What is the current consensus on the fate of the universe?

[Gold Barz]

I need to get up to date on the current consensus on the likely fate of the universe? Is it heat death, big crunch, big freeze, big bounce (didn't this get disproved a couple of years ago?)

 

[bapowell]

Currently the universe is undergoing accelerated expansion. If we were to extrapolate this as a constant rate into the future, I suppose heat death would be the ultimate fate (how's that different than big freeze?). However, we are not too certain of the rate of change of this acceleration -- if the acceleration is increasing in magnitude, then the universe ends in a Big Rip -- in which all gravitationally bound objects eventually get torn apart by the expansion.

A big crunch and a big bounce are essentially the same thing, since most theories of quantum gravity give a bounce following a contracting phase of the universe; however, those models are speculative.

Personally, I caution against extrapolating current rates of expansion far into the future. If the energy component driving the accelerated expansion today begins to decay at some point in the future then we will return to a matter dominated universe, and then our fate might be different than a simple extrapolation might suggest.

Since these are all extrapolations, no possibility is currently excluded by observational data.

 

[Chalnoth]

Heat death is currently the most likely result.

The 'big rip' scenario has tremendous theoretical difficulties, and is highly unlikely. Bounce/crunch models would require very special behavior for dark energy, as well as slightly positive spatial curvature, and are also unlikely.

 

[bapowell]

Heat death is currently the most likely result.

The 'big rip' scenario has tremendous theoretical difficulties, and is highly unlikely. Bounce/crunch models would require very special behavior for dark energy, as well as slightly positive spatial curvature, and are also unlikely.

Sure, but all that is certainly debatable. From a purely observational perspective, all are still in the running. I would not equate "tremendous theoretical difficulties" with "highly unlikely" since that implies that you are happy with the current state of gravity theory, which is widely perceived to be incomplete. With regards to bounce/crunch, what very special behavior is required? I would think that as long as the energy decays (think rolling scalar field) then a big crunch/bounce is likely. Yes, it requires positive spatial curvature, but it is just as easy to believe that the universe is positively curved as it is perfectly flat. I also don't see how "very special behavior" equates to "unlikely". For example, the inflaton potential is terribly fine-tuned and the initial conditions perhaps even more so. That's very special behavior but few people argue that inflation is unlikely to have occurred.

 

[Chalnoth]

Sure, but all that is certainly debatable. From a purely observational perspective, all are still in the running.

Since we're talking about the ultimate fate of the universe, they always will be. The current evidence, as it stands, still points towards heat death. Perhaps that will change in the future, but that's the way things stand now.

I would not equate "tremendous theoretical difficulties" with "highly unlikely" since that implies that you are happy with the current state of gravity theory, which is widely perceived to be incomplete.

It would require violation of the weak energy condition (that matter energy is always non-negative), which most consider to be impossible, especially on cosmological scales where whatever microscopic physics produce gravity are likely to be less relevant.

With regards to bounce/crunch, what very special behavior is required?

Dark energy, so far, has been decaying very slowly or not at all. A bounce/crunch would require that at some point in the future, this behavior changes so that it starts decaying more rapidly than $1/a^2$. I'm not aware that any models do this.

For example, the inflaton potential is terribly fine-tuned and the initial conditions perhaps even more so. That's very special behavior but few people argue that inflation is unlikely to have occurred.

The difference is that we have strong observational reasons to believe inflation occurred.

 

[bapowell]

Since we're talking about the ultimate fate of the universe, they always will be. The current evidence, as it stands, still points towards heat death. Perhaps that will change in the future, but that's the way things stand now.

OK, I guess I just disagree with this statement. As far as evidence is concerned, baring any arguments regarding the theoretical viability of phantom energy, a Big Rip is still in agreement with data ($$w < -1.10 \pm 0.14$$ at 68% CL from WMAP7). Also, since we have no constraints on $$\dot{w}$$, I would suggest that, observationally, big crunch models are also still OK. I am certainly not an advocate for either of these models, but strictly speaking, observational evidence does not point to heat death over other alternatives.

