Addressing Impossibilities in the Standard Cosmological Model

[turbo]

This is complete nonsense. There is nothing in standard cosmology that runs against standard physical theory -- in fact, we think it's one of the very few working cosmological model that uses known physics. Most of the others have to modify GR.

some examples:
1) inflation that turns on, then magically switches off simultaneously in causally-disconnected parts of the universe.
2) DM
3) DE
4) cosmological expansion that slowed nicely for billions of years, then accelerated (again, simultaneously in causally-disconnected parts of the universe)
5) Higgs boson
and you can add any other number of entities depending if your model depends on supersymmetry,etc.

This is taking the thread way OT, but I can offer you observational proof in one paragraph that the Higgs Boson cannot exist. As such, there is no need for me to learn all the arcana that led to the Higgs hypothesis.

 

[Evo]

This post was split off from the "I think popular science is ruining science" thread in GD.

 

[turbo]

Thank you Evo. Things were getting OT, but I could not let them go unanswered.

 

[SpaceTiger]

1) inflation that turns on, then magically switches off simultaneously in causally-disconnected parts of the universe.

Far from being "magical", there are many workable (consistent with known physics) theories of inflation. The procedure of both entering and exiting inflation are described in great detail here:

https://www.physicsforums.com/showthread.php?t=103866"

In particular, I recommend the link in post #19.

So what are the caveats? Well, the fields responsible for inflation obviously haven't been observed in the laboratory and it seems pretty unlikely that they will be in the near future. In order to produce the universe we currently observe, it had to decay into those things that are now the primary constituents of our universe (an event called "reheating"). Some theories of inflation do have predictions that are testable in the laboratory, but I'm afraid they're the minority. In order to test inflation, we'll likely have to resort to astronomical observations. If anyone is interested in the details of these observations, I'll be happy to elaborate. Perhaps hellfire will be too, since he seems to know quite a lot about the subject.

2) DM

Again, there are a lot of theories for this. The idea that it might consist of weakly-interacting particles is the most popular. We already know of particles that are weakly interacting and make up a non-negligible fraction of the energy density of the universe -- neutrinos. Considering the trouble we had to go through to detect these directly, it shouldn't be all that surprising if the dark matter particle has so far escaped our notice.

Many supersymmetry models predict a weakly-interacting particle. Not only is this not impossible, but I would even say that it's likely. We know that dark matter exists. The question now is, "is it enough?"

3) DE

The most popular explanation for the acceleration of the universe (which some include in "dark energy") is the cosmological constant. Far from being exotic, this is actually a very simple component of GR.

That said, mainly because of the fine-tuning problem, I have my doubts that it's the culprit for the current acceleration. It doesn't make the cosmological constant "impossible", as you seem to be suggesting, but it does leave it suspect. I would be more prone to suspect some kind of quintessence, which only suffers from the less severe "cosmic coincidence problem".

4) cosmological expansion that slowed nicely for billions of years, then accelerated (again, simultaneously in causally-disconnected parts of the universe)

This is basically equivalent to the dark energy point...and the causally-disconnected part is addressed by inflation.

5) Higgs boson
and you can add any other number of entities depending if your model depends on supersymmetry,etc.

The Higgs boson is a prediction of the standard model of particle physics, not the standard cosmological model. I'm pretty sure that LCDM could do without it, though some (not all) of the particle dark matter and inflationary theories would be ruled out by the lack of a Higgs boson

This thread is extremely rich with topics for discussion, so you should probably be more specific about what you'd like to understand. So far, I've only given a quick summary of why these things are not impossible. If you want to discuss observational constraints or other theoretical concerns, let us know.

 

[Garth]

We must not let the success of a posteriori adjustments to make the standard model fit the data blind us to possible alternative theories.

Inflation arises as a prediction of GUT and resolves the horizon, smoothness, density and magnetic monopole problems of the 1970's. The fact that the Higgs boson or an alternative inflaton predicted by those theories has not yet been found, that the subsequently necessary non-baryonic DM has not been identified and cosmic acceleration was only 'post-dicted' by the invocation of DE, is not to be forgotten in assessing the status of the mainstream model.

