11

The tumultous period after the original announcement that the BICEP2 experiment had supposedly detected strong evidence of cosmological inflation in the form of B-mode polarization in the cosmic microwave background brought a lot of excitement and speculation. It seems that now, more than half a year later, the dust is finally starting to settle (har-har-har).

In all seriousness though: There have been a lot of rumor going around about the possibility that BICEP2's signal could be purely due to dust. At my university, one of the founding fathers of inflationary theory has been publicly proclaiming that the BICEP2 experimenters are 'crap', and that the claimed discovery is 'nonsense'. With the new data release by Planck, it may be time for a new, possibly decisive, analysis of the claimed discovery. Was it real? Was it dust? Is there still room for speculation?

I'm looking for an in-depth exposition of the credibility of the BICEP2 'discovery' of B-mode polarizations, in light of the latest data (if anyone feels other data than Planck latest dataset is relevant here, feel free to discuss it!). How much, if any, faith should we still put in the claimed discovery? Can (part of) the signal be dismissed as coming from galactic dust? Have we seen inflation?

Community
  • 1
Danu
  • 16,302
  • Related: http://physics.stackexchange.com/q/103951/ – Art Brown Dec 20 '14 at 17:35
  • In their publication already part of the signal had been subtracted to account for the dust. – anna v Dec 20 '14 at 17:37
  • 2
    @annav: that claim, given Planck data, is kinda bogus. They've underestimated the amount of dust in their tiny sector of the sky (see this article, for instance). – Kyle Kanos Dec 20 '14 at 18:00
  • They were told before the publication that it's going to be a face plant, weren't they? It wouldn't be the first time that this kind of thing happened because somebody had their eye on the Nobel rather than the data. FTL neutrinos, cold fusion... we could probably collect material for a ten volume work about scientists too eager to publish. What does this tell us, really? The obvious: even scientists are just people. – CuriousOne Dec 20 '14 at 18:38
  • @CuriousOne Were they? Where are you getting this story? – Danu Dec 20 '14 at 18:39
  • @Danu: Don't remember anymore. That's why I added the question mark. I think I heard it from someone, but then, from experience, every time that there is a really novel experimental result there will be people who will caution. Honestly, if they would have brought this to my desk and asked for a ten cent answer, I would have told them to check it another dozen times and to find additional evidence. It just didn't sound right and it looked too easy. – CuriousOne Dec 20 '14 at 18:43
  • @CuriousOne I read it up at the time. It is a different experimental method than the Planck. At the time Planck was giving limits with large errors. Their experiment did measure polarization with five sigma from zero, not limits. The lensing they avoided by choosing the a one degree window and dust was corrected by using Planck published data at the time. After the new results they were supposed to come out with a common paper with Planck by the end of November but no sign of it yet. Disagreement? – anna v Dec 20 '14 at 19:43
  • I check for news every now and then : http://bicepkeck.org/#papers – anna v Dec 20 '14 at 19:48
  • see also http://resonaances.blogspot.com/2014/09/bicep-what-was-wrong-and-what-was-right.html and links therein – Andre Holzner Dec 20 '14 at 20:03
  • You may enjoy the talk on BICEP2, inflation and string theory presented by Paul Steinhardt at the Strings 2014 conference available on YouTube. – JamalS Dec 20 '14 at 21:00
  • @annav: No disagreement from my side, I just don't believe in fishing for data in the noise of a secondary effect for the first time detection of a new signal. You want to show me primordial gravity waves? Build a gravitational wave experiment with the required sensitivity. In addition, the timing just smelled too much like an unhealthy competition with Planck. – CuriousOne Dec 21 '14 at 01:28
  • @CuriousOne they were sitting on the data for some time. If you look at the Planck analysis it is even worse, it fits polarization on large temperature maps whereas if you look at the experimental setup of BICEP2 it is very elegant measuring tiny temperature differences in small detectors. – anna v Dec 21 '14 at 05:18
  • @annav: I bet a lot of people got frustrated with the Planck analysis, but that doesn't really help with backgrounds, neither does the elegance of ones experiment. If the background is irreducible, sensitivity is just not the problem. – CuriousOne Dec 21 '14 at 07:32
  • @CuriousOne I agree to that last, though I do wonder how irreducible can a background drawn by rough strokes can be. – anna v Dec 21 '14 at 08:37
  • @annav: This may be just one of these cases where detecting the signal is the relatively easy part and characterizing the backgrounds may be the hard one. I will keep watching, but I would like to see an independent high resolution dust map. If they can show that their signal does not correlate with the dust, then we may have something. – CuriousOne Dec 21 '14 at 08:52

1 Answers1

7

OK, I found a recent link:

Planck versus BICEP2

Despite the new data, the collaboration did not give any insights into the recent controversy surrounding the possible detection of primordial "B-mode" polarization of the CMB by astronomers working on the BICEP2 telescope. If verified, the BICEP2 observation would be "smoking-gun" evidence for the rapid "inflation" of the early universe – the extremely rapid expansion that cosmologists believe the universe underwent a mere 10^–35 s after the Big Bang. A new analysis of polarized dust emission in our galaxy, carried out by Planck earlier in September, showed that the part of the sky observed by BICEP2 has much more dust than originally anticipated, and while this did not completely rule out BICEP2's original claim, it established that the dust emission is nearly as big as the entire BICEP2 signal. Both Planck and BICEP2 have since been working together on joint analysis of their data, but a result is still forthcoming.

This article was updated on 5 December 2014

Now as the BICEP2 paper had used published Planck data available at that time on the dust in their window of the sky, maybe this means that Planck had underestimated dust at that time? The forthcoming paper should surely clear this up.

Community
  • 1
anna v
  • 233,453
  • That's a great answer, Anna. I think that your "...means that Planck had underestimated dust at that time" almost certainly sums it up: these guys seemed to have used the best data at the time, so the quality of their research is undiminished. They are more likely simply unlucky rather than "crap" (I can't believe how talented scientists, who should be rejoicing in their own achievements, still seem to get off on putting others down) and moreover, the fact that they made the error and reworked the analysis will definitely pave the way for more decisive results in the future. – Selene Routley Dec 20 '14 at 23:12