It would require violation of the weak energy condition (that matter energy is always non-negative), which most consider to be impossible, especially on cosmological scales where whatever microscopic physics produce gravity are likely to be less relevant.

This is certainly true, and I agree that phantom models face severe theoretical difficulties. However, if $w$ turns out be < -1, then we will need to start taking this more seriously. It's true that phantom quantum matter is energetically unstable and this is a problem; however, I think it's probably OK to violate the weak energy condition effectively, i.e. no fundamental fields in the theory violate the condition. Based on what you say at the end regarding the fact that inflation is 'likely' because we have lots of evidence for it, and therefore we must accept its theoretical difficulties, you would need to reconsider your argument against phantom models if, for example, Planck reliably constrained $w < -1$.

Dark energy, so far, has been decaying very slowly or not at all. A bounce/crunch would require that at some point in the future, this behavior changes so that it starts decaying more rapidly than $1/a^2$. I'm not aware that any models do this.

Why can't you? Is there a theoretical reason? I'm no expert on dark energy, but I would think it would be simply a matter of 'designing' the right potential that decays sufficiently fast at some point in the future.

The difference is that we have strong observational reasons to believe inflation occurred.

OK, but that seems kind of like an a posteriori qualification for what makes a theory 'likely'.

 

[Chalnoth]

OK, I guess I just disagree with this statement. As far as evidence is concerned, baring any arguments regarding the theoretical viability of phantom energy, a Big Rip is still in agreement with data ($$w < -1.10 \pm 0.14$$ at 68% CL from WMAP7).

Interacting dark energy models can provide apparent $w < -1$, but do not lead to a big rip scenario. They are the only somewhat realistic physical models which have been shown to produce a measured $w < -1$. So if anything, this might be evidence for interacting dark energy. But since a one sigma deviation from -1 is the expected deviation, and since that deviation could, by random chance, fall on either side of $w = -1$, there really isn't any evidence of $w < -1$.

As for why the weak energy condition is almost certainly true, see here:
http://authors.library.caltech.edu/9262/1/MORprl88.pdf

Basically, if the weak energy condition doesn't hold, then it's possible to make time machines, which lead to contradictions. There's also the thing you pointed out with phantoms making the vacuum unstable (meaning big boom).

Why can't you? Is there a theoretical reason? I'm no expert on dark energy, but I would think it would be simply a matter of 'designing' the right potential that decays sufficiently fast at some point in the future.

Well, what would cause the decay? The longer that the dark energy acts like a cosmological constant, the less the universe changes going into the future.

 

[Chronos]

Keep in mind the universe experienced an extreme inflationary epoch immediately following the big event. It died out rather quickly. It appears we are now experiencing an 'aftershock'. My hunch is it too will die out.

 

[Chalnoth]

Keep in mind the universe experienced an extreme inflationary epoch immediately following the big event. It died out rather quickly. It appears we are now experiencing an 'aftershock'. My hunch is it too will die out.

Pretty sure that doesn't work. Inflation depends upon "slow roll" which requires that the expansion of the universe put sufficient friction on the scalar field so that it takes a while to reach its potential minimum. If I remember correctly, the parameters are all wrong for the same sort of thing to be occurring now.

 

[bapowell]

Pretty sure that doesn't work. Inflation depends upon "slow roll" which requires that the expansion of the universe put sufficient friction on the scalar field so that it takes a while to reach its potential minimum. If I remember correctly, the parameters are all wrong for the same sort of thing to be occurring now.

Why? Slow-roll mimics de Sitter expansion. What is needed for late-time acceleration is a potential that is also slowly rolling, but even more so than during primordial inflation (since the epoch of late-time acceleration has lasted so much longer than the epoch of primordial inflation.) I'm not sure what parameters you're referring to.

With regards to your earlier question about what would make such a scalar field decay nowadays -- why not the same thing that made the inflaton decay? Perhaps 'decay' is a poor choice of words -- all we really need is for the late-time scalar to lose potential energy. This will occur for any suitable potential that is monotonically decreasing with the expansion of the universe. In other words, to get late time acceleration that is asymptotically approaching a matter dominated universe, I should be able to pick my favorite inflationary potential with a true vacuum minimum, adjust its magnitude and flatten it out sufficiently so that it came to dominate the energy density of the universe 5 billion years ago, and press go.

