Clocking the Early Universe
Calculations Performed at NERSC Help Scientists Close in on the Nature of Dark Energy
April 17, 2014
Margie Wylie, firstname.lastname@example.org, +1 510 486 7421
Astronomers have made the most accurate calculation yet of the expansion rate of the young Universe with help from supercomputers at the U.S. Department of Energy’s National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory (Berkeley Lab). Their findings could help scientists discover the nature of dark energy, the mysterious, repulsive force that pervades our universe causing it to expand at an accelerating rate.
By analyzing the light of distant quasars gathered by the Baryon Oscillation Spectroscopic Survey (BOSS), two teams of scientists found the 4 billion year-old Universe was expanding by one percent every 44 million years.
“This means if we look back to the universe when it was less than a quarter of its present age, we’d see that a pair of galaxies separated by a million light years would be drifting apart at a velocity of 68 kilometers a second as the universe expands,” said Andreu Font-Ribera, a postdoctoral fellow in Berkeley Lab’s Physics Division who lead one team and was a member of the other. “The uncertainty is plus or minus only a kilometer-and-a-half per second,” continued Font-Ribera, who presented the findings at the April 2014 meeting of the American Physical Society.
The science teams used two different analytical techniques and the resources of NERSC to come to their findings. One recent analysis used a tested approach with far more data than before. Font-Ribera’s team reported using a new kind of analysis late last year.
BOSS uses the 2.5-meter Sloan Foundation Telescope at the Apache Point Observatory in New Mexico. Part of the Third Sloan Digital Sky Survey, BOSS gathers the faint light of distant, bright objects, like galaxies, using “plug plates” at the telescope’s focal plane. Plug plates are aluminum disks with holes drilled to match the precise position of preselected objects.Optical fibers are plugged into these holes by hand every day to guide the light from each target to the BOSS instrument.
Using the light of over a million bright galaxies, BOSS has already given scientists a pretty good handle on some crucial parameters of the Universe up to about 6 billion years ago. But to go further back in time, astronomers needed even brighter objects.BOSS pioneered a method that relies on quasars, young galaxies powered by massive black holes. As the light from a distant quasar passes through intervening clouds of intergalactic hydrogen gas, patches of greater density absorb more light, leaving a distinctive signature in the light’s spectrum. With the light signatures from enough quasars, close enough together, the position of the gas clouds can be mapped in three dimensions – both along the line of sight for each quasar and then among the dense patches revealed by other quasars.
The more recent of the two analyses used this method, called autocorrelation, on 140,000 BOSS quasars. To eliminate noise and biases in the observational data, team-member Font-Ribera generated hundreds of synthetic data sets using NERSC's Cray XE6 supercomputer nicknamed Hopper. Simulating a BOSS dataset is no small feat: “BOSS has observed a large fraction of the observable Universe, and we needed to simulate several realizations of this huge volume with a very good accuracy.” said Font-Ribera.
The earlier study—lead by Font-Ribera—used a new method. Instead of comparing quasars’ light signatures to each other, the team correlated the light signature of quasars with the density of quasars in a given patch of the sky, a method called cross-correlation.
“Quasars are massive galaxies, and we expect them to be in the denser parts of the universe, where the density of the intergalactic gas should also be higher,” says Font-Ribera. “Therefore we expect to find more of the absorbing gas than average when we look near quasars.”
To make the correlation, Font-Ribera had to run some pretty hefty calculations at NERSC: “One of the key elements in this type of analysis is computing the uncertainties in your results. We do that by creating a covariance matrix of measurements,” he explained. This matrix compared 140,000 quasar light signatures gathered by BOSS with the density of quasars found in each of 66 different patches of sky.
Combining the two analyses, autocorrelation and cross correlation, yielded the most accurate expansion rate measurement ever taken of any epoch of the Universe. Measuring this rate over the Universe’s entire history is key to determining the nature of the dark energy. Understanding dark energy, in turn, informs our understanding of the Universe’s fundamental nature.
“These results allow us to study the geometry of the Universe when it was only a fourth its current age,” said Font-Ribera. “Combined with other cosmological experiments, we can learn about dark energy and put tight constraints on the curvature of the Universe: It’s very flat!” he concluded.
 "Baryon Acoustic Oscillations in the Ly-alpha forest of BOSS DR11 quasars," by Timothée Delubac, Julian E. Bautista, Nicolas G. Busca, James Rich, David Kirkby, Stephen Bailey, Andreu Font-Ribera, Anže Slosar, Khee-Gan Lee, Matthew M. Pieri, Jean-Christophe Hamilton, Michael Blomqvist, Jo Bovy, William Carithers, Kyle S. Dawson, Daniel J. Eisenstein, J.-M. Le Goff, Daniel Margala, Jordi Miralda-Escudé, Adam Myers, Robert C. Nichol, Pasquier Noterdaeme, Ross O'Connell, Nathalie Palanque-Delabrouille, Isabelle Pâris, Patrick Petitjean, Nicholas P. Ross, Graziano Rossi, David J. Schlegel, Donald P. Schneider, David H. Weinberg, and Christophe Yeche, has been submitted to Astronomy & Astrophysics and is available online at http://sdss3.org/science/lyaauto.pdf.
 "Quasar-Lyman-alpha Forest Cross-Correlation from BOSS DR11: Baryon Acoustic Oscillations," by Andreu Font-Ribera, David Kirkby, Nicolas Busca, Jordi Miralda-Escudé, Nicholas P. Ross, Anže Slosar, Éric Aubourg, Stephen Bailey, Vaishali Bhardwaj, Julian Bautista, Florian Beutler, Dmitry Bizyaev, Michael Blomqvist, Howard Brewington, Jon Brinkmann, Joel R. Brownstein, Bill Carithers, Kyle S. Dawson, Timothée Delubac, Garrett Ebelke, Daniel J. Eisenstein, Jian Ge, Karen Kinemuchi, Khee-Gan Lee, Viktor Malanushenko, Elena Malanushenko, Moses Marchante, Daniel Margala, Demitri Muna, Adam D. Myers, Pasquier Noterdaeme, Daniel Oravetz, Nathalie Palanque-Delabrouille, Isabelle Pâris, Patrick Petitjean, Matthew M. Pieri, Graziano Rossi, Donald P. Schneider, Audrey Simmons, Matteo Viel, Christophe Yeche, and Donald G. York, has been submitted to Journal of Cosmology and Astroparticle Physics and is available online at http://arxiv.org/abs/1311.1767M.
About Computing Sciences at Berkeley Lab
High performance computing plays a critical role in scientific discovery. Researchers increasingly rely on advances in computer science, mathematics, computational science, data science, and large-scale computing and networking to increase our understanding of ourselves, our planet, and our universe. Berkeley Lab’s Computing Sciences Area researches, develops, and deploys new foundations, tools, and technologies to meet these needs and to advance research across a broad range of scientific disciplines.