Doesn't there have to be more than one observable universe?

[trainman2001]

This was a thought provoking thread for me. I hadn't thought about the fact that if we see the image of a Quasar at the edge of our observable distance and it's light has taken 13.7 billion years to get to us, we're seeing it in a position where it was 13.7 billion years ago. That means it has been traveling outwards during that 13.7 billion years and is significantly farther away. How far away is it? Was it's speed linear? According to current thought, it's speeding up. And it also means that there are probably objects red-shifted to invisibility beyond that distance that are moving even further. It's why they've adjusted the estimate of galaxies in what we perceive as our Universe from 500 billion or so to 2 or more trillions simply based on those Hubble deep field pictures and what they implied about what was not seen beyond those distant images.

 

[kimbyd]

This was a thought provoking thread for me. I hadn't thought about the fact that if we see the image of a Quasar at the edge of our observable distance and it's light has taken 13.7 billion years to get to us, we're seeing it in a position where it was 13.7 billion years ago. That means it has been traveling outwards during that 13.7 billion years and is significantly farther away. How far away is it? Was it's speed linear? According to current thought, it's speeding up. And it also means that there are probably objects red-shifted to invisibility beyond that distance that are moving even further. It's why they've adjusted the estimate of galaxies in what we perceive as our Universe from 500 billion or so to 2 or more trillions simply based on those Hubble deep field pictures and what they implied about what was not seen beyond those distant images.

Ned Wright's cosmology calculator can be useful here:
http://www.astro.ucla.edu/~wright/CosmoCalc.html

13.7 billion years ago is a bit too far: in the very early universe, there were no stars at all.

The furthest quasars are at around a redshift of $z = 10$ or so. Based on that calculator, the light travel time was 13.2 billion years. The distance when the light was emitted (technical term: angular size distance) was around 2.9 billion light years. The distance today (technical term: comoving radial distance) is about 31.4 billion light years.

 

[Chronos]

Keep in mind it took light 13.7 billion years to reach us from the 'edge' of our observable universe so any images you view of it now are merely a snapshot from the past. If you tried to go there now it would take you at least 13.7 billion years to get there and the universe would continue to age. Yes, that means you cannot get there from here. You can go to the hospital you were born at, but, not in time to witness your own birth.

 

[Mr Wolf]

It isn't. Try it this way: where is the edge of the Earth's surface?

Well, I guess I get what your point is: that is, if I move along the surface of a sphere I don't actually reach an edge.
As far as I've read and understood (I studied Physics many years ago, but not Cosmology and Astrophysics), there's not an empty space in which the Universe expands, but space-time is "created" as the Universe expands. However, talking about these things by words and trying to visualize them is pretty difficult.

EDIT: I still keep reading 'edge' of visible Universe, and talking about the furthest objects away (13.7 billions light years). So, if there's such a distance, what does it mean?

 

[PeterDonis]

if I move along the surface of a sphere I don't actually reach an edge

Yes, exactly. A 2-sphere is a compact manifold (has a finite area) without boundary (no edge). By contrast, an ordinary piece of paper (if we idealize it as being perfectly 2-dimensional) is a compact manifold (has a finite area) with boundary (it has an edge).

there's not an empty space in which the Universe expands

Yes.

space-time is "created" as the Universe expands

No. Spacetime is a 4-dimensional geometry. It doesn't change, it doesn't get created or destroyed, it just is. The shape of the 4-dimensional geometry that describes our universe just happens to have a particular set of 3-dimensional spacelike slices in it that have an increasing spatial scale factor in the future time direction.

I still keep reading 'edge' of visible Universe

That's the distance of the farthest objects that we can see. We can only see a finite distance away because the universe has a finite age, so light has only had a finite time to travel to us from distant objects.

the furthest objects away (13.7 billions light years).

The most distant objects whose light is just reaching us now, i.e., which are at the edge of our observable universe, are more than 13.7 billion light years away, because of the geometry of spacetime. (The usual way of saying this is that the universe has been expanding while the light has been traveling, but that way of putting it can be confusing because it invites incorrect inferences like the one I corrected above.) The distance "now" of objects whose light is just reaching us now is about 47 billion light years. But the light we are seeing from those objects was emitted about 13.7 billion years ago, so the universe was much, much smaller then and the light did not cover 47 billion light-years to get to us. (Nor did it cover 13.7 billion light-years. In a curved spacetime geometry you can't use the usual special relativity intuitions about the connection between distance and light travel time.)

 

[Imager]

Again, there's as many observable universes as there are observers.