 

[Chalnoth]

Why? Slow-roll mimics de Sitter expansion. What is needed for late-time acceleration is a potential that is also slowly rolling, but even more so than during primordial inflation (since the epoch of late-time acceleration has lasted so much longer than the epoch of primordial inflation.) I'm not sure what parameters you're referring to.

Well, I suppose you could do it, but it would require a completely different potential energy than the one that drove inflation. The expansion rate has just been too slow to retard the decay of an inflaton. So it might be similar to inflation, but it wouldn't appear to be directly related to it.

With regards to your earlier question about what would make such a scalar field decay nowadays -- why not the same thing that made the inflaton decay? Perhaps 'decay' is a poor choice of words -- all we really need is for the late-time scalar to lose potential energy. This will occur for any suitable potential that is monotonically decreasing with the expansion of the universe. In other words, to get late time acceleration that is asymptotically approaching a matter dominated universe, I should be able to pick my favorite inflationary potential with a true vacuum minimum, adjust its magnitude and flatten it out sufficiently so that it came to dominate the energy density of the universe 5 billion years ago, and press go.

Well, first you're making the assumption that a "true vacuum" has zero energy, which we certainly don't know. There is no known symmetry that sets the vacuum energy to zero.

But it really has to lose potential energy quickly, not simply continue its slow roll trajectory. This means that it would have to, for instance, oscillate about the minimum of its potential (as inflaton did), producing radiation and other stuff.

 

[bapowell]

Well, I suppose you could do it, but it would require a completely different potential energy than the one that drove inflation. The expansion rate has just been too slow to retard the decay of an inflaton. So it might be similar to inflation, but it wouldn't appear to be directly related to it.

Agreed.

Well, first you're making the assumption that a "true vacuum" has zero energy, which we certainly don't know. There is no known symmetry that sets the vacuum energy to zero.

I agree there's no guarantee that the true vacuum has zero energy. But I'm just proposing a model for how you would go about getting a big crunch given that we are currently undergoing accelerated expansion. If the true vacuum has some energy (the CC) then we end up with an asymptotically de Sitter phase rather than a matter dominated one, and we get no big crunch. Of course, this could be the case -- there could be a CC afterall -- and then we get heat death.

But it really has to lose potential energy quickly, not simply continue its slow roll trajectory. This means that it would have to, for instance, oscillate about the minimum of its potential (as inflaton did), producing radiation and other stuff.

This is where I think we are missing each other. Why does the potential energy need to decay quickly? I would think, given that the current epoch of accelerated expansion has been underway for the last 5 billion years, that the rate of decay should be painstakingly slow (relative to the rate of the decay during inflation, which lasted fractions of a second.) It is true that the Hubble drag is much more substantial during inflation than it would be today, but then we just need an even flatter potential to prolong the accelerated expansion. Granted, this would be a terribly fine-tuned monstrosity, and effectively a CC during today's epoch. But we don't even know what made the inflaton as flat as it needed to be, and so I think we are missing an important part of the puzzle when it comes to cosmological scalar fields.

 

[Chalnoth]

This is where I think we are missing each other. Why does the potential energy need to decay quickly?

To get the accelerated expansion to slow, you need the energy density to decay more rapidly than $1/a^2$. Fail to do this, and the expansion will continue to accelerate.

 

[bapowell]

To get the accelerated expansion to slow, you need the energy density to decay more rapidly than $1/a^2$. Fail to do this, and the expansion will continue to accelerate.

Or you just need a potential that goes to zero at finite time. Am I missing something?

 

[Chalnoth]

Or you just need a potential that goes to zero at finite time. Am I missing something?

If the total energy density doesn't decay more rapidly than $1/a^2$, then it will never reach zero.

 

[Smock]

So are you saying that it is the energy from the Big Bang speeding up the universe or are you saying that there is dark energy pushing the universe apart? I think dark energy speeding up the universe would make the most sense. It makes more sense to me than the universe expanding quickly then slowing down and then speeding back up.