Research @ KICP
September 23, 2008
by Jeremy Tinker
Over the past twenty years, astronomers have been exploring our local universe by mapping out the positions of all nearby galaxies. As this map has expanded from a few hundred galaxies to nearly a hundred thousand galaxies, located out to distances billions of light years from our own, researchers have found expansive ''holes'' in the galaxy distribution where there appears to be nothing at all. These voids have excited and perplexed astronomers -- why are they so large and empty? Are they a challenge to our picture of structure formation in the Universe? Recent work led by University of Chicago and KICP Associate Fellow Jeremy Tinker and former Chicago undergraduate researcher Charlie Conroy has addressed this problem head on. Using data from the Sloan Digital Sky Surkey (SDSS), the research team measured the sizes and frequency of these voids and compared them to predictions created from supercomputer simulations of cosmic structure -- and found that the cold dark matter simulations and the data were in perfect agreement.
Figure: <b>A supercomputer simulation of the distribution of galaxies in the Universe.</b> <br />The simulation starts with dark matter only, initialized such that the dark matter distribution matches that seen in the cosmic microwave background radiation in the early universe. It is evolved under the force of gravity, creating voids and clusters. Galaxies are then put into the dark matter halos such that the number of bright and faint galaxies matches that observed in the SDSS. The big, red dots correspond to the brightest galaxies found in the SDSS; the smaller dots correspond to fainter galaxies; and the light blue dots are galaxies fainter than our own. Dense knots of galaxies -- clusters -- can readily be seen in the simulation. Voids are regions with few if any galaxies. The sizes of the voids in this simulation match those measured in the SDSS.
A central element of the standard cosmological theory is cold dark matter, which exerts gravity but does not emit light. Dark matter is smoothly distributed in the early universe, but over time gravity pulls it into filaments and clumps and empties out the spaces between them. Galaxies form when hydrogen and helium gas falls into collapsed dark matter clumps, referred to as ''halos,'' where the gas can collapse further to form luminous stars.
But astronomers were not sure if the areas that are devoid of galaxies are also devoid of dark matter, or if the dark matter is there, but for some reason stars just didn't form in these voids. The team used bright galaxies to trace the structure of dark matter and compared it with computer simulations to predict the number and sizes of voids. Conroy, now a graduate student at Princeton University, measured the voids in the SDSS maps. When using galaxies brighter than the Milky Way to trace structure, the biggest empty voids they found were about 75 million light years across, and the predictions from the simulations were bang-on.
The sizes of voids are ultimately set by the small variations in the primordial distribution of dark matter, and by the amount of time that gravity has had to grow these small variations into large structures. The agreement between the simulations and the measurements holds for both red (old) and blue (new) galaxies. Halos of a given mass seem to form similar galaxies, both in numbers of stars and in the ages of those stars, regardless of where the halos live.
The results were presented at the recent international symposium in Chicago, titled <a href='http://sdss2008.uchicago.edu/'>''The Sloan Digital Sky Survey: Asteroids to Cosmology.''</a> A paper detailing the analysis will appear in the The Astrophysical Journal, with the title ''Void Statistics in Large Galaxy Redshift Surveys: Does Halo Occupation of Field Galaxies Depend on Environment?''
<b>Additional References:</b><br /> 1). Jeremy L. Tinker and Charlie Conroy, <a target='_blank' href='http://arxiv.org/abs/0804.2475'>The Void Phenomenon Explained</a>.<br /> 2). Jeremy L. Tinker, Charlie Conroy, Peder Norberg, Santiago G. Patiri, David H. Weinberg, Michael S. Warren, <a target='_blank' href='http://arxiv.org/abs/0707.3445'>Void Statistics in Large Galaxy Redshift Surveys: Does Halo Occupation of Field Galaxies Depend on Environment?</a>
KICP Members: Jeremy Tinker