# All Is Dust

Joke stolen from some fellow PI postdocs.

The BICEP2 and Planck experiment teams have released a joint analysis of their data, discovering what many had already suspected: that the evidence for primordial gravitational waves found by BICEP2 can be fully explained by interstellar dust.

For those who haven’t been following the story, BICEP2 is a telescope in Antarctica. Last March, they told the press they had found evidence of primordial gravitational waves, ripples in space-time caused by the exponential expansion of the universe shortly after the Big Bang. Soon after, though, doubts were raised. It appeared that the BICEP2 team hadn’t taken proper account of interstellar dust, and in particular had mis-used some data they scraped from a presentation by larger experiment Planck. After Planck released the correct version of their dust data, BICEP2’s predictions were even more evidently premature.

Now, the Planck team has exhaustively gone over their data and BICEP2’s, and done a full analysis. The result is a pretty thorough statement: everything BICEP2 observed can be explained by interstellar dust.

A few news outlets have been describing this as “ruling out inflation” or “ruling out gravitational waves”, both of which are misunderstandings. What Planck has ruled out are inflation (and gravitational waves caused by inflation) powerful enough to have been observed by BICEP2.

To an extent, this was something Planck had already predicted before BICEP2 made their announcement. BICEP2 announced a value for a parameter r, called the tensor-scalar ratio, of 0.2. This parameter r is a way to measure the strength of the gravitational waves (if you want to know what gravitational waves have to do with tensors, this post might help), and thus indirectly the strength of inflation in the early universe.

Trouble is, Planck had already released results arguing that r had to be below 0.11! So a lot of people were already rather skeptical.

With the new evidence, Planck’s bound is relaxed slightly. They now argue that r should be below 0.13, so BICEP2’s evidence was enough to introduce some fuzziness into their measurements when everything was analyzed together.

I’ve complained before about the bad aspects of BICEP2’s announcement, how releasing their data prematurely hurt the public’s trust in science and revealed the nasty side of competition for funding on massive projects. In this post, I’d like to talk a little about the positive side of the publicity around BICEP2.

Lots of theorists care about physics at very very high energies. The scale of string theory, or the Planck mass (no direct connection to the experiment, just the energy where one expects quantum gravity to be relevant), or the energy at which the fundamental forces might unify, are all much higher than any energy we can explore with a particle collider like the LHC. If you had gone out before BICEP2’s announcement and asked physicists whether we would ever see direct evidence for physics at these kinds of scales, they would have given you a resounding no. Maybe we could see indirect evidence, but any direct consequences would be essentially invisible.

All that changed with BICEP2. Their announcement of an r of 0.2 corresponds to very strong inflation, inflation of higher energy than the Planck mass!

Suddenly, there was hope that, even if we could never see such high-energy physics in a collider, we could see it out in the cosmos. This falls into a wider trend. Physicists have increasingly begun to look to the stars as the LHC continues to show nothing new. But the possibility that the cosmos could give us data that not only meets LHC energies, but surpasses them so dramatically, is something that very few people had realized.

The thing is, that hope is still alive and kicking. The new bound, restricting r to less than 0.13, still allows enormously powerful inflation. (If you’d like to work out the math yourself, equation (14) here relates the scale of inflation $\Delta \phi$ to the Planck mass $M_{\textrm{Pl}}$ and the parameter r.)

This isn’t just a “it hasn’t been ruled out yet” claim either. Cosmologists tell me that new experiments coming online in the next decade will have much more precision, and much better ability to take account of dust. These experiments should be sensitive to an r as low as 0.001!

With that kind of sensitivity, and the new mindset that BICEP2 introduced, we have a real chance of seeing evidence of Planck-scale physics within the next ten or twenty years. We just have to wait and see if the stars are right…

## 8 thoughts on “All Is Dust”

1. JollyJoker

A comment by Jester at http://resonaances.blogspot.com
” The tensor to scalar ratio is given by r = (E/3.310^16 GeV)^4 , where E is the energy scale of inflation (more precisely, E = V^1/4 where V is the value of the inflaton potential during inflation). So, for E = 210^16 GeV you get r~0.1, for E = 10^16 GeV you get r~0.01, and for E = 0.6*10^16 GeV you get r~0.001. If E is lower than that, we will not see gravitational waves anytime soon…. ”

This seems not to match equation 14 in the pdf you linked to?

Getting down to r=0.001 seems pretty nice, but “in the next decade” 😦

Like

1. JollyJoker

Oh damn, multiplication signs are used for italics here. Well, I guess it’s obvious where they should be.

Like

Hmm, that’s odd. I got the pdf link from a cosmology friend, I’ll have to ask him what’s up.

“In the next decade” doesn’t worry me too much…it’s definitely a lot better than the timeline for getting results out of a new Chinese collider, anyway!

Like

My cosmology friend confirms that the formulas deal with slightly different things. Equation 14 on that pdf deals with the amount of change in the inflaton during inflation, while the equation in that post involves the overall energy scale.

Like

1. JollyJoker

Inserting the r=0.01 in both would then give you an energy scale of 10^16 GeV and planckian amount of change? That’s pretty much what the first page of the pdf says, but what is that “change” really? Does the energy go from Planck to zero with an average of 10^16 GeV?

Like