".
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Although I've yet to finish even my first point, I'll comment on this now.
A new, fundamental, force cannot be ruled out ... provided that force is not quantified, not specified, or even vaguely hinted at.
The moment you start to write it in words, even using poorly-defined terms, you begin to be able to start answering this question.
When you get to writing any such 'new force' in the form of an equation, you can test the idea.
Hundreds, perhaps thousands, of people have tried to come up with something specific and testable, in the way of a 'new force'. It's very easy to do (make up a new force) ... and in most, if not nearly all, cases it's also just as easy to show that such a new force won't work. Why? Because it does not fit the relevant observations, either directly or indirectly.
One of the wonderful things about astronomy is that there are millions, billions, or even trillions of excellent observations you can use to test new ideas like 'new fundamental forces' ... and most of this data is available to anyone, anywhere in the world, essentially for free (you need an internet connection, preferably broadband, a browser, and appropriate associated software ... and that's all).
.
Consider this: barely a half century ago, no one knew about quarks, not even up or down ones, much less charmed, strange, top, or bottom ones.
No one knew that there was a tau neutrino.
And so on.
And why? Many reasons, but one is that the energies physicists had available, to do their atom and nucleus smashing, were really quite low.
Today particle accelerators can get up to ~1 TeV, that's 1012 eV, and the LHC should go a factor of 10 higher.
And all that wonderful new physics was found in an energy range that is what, ~3 orders of magnitude wide (~1 GeV to ~1 TeV).
Big deal.
The Earth is hit, every day, by millions of cosmic rays that have energies at least a million times higher! That's 6 orders of magnitude above the best we can do in labs today. The highest energy cosmic ray recorded had an energy of ~1021 eV ... a full 8 orders of magnitude above the best the LHC is likely to produce.
And the universe very likely can churn out particles with energies considerably higher still, at the rate of billions of tonnes a second ...
And anyway, as has already been noted, the Standard Model, of particle physics, is known to be incomplete - neutrinos don't have mass in it, for example, yet neutrino oscillations require that at least one (two?) do.
BenBurch
Gatekeeper of The Left
DeiRenDopa,
In 1975 when I suggested that the answer to the Homestake detector solar neutrino problem may lie in the fact that in the laboratory our neutrinos are very young, with the Homestake detector being calibrated based on a smaller volume of carbon-tet right next to a reactor, and the ones from the sun have travelled for minutes, it was thought preposterous. Now we know about neutrino oscillation. New things come into our particle zoo ALL the time. I am not holding my breath for the end of physics.
In 1975 when I suggested that the answer to the Homestake detector solar neutrino problem may lie in the fact that in the laboratory our neutrinos are very young, with the Homestake detector being calibrated based on a smaller volume of carbon-tet right next to a reactor, and the ones from the sun have travelled for minutes, it was thought preposterous. Now we know about neutrino oscillation. New things come into our particle zoo ALL the time. I am not holding my breath for the end of physics.
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Thanks.
Just a clarification, if I may ... you're saying that in the ~8 orders of magnitude of particle energy, between LHC and UHECRs detected, there is very likely to be lots of 'new' particle physics?
And that, as with anything 'new' like this, there's absolutely no way to tell, in advance, what sorts of 'new particle physics' will be discovered?
Including (drum roll please) the possibility of a stable CDM particle (or a whole zoo of them)??
Last edited:
BenBurch
Gatekeeper of The Left
Well, you never know. The Cosmic Ray physics guys see things that they can't explain sometimes, but nobody ever sets much store by that because its just not repeatable or calibrated except in the most gross fashion. I expect to be surprised.
But, I've also been out of physics for decades now, so what do I know?
But, I've also been out of physics for decades now, so what do I know?
At least two I believe, based on the combination of solar (elecron type initially) and atmospheric (muon type initially) neutrino data.
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
.Well, you never know. The Cosmic Ray physics guys see things that they can't explain sometimes, but nobody ever sets much store by that because its just not repeatable or calibrated except in the most gross fashion. I expect to be surprised.
But, I've also been out of physics for decades now, so what do I know?
The cosmic ray guys (there were no gals back then, were there?) sure did a lot for particle physics - discovery of the positron, muon (and, indirectly, the muon neutrino?), pion, kaon, and more? - back in the 1930s and 1940s.