Testable, viable and concordant alternatives are therefore to be evaluated alongside the standard model.

Garth

 

[SpaceTiger]

We must not let the success of a posteriori adjustments to make the standard model fit the data blind us to possible alternative theories.

Who's saying we should ignore alternatives? This thread is about whether or not the standard model is "impossible".

 

[Garth]

Who's saying we should ignore alternatives? This thread is about whether or not the standard model is "impossible".

True, but I just thought the point ought to be made. A polemic can easily become polarised, I agree that there is nothing impossible about the standard model; just improbable IMHO!

Garth

 

[turbo]

True, but I just thought the point ought to be made. A polemic can easily become polarised, I agree that there is nothing impossible about the standard model; just improbable IMHO!
Garth

This might be a good time to ask how everyone treats these improbabilities mathematically. If the correctness of one improbable hypothesis is also dependent on the correctness of another improbable hypothesis shouldn't their probablilities be multiplied? In other words, if there is a 1% chance of A being true and a 1% chance of B being true, and if both A and B must be true to make the model work, the model's viability has been reduced to no more than .01%

The improbable things in my list above are not independent - they are inextricably linked and are model-dependent. The more such entities a model requires, the less trustworthy it becomes.

 

[turbo]

OK, let's take the "impossible" things one at a time. Let's start with Dark Matter.

GR gravitation is not predictive on galactic and cluster scales. Gravity in these massive domains seems to be far stronger than GR predicts. If we assume that GR is infallible, we come to the conclusion that there must be about 10 times more matter in each of these systems than we can detect observationally. This is a serious problem, because we are already pretty adept at detecting stuff and if DM was made out of any stuff that we are currently aware of, we would already have detected it. The way out of this problem is to say that DM is non-baryonic and weakly interactive, which leads to yet another problem. If the stuff is so weakly interactive, how does it manage to distribute itself "just so" in every relevant system? How can it form spherical halos with central voids around galaxies and then arrange itself in blobs and tendrils in clusters? This stuff is so elusive that we have never detected it, yet it is remarkably obedient.

After decades of diligent searching for DM, is it not time to admit that our understanding of gravitation is inadequate to explain the behavior of galaxies and clusters?

 

[Garth]

Nobody is taking GR as infallible, although perhaps it should be questioned more, however, galactic rotation curves and cluster dynamics are calculated under Newtonian gravity. (Hence the interest in Cooperstock & Lieu). The extra mass required by Newtonian dynamics also lenses (under GR gravity) distant quasars and so is doubly detected.

Once you get into explaining the detailed galactic structure you also have to include galactic magnetic and possibly electric fields as well as gravitational ones. Galactic MHD anyone?

Garth

 

[SpaceTiger]

If the stuff is so weakly interactive, how does it manage to distribute itself "just so" in every relevant system?

It's "weakly interactive" only for non-gravitational interactions. It still has mass and it will still clump gravitationally.

How can it form spherical halos with central voids around galaxies and then arrange itself in blobs and tendrils in clusters?

The spherical halos are actually a direct consequence of a low interaction cross section. If the dark matter were arranging itself into a disk or some other shape, then we would have a real inconsistency. In the absence of dissipation (which is a form of non-gravitational interaction), the halo of particles should end up with approximately spherical symmetry (for lack of a preferred direction). For another example of dissipationless "particles" that arrange themselves into a spherical configuration, take a look at a globular cluster.

I'm not sure what you mean by "central voids". There is no hole in the dark matter distribution in the cores of galaxies.

This stuff is so elusive that we have never detected it, yet it is remarkably obedient.

Actually, this could be viewed as a favorable thing for the standard model. The weak interaction explains many things at once, including the lack of observability, the lack of detectability in the laboratory, and the approximately spherical halos around galaxies and clusters.

After decades of diligent searching for DM, is it not time to admit that our understanding of gravitation is inadequate to explain the behavior of galaxies and clusters?

This has been under consideration for a long time (read up on MOND, for example). It doesn't, however, fit the data as well as the dark matter model.

 

[Mike2]

The weak interaction explains many things at once, including the lack of observability, the lack of detectability in the laboratory, and the approximately spherical halos around galaxies and clusters.

Excuse me, but I would think that the non-spherical nature of spiral galaxies would effect the distribution of dark matter halos so that the halo would not be spherical either, right?

 

[Chronos]

Perhaps it would be more effective to address the evidence in favor of the alternatives... which is mostly nonexistent. The standard [cosmological] model is not based on a few conjectures, like DM, DE or the Higgs boson, it is based on all of physics. The real pillar of the standard model [cosmological] is GR. Fell that tree and the forest is yours for the taking.

 

[Nereid]

If I may, can I ask that we distinguish between what's 'solid' in terms of a great deal of good observational or experimental results and what's contained in various mainstream (and not so mainstream) theories?

For example: for each of turbo-1's five OP points (actually only four, as ST has already noted), what's the status of the observational/experiemental basis?

Personally, I'd like to begin by getting some consensus on what this is, before we dive into the theories.

This might be a good time to ask how everyone treats these improbabilities mathematically. If the correctness of one improbable hypothesis is also dependent on the correctness of another improbable hypothesis shouldn't their probablilities be multiplied? In other words, if there is a 1% chance of A being true and a 1% chance of B being true, and if both A and B must be true to make the model work, the model's viability has been reduced to no more than .01%

The improbable things in my list above are not independent - they are inextricably linked and are model-dependent. The more such entities a model requires, the less trustworthy it becomes.

A very good, and very deep question, turbo-1!

I suspect that discussing it will take us way, way outside modern cosmology, into the heart of what the nature of modern science is, for example. Personally, I'd love to participate in such a discussion (partly because turbo-1 and I have already danced around the edges of just such a discussion, in another forum).

How do folk feel about starting another thread, in PF's Philosophy of Science section?

 

[Chronos]

I'm game, Nereid.

 

[turbo]

I'm not sure what you mean by "central voids". There is no hole in the dark matter distribution in the cores of galaxies.

Dynamical simulations have pointed to a cuspy distribution of DM - low density near the galactic core and increasing with radius.
http://arxiv.org/abs/astro-ph/9710039
http://www.citebase.org/cgi-bin/citations?id=oai:arXiv.org:astro-ph/0108505

Regarding a modification of gravity:

This has been under consideration for a long time (read up on MOND, for example). It doesn't, however, fit the data as well as the dark matter model.

No, MOND doesn't cut it - it is an ad-hoc modification that is predictive on limited scales, but it does suggest that all is not well when we try to apply our laws of gravity to very massive structures. I think that it is instructive to remember that our understanding of gravitation arose from the study of fairly idealized systems, balls rolling down ramps, planets orbiting the sun, suns orbiting one another, etc. Newton's attractive gravitational force and Einstein's curved space-time both work well in these "ideal" domains, but they do not predict the behavior of galaxies accurately, nor can they explain the distribution of tendril-like overdense regions of space discovered by lensing surveys (again neatly attributed to DM).
http://www.cfht.hawaii.edu/News/Lensing/

To date, the DM model has accomplished one thing - quantifying in detail the extent to which Newtonian and GR gravitation fail to predict the behavior of large systems of luminous matter. The process has been to 1) measure the matter shortfall (assuming our understanding of gravitation is correct), 2) attribute the missing mass to an experimentally undetectable entity, and then 3) look for the entity.

 

[turbo]

A very good, and very deep question, turbo-1!

I suspect that discussing it will take us way, way outside modern cosmology, into the heart of what the nature of modern science is, for example. Personally, I'd love to participate in such a discussion (partly because turbo-1 and I have already danced around the edges of just such a discussion, in another forum).

How do folk feel about starting another thread, in PF's Philosophy of Science section?

It is done here:

https://www.physicsforums.com/showthread.php?p=870900#post870900

I have included a couple of simple examples to illustrate how a model with many parameters, even if each is individually at a high confidence level, may prove to be untenable.