A couple of question to check if I am understanding what is being said:

Assuming an infinite universe, is there any upper limit to the number of observable universes? I say there isn't. The way I am thinking there would be an infinite amount 92 billion LY bubbles across the universe.

Assuming a finite universe, wouldn't the observable universes be forced to overlap each other? I say yes.

And a thank you to Brandersnatch for the many great answers he has given over the years!

 

[kimbyd]

Assuming an infinite universe, is there any upper limit to the number of observable universes?

Maybe, though it depends a bit upon what you mean.

In the quantum view, the number of non-identical observable universes may be finite (but excessively large). For instance, if the cosmological constant really is a constant, then there is a finite number of possible observable universes based upon the horizon size set by that cosmological constant. If the cosmological constant can actually take some finite number of values values, then you get a finite number of possible observable universes for each of those values, for a total number that remains finite. Though this may only work if $\Lambda=0$ is not possible (I think $\Lambda=0$ may be infinite, even in the quantum case, while I'm pretty sure any other values are finite, with the caveat that I'm not certain about negative values).

In the non-quantum view, the number is definitely unlimited (because any observer location will result in a different observable universe, and in the non-quantum view there are an infinite number of observer locations even within a finite region). This actually leads to problems with theoretical predictions: you can't calculate certain kinds of probabilities in an infinite universe (this is known as the "measure problem").

Assuming a finite universe, wouldn't the observable universes be forced to overlap each other? I say yes.

No need to assume a finite universe. Any observer within our observable universe will have its own observable universe which overlaps our own.

The question, rather, is whether there can be any regions which are entirely disconnected, where there can be no overlap between the two regions. That is unknown. It depends entirely upon the physics which resulted in the low-entropy state which occurred early in our observable universe.

 

[Bandersnatch]

The question, rather, is whether there can be any regions which are entirely disconnected, where there can be no overlap between the two regions. That is unknown.

Isn't this pretty much a given in the concordance model? I always thought these entirely causally disconnected observers are spaced every ~130 Gly or so (twice the comoving event horizon distance).

 

[kimbyd]

Isn't this pretty much a given in the concordance model? I always thought these entirely causally disconnected observers are spaced every ~130 Gly or so (twice the comoving event horizon distance).

What I meant is whether or not there can be any distinct regions (not necessarily just observable regions) which have no overlap anywhere.

Certainly it's possible for there to be non-overlapping observable regions in the concordance model, just by being far enough away from one another. But you could, at least as a mental exercise, connect any two observable regions by selecting a set of observers between the two regions, where each observer overlaps with the other, eventually reaching the far-away observer.

My question is: is there a way for that not to be possible? Can there be regions entirely disconnected?

The reason why the physics of the early universe may hold the answer to this question, even though it can never be tested directly, is that many models demand exactly this. For instance, if the start of our universe was a quantum vacuum fluctuation in another region of space-time, observers in that space-time would see what appears to be a microscopic black hole which pops into existence then rapidly evaporates, while observers within our space-time see billions of years of history. These two space-times are entirely disconnected, and there is no way to bridge the two.

Right now we have no way to know if this is possible or not. And we don't know if we'll ever be able to learn enough about the universe to say whether or not this is possible (since it is, by construction, impossible to directly test).

 

[Chronos]

No observer that resides in a region of space visible to us can ever report 'seeing' anything that is not also visible to us. By the time a message about a 'new' event can reach any other observer, the light from that event will also have reached them, no matter where in space it occurs. This is simply due simply to the finite speed of light. You can never 'see' any event that is older than your observable universe and your observable universe is always identical in size to that of any remote observer's observable universe at the same time [age of the universe]. This also suggests any even that is causally disconnected from one observer must also be causally disconnected from all other potential observers.

 

[PeterDonis]

No observer that resides in a region of space visible to us can ever report 'seeing' anything that is not also visible to us.

To make this more precise: at the event on our worldline where we receive a report from some other observer about something they have seen, we must also be receiving (or have already received) light signals directly from the something they report having seen. In other words, any event that is in the past light cone of the observer that reported to us when they sent the report, is also in our own past light cone when we receive their report.

However, the following is also true: on any spacelike slice of constant time in the universe, the events on the worldlines of two spatially separated observers where they intersect that spacelike slice will have different past light cones, and therefore, at those events, the "observable universes" of the two will be different.

In other words, the precise referent of the term "observable universe" depends on which event on which observer's worldline you pick. You can make the "observable universes" of the same two observers the same or different by appropriate choices of events.

 

[kimbyd]

To clarify, the hypothetical connection I was proposing wasn't intended to suggest that you could use it to communicate beyond the observable universe. Rather, it's just a hypothetical construction to illustrate connectedness.