While particle discoveries have come, since then, from accelerators, the existence of natural particle accelerators far beyond the capabilities of Earthly ones is surely a nice spur for doing more research.
But my main point was that a great deal of 'new physics' was found in the ~GeV and ~TeV range (and much 'different' physics between ~eV and ~MeV, and between ~MeV and ~GeV); surely the universe cannot be so mean as to have a complete desert of 'different' physics over the next ~8 orders of magnitude, can it?
The existence of UHECRs shows particles can be accelerated to these energies, even if we still don't know (for sure) whether those particles are protons, ... iron nuclei, ... or something mildly (or quite) exotic ...
This is a pretty lame argument, and I'd be happy to see it die. The Standard Model is perfectly happy having the neutrinos be massive; the original practice of writing the masses as Zero (rather than as "0 < m < experimental limit") was quite simply wrong. Rather than saying "the standard model is incomplete" over and over again, we simply fixed it.
(There are *other* reasons for the SM to be incomplete, pending an understanding of the strong-CP problem, the Higgs mechanism, etc., but neutrino masses aren't on the list.)
sol invictus
Philosopher
- Joined
- Oct 21, 2007
- Messages
- 8,613
I had in mind electro-weak symmetry breaking and the hierarchy problem.
But (as you probably know) the smallness of neutrino masses also implies new physics at a higher scale.
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Hmm ...
Perhaps some clarification is needed?
There's lots of stuff out on the internet, and in popsci magazines like Scientific American, which talks about neutrino masses being zero in the Standard Model (SM). Even this Wikipedia article* ... but it does make a distinction between "SM classic" and several ways the SM can be tweaked to accommodate them.
To what extent does tweaking the SM to accommodate non-zero neutrino masses automatically lead to something potentially testable, with the LHC for example? For sure the focus is on the Higgs mechanism far more than it is on which tweak of the SM is best for neutrino masses ...
* cited with trepidation, given all the well-known problems with this source ...
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Why must galactic halos be CDM (and not, say, sand)?
There's a lot of mass in the halos of a lot of galaxies ... that's a robust conclusion from a lot of astronomical observations.
Galaxy halos are 'dark' in the sense that they do not emit light, nor do they absorb it.
So what are the observations that rule out 'baryons' as the mass in galaxy halos? That's what this post is about.
First, a definition and some clarifications.
By 'non-baryonic matter' I mean matter which is not the kind of stuff that people, rocks, inter-stellar dust, (hydrogen) gas, normal stars, white dwarf stars, and neutron stars are made of. Neutrinos are an example of non-baryonic mass. Are black holes non-baryonic matter? I will leave that question open, for now.
There is certainly baryonic matter in galaxy halos, and a lot of it ... there are stars, there is gas, and some dust ... and there are almost certainly rogue planets, small chunks of iron-nickel, boulders, mountains, and so on. The question I will examine in this post is whether any one kind of baryonic matter could comprise as much as ~10% of galaxy halo mass, or whether the combination of all kinds of baryonic matter could comprise ~80%+ of it (why not ~100%? a backward recognition of observational uncertainty).
Last clarification: the more you require consistency across a range of astronomical observations, the more clearly you can rule out a particular kind of baryonic mass. For example, if you use only the deep HST images, you can't rule out a large population of (very) old, long 'dead' low mass stars in the MW halo (they'd be too faint for the HST to detect); however, if you include all the relevant observations of all stars, to date, you would have good grounds for saying that no such stars exist, period (not to mention that any such stars would have shown up in projects like OGLE (see later)). The same applies to moving away from the cosmological principle - the more you insist that the MW is special, the easier it is to make a case that the halo of some distant spiral is comprised of lots of red and brown dwarfs, for example (no CDM needed).