 

[SpaceTiger]

Excuse me, but I would think that the non-spherical nature of spiral galaxies would effect the distribution of dark matter halos so that the halo would not be spherical either, right?

That's one of several reasons I said "approximately" spherical. You have to keep in mind that we're seeing something like 10% of the total mass of the galaxy when we look at images of spiral galaxies. Although the visible baryonic component has arranged itself into a disk due to non-gravitational forces, the stuff that we can't see (~90%) would only be slightly perturbed by it in the outer parts of the galaxy, where we're most interested in the properties of the dark matter halo. There's still a lot of uncertainty about the baryon-dominated inner parts (partially because of this "cusp" problem turbo mentions), but even there, it's possible to maintain an approximately spherical halo of stars and dark matter.

 

[Garth]

You have to keep in mind that we're seeing something like 10% of the total mass of the galaxy when we look at images of spiral galaxies. Although the visible baryonic component has arranged itself into a disk due to non-gravitational forces, the stuff that we can't see (~90%) would only be slightly perturbed by it in the outer parts of the galaxy, where we're most interested in the properties of the dark matter halo. .

\Omega_{observed matter}= 0.003, \Omega_baryon = 0.04

ST what are the latest hypotheses about the dark baryonic matter? What form does it take, what equation of state might it have - dust or gas??

Garth

 

[SpaceTiger]

Dynamical simulations have pointed to a cuspy distribution of DM - low density near the galactic core and increasing with radius.

There seems to be a slight misunderstanding of "cusp problem" here. It's not that we're seeing a "void" or "hole" of dark matter at the center of galaxies, it's that it's less dense than predicted by some simulations. It's still more dense than in the outer parts of the galaxy. We already discussed this in a recent thread:

Cuspy Halo Problem

No, MOND doesn't cut it - it is an ad-hoc modification that is predictive on limited scales

It certainly was originally, but recently, it has been put into a more complete relativistic framework by Bekenstein:

http://lanl.arxiv.org/abs/astro-ph/0403694"

Word is that it's failing to fit the third peak of the CMB:

http://lanl.arxiv.org/abs/astro-ph/0508048"

I think that it is instructive to remember that our understanding of gravitation arose from the study of fairly idealized systems, balls rolling down ramps, planets orbiting the sun, suns orbiting one another, etc. Newton's attractive gravitational force and Einstein's curved space-time both work well in these "ideal" domains, but they do not predict the behavior of galaxies accurately, nor can they explain the distribution of tendril-like overdense regions of space discovered by lensing surveys (again neatly attributed to DM).

Actually, this was the whole idea behind MOND, to provide an emperical fit to the galactic/cosmological regime (i.e. small accelerations). Now that it has been extended to a full relativistic theory of gravity, I'm not sure why you think it doesn't fit what you're describing.

To date, the DM model has accomplished one thing - quantifying in detail the extent to which Newtonian and GR gravitation fail to predict the behavior of large systems of luminous matter.

That's incorrect. The dark matter paradigm has contributed to successful predictions about gravitational lensing, the CMB, and large scale structure. It also successfully explains many things we previously knew, such as roughly flat rotation curves and the large velocity dispersions of galaxy clusters.

 

[SpaceTiger]

ST what are the latest hypotheses about the dark baryonic matter? What form does it take, what equation of state might it have - dust or gas??

To my knowledge, most people have put aside dark baryonic matter because it's inconsistent with both WMAP and nucleosynthesis. If one were to ignore those constraints, there would still be some regimes that couldn't be ruled out (like if the dark matter were made of textbook-sized objects), but we've ruled a lot of the things one might naively expect, like large planets or brown dwarfs.

EDIT: Above response is irrelevant to the question. He wasn't asking about having all of the dark matter be baryonic. See below

 

[SpaceTiger]

How do folk feel about starting another thread, in PF's Philosophy of Science section?

