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

I Hawking points discovered in CMB

  1. Aug 11, 2018 #1
    This paper seems to claim, that there is found powerful observational evidence for some anomalies in the CMB that seems to suggest a conformal cyclic cosmology and so called Hawking points.

    As I understand it, it also claims, that these points, were present in the very early universe and that they contained enormous amounts of energy. Does that mean that there is found evidence for big black holes in the very early universe? I would love if someone would explain in layman terms, what this is, and if the powerful evidence seems solid.

    If I have posted this in a wrong thread or forum, I am sorry. Did try to search for something about Hawking points. Have never heard of those before, and actually does not understand why it is not called black holes?
    Last edited by a moderator: Aug 11, 2018
  2. jcsd
  3. Aug 11, 2018 #2
    Sorry forgot the link... https://arxiv.org/pdf/1808.01740.pdf
  4. Aug 12, 2018 #3


    User Avatar
    Science Advisor
    Gold Member

    Prior to looking at the paper, let me just say that CCC has had a terrible history. The papers that Penrose and collaborators put out to support the idea some years ago were abominably bad...as in didn't bother to learn the basic statistical properties of the data set they were investigating bad.

    This is also in addition to the the fact that there are a lot of papers which claim statistical anomalies in the CMB that just don't have statistical support at all. That said, now to peruse the paper...
  5. Aug 12, 2018 #4


    User Avatar
    Science Advisor
    Gold Member

    Well, that was disappointing. They're basically doing the exact same analysis they did years ago, through with slightly different math. At least they might be getting the CMB simulations correct this time.

    My bet is if they're finding anything at all, they're just finding the residuals of bright point sources that weren't entirely removed from the CMB data (or, in some cases, the residuals of regions which were removed from the source data and filled in due to the existence of a bright point source there).
  6. Aug 13, 2018 #5
    Thank you for your answer.
    Not what I had hoped for obviously, but nice to be able to ask for some eduacated answers in here. :-)
  7. Aug 13, 2018 #6


    User Avatar
    Science Advisor
    Gold Member

    No worries.

    This kind of analysis is particularly challenging to do correctly because it diverges so dramatically from the way this data is usually analyzed. Because the ways in which the CMB data is processed are verified using very different analysis methods, it's very possible that their measurements will be sensitive to subtle systematic effects that don't have an impact on other methodologies.

    Note that it's also possible that dust in our galaxy is what they're seeing. This would explain the apparent coincidence they note regarding the BICEP 2 erroneous B-mode detection and the "Hawking point".

    The things I'd want to see before accepting that they're measuring something real are:
    1. A comparison that takes into account the noise properties of the CMB maps. Pixel-pixel correlations will be the most critical. Cross-checks using independent measurements would be sufficient to ensure noise isn't contributing.
    2. Examination of foreground models, including both far-away point sources and gas/dust within our galaxy to see if those might be the cause.
    3. A full Bayesian analysis which does a good job of estimating the the probability of seeing these kinds of things given a standard CMB model. The method they used isn't terrible, but is still too limited to get a solid handle on the relative probabilities (basically: the statistics for anomaly detection are very hard to get right).
    4. Independent cross-check by a team of people who are experienced in working with CMB data.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted