Age of the Universe - is it meaningful?

[Saddlestone]

The age of the universe is equated to the age of the oldest objects in it - and estimated at about 13.75 billion years. But in whose frame of reference is this measurement valid? Surely the oldest objects have spent much of their "lives" either under intense gravitational fields, or traveling at very close to the speed of light, or both, and therefore their experience of time is different to younger objects.

Is it valid to give an age to a diverse set of objects (ie the universe), all of which have different "perceptions" of time, depending on their densities and speeds since they were created?

 

[mathman]

I presume the frame of reference is ours on earth.

 

[marcus]

The age of the universe is equated to the age of the oldest objects in it ...?

No that is not how the age is defined. The 13.7 billion years you hear about as the age is not the age of any object one can point to. It is the age of the expansion process.

The expansion started with a bounce according to some models and in those models the age (13.7 billion) is the TIME SINCE THE BOUNCE. In models without a bounce there is some other way of marking the start of expansion. In all cases the age is the time since start of expansion.

Time is understood to be cosmologist's standard universe time, that the conventional Friedmann model runs on.

This time is that perceived by observers anywhere in the U who are AT REST with respect to the ancient light (CMB). That is the same as being at rest relative to the uniform average distribution of matter. Or relative to the expansion process itself. Or at rest with respect to ancient matter----the hot approximately uniform gas that emitted the CMB. The ancient matter BEFORE it began seriously to condense and fall together into clumps. The CMB is a picture of that ancient uniform hot gas world, so it is a good restframe definer.

CMB REST frame is a beautiful and simple idea that is is used throughout cosmology and you keep seeing it cropping up. An observer is at CMB rest if he does not see any Doppler hotspot/coldspot dipole caused by his motion relative to CMB.

When we map the CMB we see a Doppler dipole which is then removed from the data so that we get the data AS IF observed by someone at CMB rest.
Otherwise the dipole would overwhelm thousand-fold the delicate temp fluctuations which we are interested in.

CMB REST is what let's us define universe standard time. Because observers all over the universe would, if they are at rest, be able to synchronize their clocks. The only problem would be somehow compensating for some observers being deeper in gravity wells. Let's just assume they are in intergalactic space and the effect is negligible.

If you are not at CMB rest, then you are not at rest relative to the expansion process itself. The galaxies ahead of you will seem to be receding slower than they should and those behind will be receding faster. The Hubble distance expansion rate will not be the same in all directions!

Uniformly (on average) distributed matter introduces a concept of absolute rest which is not available in its absence, in bare GR.

 

[marcus]

In practice you just fit the Friedman equation model to the data and the Friedman model tells you the time since start of expansion. That's where the 13.7 comes from.

The Friedman model runs on standard universe time (not the time as seen by one particular planet, which may have had an obscure or checkered past ).

So what people are quoting when they say 13.7 or 13.76 or whatever is actually a figure in universe standard time terms, as would be measured by an observer who had always been at rest and who is not too deep in anyone clump's gravity well.

A perfect Friedman model observer would, I guess, see matter as homogeneous and isotropic---that is to say, unclumpy or effectively "spaced out". Then his clock would keep time with the model itself.

 

[Saddlestone]

OK maybe I can accept that definition now that the universe is big and relatively old and the concept of ancient light may be valid.

How about when the universe was very young, say less than 10 seconds old, and very dense and hot, and light itself didn't exist. If you look for example on Wikipaedia, it states: Inflationary Epoch: Unknown duration, ending 10–32 seconds after the Big Bang. How is this time period defined?

Presumably as the very young universe was expanding, space-time itself was being created at the very edge of the expansion and so there is no way of being able to define time outside the expansion process itself.

 

[marcus]

OK maybe I can accept that definition now that the universe is big and relatively old and the concept of ancient light may be valid.
...

That seems like a reasonable attitude. The period where there is ancient light goes back to about 380,000 years from the start of expansion.

The Friedman equation which is our model still continues to apply without change for a while as we go back earlier. But we have no light to observe to let us check the model! For quite a while back before year 380,000 it is just physics as usual. Rarified hot hydrogen gas. We know what that is like. Temperatures of 4, 5, 6000 degrees kelvin. Glowing too much to be properly transparent. All the light that gets to us comes from 380,000 or later.

There would have been plenty of light earlier, dazzingly bright. But we don't see that light because in those days the gas was too hot to let the light pass undisturbed. You probably know about this. Conditions like inside a very hot furnace or on the surface of a star. Well studied and well understood.

So we can run the Friedman model back in time and describe what it says, based on the physics we know from laboratory and other tests. But we can't verify.

Well, there are some confirming evidences, like the relative abundance of isotopes of various elements, but it is different because we can't actually SEE.

I think of fairly conventional timekeeping, in a universe we can see, as only extending back to around year 380,000 of the expansion. So there could be an uncertainty of on the order of 1000 years, or even 10,000 years. That doesn't seem like much uncertainty to me, when we are talking billions of years! So I don't worry about it.

Maybe someone else would like to respond to what you have to say about fractions of a second.
AFAICS that is just extrapolation based on the best physics we have so far. Inflation scenarios are guesswork. There is a lot more to be learned about the very very early period of expansion.

Good luck researching it!

 

[cepheid]

Isn't there a cosmology FAQ on this one?

EDIT: not exactly, but it's sort of mentioned here:
https://www.physicsforums.com/showthread.php?t=506989

and kind of alluded to here:
https://www.physicsforums.com/showthread.php?t=506990

 

[Saddlestone]

I guess what concerns me is that statements are made about events happening at certain times in the early universe without reference to the observer who is measuring the time. Einstein would turn in his grave over such sloppiness!

On a slightly different point, which Einstein would have found to be a key question: when did "c" as the natural speed limit in the universe come into existence? Does the speed limit rely on the existence of photons for it to be valid? Ie before photons existed, could matter move around at any speed the expansion drove it to, and without any relativistic effects?

If this is so, then the natural follow on question is: once "c" came into existence as a speed limit, what happened to all the matter (probably most/all of the universe) already moving at greater than c? Maybe this should be a different post subject?

 

[Cosmo Novice]

I guess what concerns me is that statements are made about events happening at certain times in the early universe without reference to the observer who is measuring the time. Einstein would turn in his grave over such sloppiness!

On a slightly different point, which Einstein would have found to be a key question: when did "c" as the natural speed limit in the universe come into existence? Does the speed limit rely on the existence of photons for it to be valid? Ie before photons existed, could matter move around at any speed the expansion drove it to, and without any relativistic effects?

If this is so, then the natural follow on question is: once "c" came into existence as a speed limit, what happened to all the matter (probably most/all of the universe) already moving at greater than c? Maybe this should be a different post subject?

Objects can move away from us faster than c due to expansion, this is not contradictory of GR. Expansion is the geometric increase in distances between gravitationally bound regions of spacetime, acceleration can exceed c without breaking any cosmic laws.

 

[Saddlestone]

I'm thinking of the natural speed limit of say a rocket leaving earth, which is limited to c.

When in the development of the universe did this speed restriction come about? Has it always been limited to the speed of light and does it depend on light itself (ie photons) existing?

Has c always been constant?

 

[phinds]

First, your references to "the speed of light" are slightly misleading. Light does not set its own speed limit, it follows the "universal speed limit" which is of necessity followed by ANY massless object, which of course happens to include photons.

I believe that as far as is known, that speed limit came into being at the start of the universe along with all other physical laws. It did NOT come into being at some later time.

As saddlestone already pointed out, the objects that appear to be moving faster than c are moving away from each other due to the expansion of the universe and this does not violate the speed limit since the speed limit doesn't apply to the expansion.

 

[D H]

I guess what concerns me is that statements are made about events happening at certain times in the early universe without reference to the observer who is measuring the time. Einstein would turn in his grave over such sloppiness!

It's not sloppiness. It is mostly superfluous information.

It is superfluous information to the public at large who don't know that time is not universal, and that even the concepts of past, present, and future can become a bit muddied. The author of any science article aimed at the lay audience has to walk a delicate line between over-simplifying and presenting too much information. The picture presented will always be incomplete; the only way to obtain a complete understanding is to get a PhD in the subject matter of the article.

It is also superfluous information to cosmologists who understand these complexities and have developed an agreed-upon concept of how to measure cosmological time.

It is only important to people who know enough about relativity and what it does to time to ask "in what frame?"

 

[Schulz]

OK maybe I can accept that definition now that the universe is big and relatively old and the concept of ancient light may be valid.

How about when the universe was very young, say less than 10 seconds old, and very dense and hot, and light itself didn't exist. If you look for example on Wikipaedia, it states: Inflationary Epoch: Unknown duration, ending 10–32 seconds after the Big Bang. How is this time period defined?

Presumably as the very young universe was expanding, space-time itself was being created at the very edge of the expansion and so there is no way of being able to define time outside the expansion process itself.

Learning a lot from this thread.

 

[rop]

Just wondering...

If you accept the Dark-Energy/Accelerating-Universe hypothesis, shouldn't the "Age of the Universe" be revised to a much older age than before?

I mean, "counting backwards"... if the expansion-speed is accelerating into the future, it seems the "implosion-speed" going back into the past will decrease correspondingly, and it will therefore be much longer time into the past before you reach the Big-Bang-singularity...?

Right or wrong?

 

[phinds]

I don't follow your logic, but clearly it's giving you the wrong answer so I'd hazzard a guess that it must be wrong.

Perhaps it would clarify things for you to know that the acceleration only started about 6 to 8 billion years ago when matter got spread out enough for dark energy to overcome gravity.

 

[cepheid]

Just wondering...

If you accept the Dark-Energy/Accelerating-Universe hypothesis, shouldn't the "Age of the Universe" be revised to a much older age than before?

I mean, "counting backwards"... if the expansion-speed is accelerating into the future, it seems the "implosion-speed" going back into the past will decrease correspondingly, and it will therefore be much longer time into the past before you reach the Big-Bang-singularity...?

Right or wrong?

The age of the universe in models with dark energy/a cosmological constant is indeed larger than the age in models without (this depends on what specific values you enter in for various cosmological parameters in the model). However, the current value does not need "revision", because it is already based on the best estimates of the values of cosmological parameters like Ω_m and Ω_Λ from experiments like WMAP that made their observations well after the EDIT: [STRIKE]discovery[/STRIKE] first observational evidence of the accelerated expansion. It's important to realize that parameters like the above are what these experiments are able to measure somewhat "directly", and that other things like the age of the universe are derived parameters.

Putting it another way, the age of the universe is model-dependent, and the current estimate of 13.7 Gyr is based on a model that already incorporates dark energy (since the values of the model parameters are from the WMAP best estimates, which strongly support a non-zero Ω_Λ

 

[twofish-quant]

The age of the universe is equated to the age of the oldest objects in it - and estimated at about 13.75 billion years. But in whose frame of reference is this measurement valid?

You can define a general frame of reference consisting of a clock that is distance from any strong gravitation fields.

Surely the oldest objects have spent much of their "lives" either under intense gravitational fields, or traveling at very close to the speed of light, or both, and therefore their experience of time is different to younger objects.

Nope. You need really, really strong gravitational fields to start having noticeable clock differences, and most ordinary stars don't have fields that are strong enough to make a difference. Likewise, the stuff that travels close to the speed of light is usually very light, (i.e. particles instead of stars).

Is it valid to give an age to a diverse set of objects (ie the universe), all of which have different "perceptions" of time, depending on their densities and speeds since they were created?

Depends on how precise you want to be. If you want to measure in milliseconds, then keep track of different time measures turns out to be really hard. If you are measuring things to +/- 100 million years, most people think that the effects are too small to matter, although there is one scientist that thinks that these sorts of effects will give the illusion of an accelerating universe.

 

[twofish-quant]

How about when the universe was very young, say less than 10 seconds old, and very dense and hot

10 seconds is still in "familar" territory. At 10 seconds, you have protons and neutrons.

If you look for example on Wikipaedia, it states: Inflationary Epoch: Unknown duration, ending 10–32 seconds after the Big Bang. How is this time period defined?

You imagine the expansion of the universe as a movie, you run the movie backwards. If nothing weird happens then you can calculate a "time zero." The problem is that it's certain that something weird happens at time zero. In the inflationary era, we still can talk about space and time. At "Planck's time" we *know* that are theories break down.

 

[twofish-quant]

I guess what concerns me is that statements are made about events happening at certain times in the early universe without reference to the observer who is measuring the time. Einstein would turn in his grave over such sloppiness!

You can calculate proper time from an assumed zero. It turns out that observer effects are not important when you try to match things to observations so it doesn't much matter right now.

 

[rop]

The age of the universe in models with dark energy/a cosmological constant is indeed larger than the age in models without (this depends on what specific values you enter in for various cosmological parameters in the model). However, the current value does not need "revision", because it is already based on the best estimates of the values of cosmological parameters like Ω_m and Ω_Λ from experiments like WMAP that made their observations well after the EDIT: [STRIKE]discovery[/STRIKE] first observational evidence of the accelerated expansion. It's important to realize that parameters like the above are what these experiments are able to measure somewhat "directly", and that other things like the age of the universe are derived parameters.

Putting it another way, the age of the universe is model-dependent, and the current estimate of 13.7 Gyr is based on a model that already incorporates dark energy (since the values of the model parameters are from the WMAP best estimates, which strongly support a non-zero Ω_Λ

Thank you Cepheid -- interesting to hear that...

Do you know, what was the earlier estimate? And what year came the new estimate approx?

 

[Saddlestone]

Depends on how precise you want to be. If you want to measure in milliseconds, then keep track of different time measures turns out to be really hard. If you are measuring things to +/- 100 million years, most people think that the effects are too small to matter, although there is one scientist that thinks that these sorts of effects will give the illusion of an accelerating universe.

The significant events in the early universe, eg see:
http://en.wikipedia.org/wiki/Timeline_of_the_Big_Bang
are measured in very small intervals. For example, at 10>-32 seconds after the big bang the Inflationary Epoch era ended, then the Quark Epoch lasted from 10>-12 to 10>-6 seconds, etc and therefore events aren't being measured +/- 100Million years or even milliseconds, but apparently to great precision. I'm assuming Wikipedia is taking these timings from reputable sources.

How accurate can these timings be if the universe is incredibly dense at this time, and hence under the effects of really strong gravitational fields?

 

[yogi]

No that is not how the age is defined. The 13.7 billion years you hear about as the age is not the age of any object one can point to. It is the age of the expansion process.

The expansion started with a bounce according to some models and in those models the age (13.7 billion) is the TIME SINCE THE BOUNCE. In models without a bounce there is some other way of marking the start of expansion. In all cases the age is the time since start of expansion.

Time is understood to be cosmologist's standard universe time, that the conventional Friedmann model runs on.

This time is that perceived by observers anywhere in the U who are AT REST with respect to the ancient light (CMB). That is the same as being at rest relative to the uniform average distribution of matter. Or relative to the expansion process itself. Or at rest with respect to ancient matter----the hot approximately uniform gas that emitted the CMB.

The ancient matter BEFORE it began seriously to condense and fall together into clumps. The CMB is a picture of that ancient uniform hot gas world, so it is a good restframe definer.

CMB REST frame is a beautiful and simple idea that is is used throughout cosmology and you keep seeing it cropping up. An observer is at CMB rest if he does not see any Doppler hotspot/coldspot dipole caused by his motion relative to CMB.

When we map the CMB we see a Doppler dipole which is then removed from the data so that we get the data AS IF observed by someone at CMB rest.
Otherwise the dipole would overwhelm thousand-fold the delicate temp fluctuations which we are interested in.

CMB REST is what let's us define universe standard time. Because observers all over the universe would, if they are at rest, be able to synchronize their clocks. The only problem would be somehow compensating for some observers being deeper in gravity wells. Let's just assume they are in intergalactic space and the effect is negligible.

If you are not at CMB rest, then you are not at rest relative to the expansion process itself. The galaxies ahead of you will seem to be receding slower than they should and those behind will be receding faster. The Hubble distance expansion rate will not be the same in all directions!

Uniformly (on average) distributed matter introduces a concept of absolute rest which is not available in its absence, in bare GR.

Marcus, your posts are the best tutorials on the interent

 

[Imax]

Some models of the universe look at time as we see it here on Earth and assume that time is linear. Observations of distant galaxies indicate that the universe is expanding and that expansion is accelerating. Assuming time is linear and using observations, you can extrapolate backwards in time to the Big Bang, some 13.75 billion years ago.