It is indeed an interesting question, but if we actually intend to put numbers on this, would it not be more appropriate for the "Set Theory, Logic, Probability & Statistics" forum? I have a feeling like this touches on some of the core issues of the frequentist-bayesian debate. My first impression is that the former would view this as impossible, while the latter would just view it as impractical. The main issue here, I think, is not how we treat the probabilities of A and B once they're computed, but rather how we compute these probabilities in the first place. Assuming A and B are theories (or assertions of theories), then one would almost certainly have to perform Bayesian inference to obtain a probability.

 

[turbo]

Actually, this was the whole idea behind MOND, to provide an emperical fit to the galactic/cosmological regime (i.e. small accelerations). Now that it has been extended to a full relativistic theory of gravity, I'm not sure why you think it doesn't fit what you're describing.

It cannot describe this:

http://www.cfht.hawaii.edu/News/Lensing/#IC

The filamentous structures modeled here from gravitational lensing are indicative that there is something modifying the local optical properties of space. The conventional interpretation is that these filaments show how dark matter is distributed - I suggest that we are seeing the optical effects of vacuum polarization or some very similar process that can modify the electromagnetic properties of these filamentous zones.

That's incorrect. The dark matter paradigm has contributed to successful predictions about gravitational lensing, the CMB, and large scale structure. It also successfully explains many things we previously knew, such as roughly flat rotation curves and the large velocity dispersions of galaxy clusters.

The point is that we know that either gravitational attraction is enhanced in massive domains OR there is more mass there than we can see. Either of these will lead you to infer stronger gravitational lensing than expected in GR given the mass inferred by the luminosity of the observed matter. Either of these will motivate you to infer higher cluster binding energies and modified galactic rotation curves. Coming up with a modified theory of gravitation will be a daunting task, no doubt, but if QT is to be salvaged and united with gravity, our theory of gravity must be reformulated in a language in which QT can be expressed. Right now, they don't communicate at all, and I regard the necessity for DM to be a flaw that indicates that maybe QT is OK, but gravitation needs to be formulated differently. Maybe I spend too much time watching Penrose lectures...

 

[SpaceTiger]

It cannot describe this:
http://www.cfht.hawaii.edu/News/Lensing/#IC

I don't think it's known what relativistic MOND predicts for lensing by large scale structure, so this may be incorrect.

The filamentous structures modeled here from gravitational lensing are indicative that there is something modifying the local optical properties of space. The conventional interpretation is that these filaments show how dark matter is distributed - I suggest that we are seeing the optical effects of vacuum polarization or some very similar process that can modify the electromagnetic properties of these filamentous zones.

This doesn't belong here, you should submit to the Independent Research forum.

The point is that we know that either gravitational attraction is enhanced in massive domains OR there is more mass there than we can see. Either of these will lead you to infer stronger gravitational lensing than expected in GR given the mass inferred by the luminosity of the observed matter. Either of these will motivate you to infer higher cluster binding energies and modified galactic rotation curves.

That's correct, but so far none of the self-consistent alternatives to gravity can successfully reproduce all of the observations quantitatively. It's not enough to simply have qualitative agreement. It might happen in the future, it's hard to say, but until then, dark matter will likely remain the favored theory.

Coming up with a modified theory of gravitation will be a daunting task, no doubt, but if QT is to be salvaged and united with gravity, our theory of gravity must be reformulated in a language in which QT can be expressed. Right now, they don't communicate at all, and I regard the necessity for DM to be a flaw that indicates that maybe QT is OK, but gravitation needs to be formulated differently.

It's possible that the IR limit of quantum gravity could resolve the dark matter problem, but the quantum extension of GR is certainly not necessarily discrepant with Newtonian gravity at galactic scales.

 

[Garth]

To my knowledge, most people have put aside dark baryonic matter because it's inconsistent with both WMAP and nucleosynthesis. If one were to ignore those constraints, there would still be some regimes that couldn't be ruled out (like if the dark matter were made of textbook-sized objects), but we've ruled a lot of the things one might naively expect, like large planets or brown dwarfs.

I think you may have misunderstood me. I am not referring to the Freely Coasting Model and its prediction that all DM is baryonic but to the standard model. If in the standard model \Omega_baryonic = 0.04 but only
\Omega_{observed matter} = 0.003, then, even if there is a relative abundance bias problem, there must be a lot of baryonic matter that is not seen, i.e. 'dark' in some respect. Is it thought that this is in the Galactic halo or IGM or ICM as gas, dust, plasma or what?

