Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Surface of Last Scattering

  1. Jul 8, 2010 #1
    This came up when I was trying to explain how we receive light from decoupling era: I know that the CMB we are getting is from some Last Scattering Surface (LSS) and this surface is moving back in time in some sense. But the question is: will there come a time that we are past the decouping era (back in time) and there would be no further ambient CMB in the sky? In other words, will the LSS vanish after a certain point in our past light cone?
    Last edited: Jul 8, 2010
  2. jcsd
  3. Jul 8, 2010 #2
    As time passes CMB is redshifted towards infinite wavelength and therefore someday it will be virtually undetectable, but it will never completely vanish, unless you are prepared to wait infinite time.

    The only way that CMB abruptly stops is if cosmological principle is not true (i.e. there is edge of universe), which would be highly surprising.
  4. Jul 8, 2010 #3
    Even if there is no spatial edge there is still a temporal one, that is, the time before decoupling.
  5. Jul 8, 2010 #4
    I agree, but what it has to do with our perception of CMB?
  6. Jul 8, 2010 #5
    I don't understand what you exactly mean by "our perception of CMB" but my question is essentially that if there is a temporal edge, then is it true that past a certain time we would see NO CMB in the sky?
  7. Jul 8, 2010 #6
    Well Cmb is called "oldest light in the sky", and the surface of last scattering is the farthest object we can see (some 45 Billion light years away). Before that universe was opaque for photons. We could theoretically see beyond that with some kind of neutrino telescope, but that is out of realm of possible for now.
  8. Jul 8, 2010 #7
    So as long as we are dealing with photons, there will come a time that no CMB (in photonic sense of the term) would exist in the sky, is that correct?
  9. Jul 8, 2010 #8
    CMB is all about photons. I mentioned neutrinos as possible way to peer in the times much closer to big bang.

    I really can't tell you for sure when CMB will be undetectable even in principle. I guess that it certainly can't be detected when its wavelength becomes equal or larger then cosmological horizon.
  10. Jul 8, 2010 #9
    The issue comes down the size of the universe.

    If the universe is infinite, it is not about time. It doesn't matter how long ago the universe was created as there will always be some point a bit further away than what we currently detect that will provide the new photons for the CMB of tomorrow. If today's photons come from a sphere that is of X radius from our vantage point as the center, then tomorrow's will come from a sphere with X + 1 light day radius.

    If the universe if finite then the CMB might start fading out as the light from the farthest reaches of the universe hits us. The light from those regions where the edge of the universe is closest would stop first, in this event.

    So far, there's no end of the CMB in sight so there's no reason be believe that the universe is finite.
    Last edited: Jul 8, 2010
  11. Jul 8, 2010 #10
    Nop inflector, that is not correct. Finite universe has no edges.
  12. Jul 8, 2010 #11


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    This is your wrong assumption. What you said in the first post. This is the source of the confusion.
    The LSS is not "moving back in time"

    The light always originated when the U expansion was about 380 thousand years old. What Cmb we received in past did originate then, what we now receive originated then, and in future what we will receive will also be light that originated at that same moment. It was the moment the fog cleared.

    The LSS is simply where the matter is now located that emitted the Cmb light that we now receive. Naturally as time goes on, light has had more time to travel, so the emitting matter must be more and more distant. Also the distance to the LSS is subject to the general expansion of distance.
    So the distance to LSS does gradually increase. But the time of emission is always the same...year 380 thousand, when the medium cooled enough to become effectively transparent.

    Get rid of the idea that the LSS is "moving back in time" (ie somehow getting "earlier") and I think you will be all right.
    Last edited: Jul 8, 2010
  13. Jul 8, 2010 #12
    It may or it may not. If the universe if curved you are correct but we don't know that.
  14. Jul 9, 2010 #13
    I beg to differ. The light cone is not attached to the event (emission from LSS), its origin coincides with the observer at all times. In this sense, as time goes by, whatever the event is, it will move down the past light cone (which is synonymous to "moving back in time"). I know that little, what I don't know is that assuming the universe is compact without boundary, what will happen to CMB after sufficiently long (yet finite) time.
  15. Jul 10, 2010 #14
    I see your point and apologize for misjudgment. But the question is still open to me: assuming a compact universe with no boundary, what will happen to the observed CMB in later times?
  16. Jul 10, 2010 #15


    User Avatar

    Staff: Mentor

    No. Time and distance are coupled. We see objects that are near and new and we see objects that are old and far. The fact that time is progressing does not alter which objects we see and which we don't: we'll always be able to see the oldest "object" in the universe.
    The SLC will just continue to move further and further away and red shift more and more.
  17. Jul 10, 2010 #16
    I agree, but the "always" condition requires certain class of geometries, I think. For example assume a flat and finite universe with boundary, then you won't receive such signals (from a certain moment in the past) forever.
  18. Jul 10, 2010 #17


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Vizart, are you OK with the idea of a compact U with no boundary which expands forever due to positive cosmological constant Lambda?

    That is one version of the widely used "standard" or "consensus" universe model, the LambdaCDM. The other version of the standard cosmo model is spatially infinite.

    No wise person would claim to know the right model, but one has to have some model to fit the data to and the LCDM gives a remarkably good fit (either spatial finite or spatial infinite) and has become the preferred one to use.

    So if you are talking about LCDM, then it is easy to answer your question. The observed CMB just slowly fades. Wavelengths lengthen and the temperature declines forever---as the U expands forever.

    A prominent cosmologist, Larry Krauss, wrote a paper describing this and other aspects of the indefinite future (according to LCDM). He pointed out that in the far future the cosmologists if there are any would not be able to detect the Background because it will be too redshifted. He pointed out that although the U keeps expanding the imagined future scientists will not have all the clues that we have, and may not realize it is expanding. If you are curious in such far-distant matters, the Krauss paper is called "return of a static universe and the end of cosmology"

    If you google "Krauss static" you will get the free online preprint: http://arxiv.org/abs/0704.0221
    Or google "Krauss end cosmology".

    Working cosmologists normally assume no boundary. I wouldn't speculate about the case where there was a boundary. What would it be like?
    Keeping within the no-boundary assumption, there is the spatial compact and spatial infinite distinction. It doesn't seem to make any difference---the CMB just keeps on redshifting as the U expands. I'm probably repeating what others have already replied in this thread.
    Last edited: Jul 10, 2010
  19. Jul 11, 2010 #18


    User Avatar
    Science Advisor
    Gold Member

    What do you mean by 'compact'? There is no compelling evidence to date suggesting the universe is finite or infinite. All we know is it is temporally finite.That is unsurprising.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook