Explaining the Horizon Problem in cosmology.
July 4, 2010 7:14 PM Subscribe
Cosmology / Physics question: Explain the "Horizon Problem" to me, in regards to measurements of Cosmic Mircowave Background Radiation - why would physicists regard different temperatures in differnet directions as their null hypothesis?
I've been reading ("String Theory for Dummies" - pardon me, I'm a biologist), and the question of the Horizon Problem came up. If we measure the CMBR in one direction (14 billion years old) and the CMBR in another direction (14 billion years old), they come up to the same temperature. But since the points we are measuring are 28 billion lightyears apart, there is no way they could have "communicated" their temperature, so physicists have had to come up with more complex models to explain the evolution of the universe.
However - why would we expect them to have different temperatures? The radiation in both directions was produced by the same process.
If I've got a big gas burner (the big bang), and I put two saucepans with water on it, I would expect the water in both pans to heat up and boil at the same time, because the same energy source is heating them. There doesn't need to be "communication" or heat exchange between the two saucepans for them to heat up the same way.
What am I missing? Please try to avoid calculus in your answer TIA.
I've been reading ("String Theory for Dummies" - pardon me, I'm a biologist), and the question of the Horizon Problem came up. If we measure the CMBR in one direction (14 billion years old) and the CMBR in another direction (14 billion years old), they come up to the same temperature. But since the points we are measuring are 28 billion lightyears apart, there is no way they could have "communicated" their temperature, so physicists have had to come up with more complex models to explain the evolution of the universe.
However - why would we expect them to have different temperatures? The radiation in both directions was produced by the same process.
If I've got a big gas burner (the big bang), and I put two saucepans with water on it, I would expect the water in both pans to heat up and boil at the same time, because the same energy source is heating them. There doesn't need to be "communication" or heat exchange between the two saucepans for them to heat up the same way.
What am I missing? Please try to avoid calculus in your answer TIA.
Response by poster: From that page: "Isotropy was simply an initial condition specified by the Standard Big Bang model. But such ad hoc assumptions don't make for a very satisfying theory."
Why is isotropy not a very satisfying theory? In the terms I'm used to, it sounds like a decent null hypothesis. "The event that occured resulted in equal distribution of energy in all directions. If we detect different energy in different directions, then something interesting is happening," Instead, physicists seem to be going for the opposite - "We detect the same energy in different directions, therefore something interesting is happening." Why?
posted by Jimbob at 7:58 PM on July 4, 2010
Why is isotropy not a very satisfying theory? In the terms I'm used to, it sounds like a decent null hypothesis. "The event that occured resulted in equal distribution of energy in all directions. If we detect different energy in different directions, then something interesting is happening," Instead, physicists seem to be going for the opposite - "We detect the same energy in different directions, therefore something interesting is happening." Why?
posted by Jimbob at 7:58 PM on July 4, 2010
Best answer: Here's a thread at the site Physics Forums on the same topic:
Why do we need inflation to explain the homogeneity of CMB?
Note that it was started only forty or fifty days ago, so you could probably join the site and ask questions right in that thread if you wanted.
What's being said there is that assuming homogeneous initial conditions would be excessively strident:
posted by XMLicious at 8:16 PM on July 4, 2010
Why do we need inflation to explain the homogeneity of CMB?
Note that it was started only forty or fifty days ago, so you could probably join the site and ask questions right in that thread if you wanted.
What's being said there is that assuming homogeneous initial conditions would be excessively strident:
But I'm equally perplexed as to why that seems so obvious to so many people. I mean, we're at least assuming that things like the speed of light are the same in all directions, right? So why is one more form of isotropy a no-no?Sure, the initial conditions could have been perfectly homogeneous. There is nothing that forbids this in principle. But, as you say, the IC's could have been inhomogeneous as well. In fact, there are lots more ways of not being homogeneous than there are of being homogeneous. So, given a generic initial state, one would not expect it to be perfectly homogeneous. By just putting such a condition in 'by hand' is considered by many to be a serious fine tuning of initial conditions.— user bapowell
posted by XMLicious at 8:16 PM on July 4, 2010
Best answer: The way I understand it, nothing can be perfectly smooth or uniform due to fluctuations at quantum scales. If you take that initial tiny little ball of plasma and expand it to the current size of the universe without inflation, then those quantum fluctuations would account for more variation in the CMBR than we observe. In fact, the small variances that we do observe seem to fit closely with what you would expect from the quantum variations in that early mass of plasma coupled with inflation. In other words, it wouldn't have been possible to have an early universe regular enough to make what we see now without inflation helping out. In your saucepan example you aren't magnifying the differences by a factor of 1078: if you did then you would no longer conclude that they are the same to 10 ppm (or whatever the CMBR is.)
posted by Rhomboid at 9:37 PM on July 4, 2010
posted by Rhomboid at 9:37 PM on July 4, 2010
...that initial tiny little ball of plasma...
Since we're talking about cosmology I figured I'd mention that this is one of the things that can cause misunderstandings: the tiny ball of plasma at the point of the big bang corresponds to our current-day observable universe, the stuff we can actually see. My understanding is that the universe itself in entirety may have extended infinitely in every direction even at the time of the big bang and may do so now - that cosmology today doesn't make any definite statements about the universe having 3-dimensional boundaries or edges.
There's a great Scientific American article that gets passed around, "Misconceptions about the Big Bang". (Unfortunately all my links to non-paywalled copies of it seem to be dead.)
posted by XMLicious at 10:18 PM on July 4, 2010 [1 favorite]
"But I'm equally perplexed as to why that seems so obvious to so many people. I mean, we're at least assuming that things like the speed of light are the same in all directions, right? So why is one more form of isotropy a no-no?"
Well, we have pretty tight constraints on the variability of the speed of light over a wide range of times and locations (albeit not as far back as the very early universe), and it's a fundamental component of the laws of physics as we know them that it is invariant like that. On the other hand, we can clearly see right now that temperature and density and so on vary wildly from place to place, so there's no mechanism in play right now to make that the case.
So to have it isotropic initially requires not just that mechanism to be in place, but for it to be in place at only that initial moment in time, and you then have to explain how that mechanism works.
Inflation does postulate a somewhat weird mechanism that's turned on initially and then switches off, but it has the advantage of allowing standard processes to take place to get things into thermal equilibrium and also solves the flatness problem, solves the monopole problem, and suggests other observational signatures that might back it up. The mechanism by which it might have occurred is also fairly straightforward to add in to bog-standard theories.
I think it's a somewhat more powerful mechanism in terms of what problems it can solve and is fairly straightforward to write down the mathematics for compared to the fine-tuning of the initial conditions like that needed to have the same temperature everywhere very early on.
posted by edd at 7:28 AM on July 5, 2010 [1 favorite]
Well, we have pretty tight constraints on the variability of the speed of light over a wide range of times and locations (albeit not as far back as the very early universe), and it's a fundamental component of the laws of physics as we know them that it is invariant like that. On the other hand, we can clearly see right now that temperature and density and so on vary wildly from place to place, so there's no mechanism in play right now to make that the case.
So to have it isotropic initially requires not just that mechanism to be in place, but for it to be in place at only that initial moment in time, and you then have to explain how that mechanism works.
Inflation does postulate a somewhat weird mechanism that's turned on initially and then switches off, but it has the advantage of allowing standard processes to take place to get things into thermal equilibrium and also solves the flatness problem, solves the monopole problem, and suggests other observational signatures that might back it up. The mechanism by which it might have occurred is also fairly straightforward to add in to bog-standard theories.
I think it's a somewhat more powerful mechanism in terms of what problems it can solve and is fairly straightforward to write down the mathematics for compared to the fine-tuning of the initial conditions like that needed to have the same temperature everywhere very early on.
posted by edd at 7:28 AM on July 5, 2010 [1 favorite]
This thread is closed to new comments.
posted by zengargoyle at 7:49 PM on July 4, 2010