Garth

 

[SpaceTiger]

\Omega_{observed matter} = 0.003, then, even if there is a relative abundance bias problem, there must be a lot of baryonic matter that is not seen, i.e. 'dark' in some respect. Is it thought that this is in the Galactic halo or IGM or ICM as gas, dust, plasma or what?

Ok, yes, my apologies. Do you have a reference for the number you're quoting? I agree that there is plenty of unobserved baryonic matter, but I was not under the impression that there was a single concordance number for \Omega_{M, observed}.

 

[Garth]

Ok, yes, my apologies. Do you have a reference for the number you're quoting? I agree that there is plenty of unobserved baryonic matter, but I was not under the impression that there was a single concordance number for \Omega_{M, observed}.

The baryon content of the Universe

We estimate the baryon mass density of the Universe due to the stars in galaxies and the hot gas in clusters and groups of galaxies. The galaxy contribution is computed by using the Efstathiou, Ellis & Peterson luminosity function, togetherwith van der Marel and Persic & Salucci’s mass-to-light versus luminosity relationships. We find stars \Omega_b ~ 0.002. For clusters and groups we use the Edge et al. X -ray luminosity function, and Edge & Stewart and Kriss, Cioffi & Canizares’ (gas mass)-luminosity relations. We find gas \Omega_b ~ 0.001. The total amount of visible baryons is then \Omega_b ~ 0.003, i.e. less than 10 per cent of the lower limit predicted by standard primordial nucleosynthesis, implying that the great majority of baryons in the Universe are unseen.

The baryonic mass function of galaxies

Garth

 

[SpaceTiger]

I think that second abstract answers your question:

Most (∼ 90%) of the baryons in the Universe are not in galaxies. They probably exist in a warm/hot intergalactic medium. Searching for direct observational evidence and deeper theoretical understanding for this will form one of the major challenges for astronomy in the next decade.

The number you quote is actually \Omega_{b,g}, the total baryonic content of galaxies. The reason I say there isn't a concordance value for \Omega_{M,observed}, or equivalently, \Omega_{b, observed}, is that a large fraction of the baryonic content of the universe is thought to lie in gas in between galaxies. There have been some crude measurements of this that indicate that most of the remaining baryonic mass lies in the WHIM (Warm-Hot Intergalactic Medium). Here's one such reference:

http://lanl.arxiv.org/abs/astro-ph/0501126"

 

[Nereid]

It is indeed an interesting question, but if we actually intend to put numbers on this, would it not be more appropriate for the "Set Theory, Logic, Probability & Statistics" forum? I have a feeling like this touches on some of the core issues of the frequentist-bayesian debate. My first impression is that the former would view this as impossible, while the latter would just view it as impractical. The main issue here, I think, is not how we treat the probabilities of A and B once they're computed, but rather how we compute these probabilities in the first place. Assuming A and B are theories (or assertions of theories), then one would almost certainly have to perform Bayesian inference to obtain a probability.

I feel we should certainly examine these aspects - a good discussion of the frequentist-bayesian divide is long overdue, IMHO.

However, having debated turbo-1 http://www.bautforum.com/showthread.php?t=35505", on a closely related topic, I reckon a discussion on just the apparent content of turbo-1's post would quickly uncover a much deeper gulf, which is (basically) quite different understandings of the nature of modern cosmology (for which turbo-1's new thread is in the right place).

 

[Chronos]

I've waded through all the calculations and am convinced that the 'dark baryonic' proposition is ruled out [at least to the 4 sigma level]. It's absolutely impossible to create that much baryonic matter in a 'Big Bang'. Worse yet, an entirely baryonic universe is ruled out at better than 3 sigma if you apply the same parameters to that calculation. So the option is... an infinitely old universe with no 'Big Bang'. An unlikely option, IMO. That annoying CMB rules that model out at about... 6 sigma.