THEORY OF RECYCLED UNIVERSE CALLED INTO QUESTION
By Lisa Grossman
Wired News
May 13 2011
In November, cosmologists claimed to see echoes of violent collisions
that happened before the Big Bang in the form of circular patterns in
the early universe’s relic radiation. But two new analyses of the same
data, which are the first papers on the subject to be published in
peer-reviewed journals, assert that those circles are nothing special.
“We found there was nothing strange in the [cosmic microwave
background] data at all,” said astrophysicist Ingunn Wehus of the
University of Oslo, coauthor of a paper published online in the
Astrophysical Journal Letters May 9. The difference in their analyses,
she says, is “We do it correctly, and they do not.”
The original claim, made in research published on arXiv.org by
theoretical physicist Roger Penrose of the University of Oxford
in England and Vahe Gurzadyan of the Yerevan Physics Institute and
Yerevan State University in Armenia, made a small media splash (and
was one of Wired Science’s Top Scientific Breakthroughs of 2010).
Penrose had previously championed the idea that the universe got
its start well before the Big Bang, and has been cycling through an
endless series of bangs for eons. As evidence for this strange claim,
he and Gurzadyan pointed out funny concentric circles in the universe’s
baby photos, the cosmic microwave background. The CMB shows a universe
that looks more or less the same in every direction, with a nearly
uniform temperature of about 3 degrees Kelvin.
But some spots are hotter or colder than others. These fluctuations,
which ultimately led to the clumps of matter that make up galaxies
and other cosmic structures today, are not as random as they look,
Penrose and Gurzadyan claimed. Making a statistical search of the
CMB revealed concentric circles where the tiny temperature variations
between one spot and its neighbors are smaller than average.
Those circles are sure signs of pre-Big Bang activity, Penrose says.
He suggests they were generated by collisions between supermassive
black holes in an earlier eon, which gave off an intense burst
of energy. The burst would radiate outward in a uniform sphere of
gravitational waves, which would leave circles on the CMB when they
entered the epoch we live in.
“Because they claimed this, they got a lot of media attention.
Everybody was talking about this,” Wehus said. “It just seemed strange
that nobody else had noticed this before. It’s a very simple thing
to check. Since nobody else had checked it, we just decided to do it.”
Wehus and University of Oslo physicist Hans Eriksen redid Penrose’s
analysis of data from NASA’s Wilkinson Microwave Anisotropy Probe
(WMAP), which spent nine years mapping the glow of the first atoms
to release their radiation 380,000 years after the Big Bang. Another
independent group led by Adam Moss of the University of British
Columbia made a similar analysis, and published their results in the
Journal of Cosmology and Astroparticle Physics April 26.
To their surprise, both groups actually saw the same circles that
Penrose did. The circles are really there.
But then the skeptical researchers built thousands of random
simulations of the CMB, built up from the principles of the commonly
accepted standard model of cosmology. The circles showed up there too,
in the same numbers.
“In our case we found that the rings are in all the simulations,
so they’re just a feature of the standard model,” Wehus said. “It’s
not a signature of new physics.”
Moss and colleagues even found concentric equilateral triangles in the
CMB, a feature for which Penrose’s cyclic cosmology has no explanation.
“There is nothing special about the presence of low-variance circles
on the sky,” Moss concludes. “If there are signals of extraordinarily
early times buried in the CMB, they have not yet been found, and we
will have to keep looking.”
Penrose and Gurzadyan compared their results to simulations, too, but
Wehus and Moss claim they set their simulations to the wrong baseline.
Wehus and Moss assumed that the average variations in the CMB were set
by the laws of the standard model of cosmology; Penrose’s original
paper apparently used white noise. Even an updated version of the
paper, posted to arXiv on April 29, failed to hit the mark, Wehus says.
“Some way or another they screwed up their simulations,” Wehus said.
“They used wrong simulations.”
This doesn’t necessarily mean the cyclic universe theory is wrong,
she adds.
“We are not knocking down the idea of Penrose, of there being a cyclic
universe,” she said. “We’re just saying there’s no evidence for it.”
Penrose is sticking to his story. In the most recent paper, he looks
for concentric sets of three or more circles in both the WMAP data
and a simulated sky. The patterns and colors for the simulated sky
look random, he says, but the patterns on the actual sky do not.
“Such a pattern is consistent with [a cyclic cosmology], but hard
to square with the standard inflationary view of the origin of the
temperature variations,” Penrose wrote to Wired.com in an e-mail. “I
think that Eriksen and Werhus may have read that part of our paper
rather hastily … evidently not having understood what we were doing.”
“I suppose there may well be further argument about all this – which
is to be expected, of course – and maybe we have missed something
important,” he added. “But it seems to me that here is something to
be taken very seriously.”