In a nutshell then:
hot gas/plasma: no; x-ray emission from galaxy halos is too small, and the x-ray spectra of distant sources show no 'local' (absorption) lines
cold gas: no; no 21 cm emission (from H), the spectra of distant sources show no 'local' (absorption) lines
small clumps of cold gas: no; they'd collide to form bigger clumps, they'd form stars, and too few micro-lensing events (see below)
dust: no; distant sources are not reddened (the same principle that gives you red sunsets), nor is there any far-IR emission (which you'd expect, dust should radiate more or less as a blackbody)
faint stars (red dwarfs, brown dwarfs): no; very deep Hubble images of random halo fields show no such stars, too few micro-lensing events
faint stars (white dwarfs): no; as above, plus no evidence of the matter that should accompany them (the gas, dust, etc that would form during the process by which the stars became white dwarfs)
super-Jupiters, Jupiters, etc adrift in the halo: no; too few micro-lensing events
sand, rocks, boulders, mountains, dwarf planets: no; whatever elements they'd be made of, erosion should have enriched the halo gas to an extent it'd be clearly detected; here on Earth we are not being bombarded by such things (there'd be lots of them, if they made up even 10% of the MW halo mass, and they'd be arriving at speeds far in excess of the run-of-the-mill meteorite)
hydrogen snowballs: no; they'd sublime in quick order (all that light in space, all those cosmic rays), and the resulting clouds of H would be easily detected.
A word on micro-lensing: if you stare at enough distant stars (in the Magellanic Clouds, for example) for long enough, you can expect one to show an achromatic, symmetric brightening and fading. This is a micro-lensing event, caused by a foreground object (hopefully a MACHO - massive compact halo object) moving into the line of sight of the distant star and out of it again - another example of (GR) gravitational lensing. This technique is sensitive enough to detect compact objects, from Neptune-mass 'planets' through brown and red dwarfs to ordinary stars (and plans are afoot for facilities capable of detecting Earth- and even Mercury-mass planets) in the MW halo, or at least that part of the halo close to us. There have been several projects of this kind - OGLE, EROS, MACHO, super-MACHO, MOA - the combined results are that there are too few MACHOs to account for even a small fraction of the MW halo mass.
This has been a very brief summary; I'd be happy to expand on any aspect.
Next: possible alternative (non-textbook) physics explanations.
There's a lot of mass in the halos of a lot of galaxies ... that's a robust conclusion from a lot of astronomical observations.
Galaxy halos are 'dark' in the sense that they do not emit light, nor do they absorb it.
So what are the observations that rule out 'baryons' as the mass in galaxy halos? That's what this post is about.
First, a definition and some clarifications.
By 'non-baryonic matter' I mean matter which is not the kind of stuff that people, rocks, inter-stellar dust, (hydrogen) gas, normal stars, white dwarf stars, and neutron stars are made of. Neutrinos are an example of non-baryonic mass. Are black holes non-baryonic matter? I will leave that question open, for now.
There is certainly baryonic matter in galaxy halos, and a lot of it ... there are stars, there is gas, and some dust ... and there are almost certainly rogue planets, small chunks of iron-nickel, boulders, mountains, and so on. The question I will examine in this post is whether any one kind of baryonic matter could comprise as much as ~10% of galaxy halo mass, or whether the combination of all kinds of baryonic matter could comprise ~80%+ of it (why not ~100%? a backward recognition of observational uncertainty).
Last clarification: the more you require consistency across a range of astronomical observations, the more clearly you can rule out a particular kind of baryonic mass. For example, if you use only the deep HST images, you can't rule out a large population of (very) old, long 'dead' low mass stars in the MW halo (they'd be too faint for the HST to detect); however, if you include all the relevant observations of all stars, to date, you would have good grounds for saying that no such stars exist, period (not to mention that any such stars would have shown up in projects like OGLE (see later)). The same applies to moving away from the cosmological principle - the more you insist that the MW is special, the easier it is to make a case that the halo of some distant spiral is comprised of lots of red and brown dwarfs, for example (no CDM needed).
In a nutshell then:
hot gas/plasma: no; x-ray emission from galaxy halos is too small, and the x-ray spectra of distant sources show no 'local' (absorption) lines
cold gas: no; no 21 cm emission (from H), the spectra of distant sources show no 'local' (absorption) lines
small clumps of cold gas: no; they'd collide to form bigger clumps, they'd form stars, and too few micro-lensing events (see below)
dust: no; distant sources are not reddened (the same principle that gives you red sunsets), nor is there any far-IR emission (which you'd expect, dust should radiate more or less as a blackbody)
faint stars (red dwarfs, brown dwarfs): no; very deep Hubble images of random halo fields show no such stars, too few micro-lensing events
faint stars (white dwarfs): no; as above, plus no evidence of the matter that should accompany them (the gas, dust, etc that would form during the process by which the stars became white dwarfs)
super-Jupiters, Jupiters, etc adrift in the halo: no; too few micro-lensing events
sand, rocks, boulders, mountains, dwarf planets: no; whatever elements they'd be made of, erosion should have enriched the halo gas to an extent it'd be clearly detected; here on Earth we are not being bombarded by such things (there'd be lots of them, if they made up even 10% of the MW halo mass, and they'd be arriving at speeds far in excess of the run-of-the-mill meteorite)
hydrogen snowballs: no; they'd sublime in quick order (all that light in space, all those cosmic rays), and the resulting clouds of H would be easily detected.
A word on micro-lensing: if you stare at enough distant stars (in the Magellanic Clouds, for example) for long enough, you can expect one to show an achromatic, symmetric brightening and fading. This is a micro-lensing event, caused by a foreground object (hopefully a MACHO - massive compact halo object) moving into the line of sight of the distant star and out of it again - another example of (GR) gravitational lensing. This technique is sensitive enough to detect compact objects, from Neptune-mass 'planets' through brown and red dwarfs to ordinary stars (and plans are afoot for facilities capable of detecting Earth- and even Mercury-mass planets) in the MW halo, or at least that part of the halo close to us. There have been several projects of this kind - OGLE, EROS, MACHO, super-MACHO, MOA - the combined results are that there are too few MACHOs to account for even a small fraction of the MW halo mass.
This has been a very brief summary; I'd be happy to expand on any aspect.
Next: possible alternative (non-textbook) physics explanations.
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Galaxies: possible alternative explanations of the observations relevant to CDM.
Alternative as in 'not found in standard physics textbooks'.
There is only one, that I know of, and that is MOND (MOdified Newtonian Dynamics). This is spectacularly successful at modelling galaxy rotation curves. As it does away with CDM entirely, it is also, by default, good at explaining the MW halo microlensing observations. Its supporters also claim it does a good job of explaining the (nearby) dwarf galaxy observations; I've not checked carefully enough to see how well it addresses other galaxy observations (I'll leave questions about MOND and rich clusters, and MOND and cosmology, until later).
Being a classical theory (or model), MOND cannot be the whole story, no matter how successful it might be at fitting galaxy rotation curves. Various attempts have been made to extend it, to incorporate relativity.
The MOND pages is an excellent source of all things MOND, being maintained by one of its leading proponents.
If any MOND fan is reading this, I wonder if you could answer one question I have, about MOND? Why has no MOND supporter sought to try to fit the M31 and M81 galaxy rotation curves? Not only are these two the first observed, but they are the most detailed, as these two galaxies are the in the top five of large, nearby, easily observed spirals.
Alternative as in 'not found in standard physics textbooks'.
There is only one, that I know of, and that is MOND (MOdified Newtonian Dynamics). This is spectacularly successful at modelling galaxy rotation curves. As it does away with CDM entirely, it is also, by default, good at explaining the MW halo microlensing observations. Its supporters also claim it does a good job of explaining the (nearby) dwarf galaxy observations; I've not checked carefully enough to see how well it addresses other galaxy observations (I'll leave questions about MOND and rich clusters, and MOND and cosmology, until later).
Being a classical theory (or model), MOND cannot be the whole story, no matter how successful it might be at fitting galaxy rotation curves. Various attempts have been made to extend it, to incorporate relativity.
The MOND pages is an excellent source of all things MOND, being maintained by one of its leading proponents.
If any MOND fan is reading this, I wonder if you could answer one question I have, about MOND? Why has no MOND supporter sought to try to fit the M31 and M81 galaxy rotation curves? Not only are these two the first observed, but they are the most detailed, as these two galaxies are the in the top five of large, nearby, easily observed spirals.
Wangler
Master Poster
- Joined
- Feb 20, 2008
- Messages
- 2,228
DeiRenDopa
Master Poster
- Joined
- Feb 25, 2008
- Messages
- 2,582
Thanks!
Well, that explains why M31 is not listed in the MOND pages!
A pity really, the '84-0-11' hit-rate touted there is unreliable, and unless McGaugh has been slow and/or sloppy in updating the MOND pages, it counts rather badly for his credibility. I note that Bekenstein did cite the Corbelli and Salucci paper, in his 2006 review paper (good for him).
Now for M81 ...
Oh, and to conclude: as McGaugh is not shy in pointing out, MOND has essentially no free parameters, so failure to fit the M31 rotation curve (if confirmed) is fatal to MOND, period.