Computer Sim Shows Cosmic Evolution Over Billions and Billions of Years
The Illustris simulation models 41,416 galaxies out of the billions of galaxies in the universe. That's a small subset of the total, but the simulation closely matches the rate at which certain types of galaxies develop across the universe as a whole. "The work looks like a solid step forward in simulating the universe," said Arizona State University prof James Rhoads.
May 8, 2014 2:54 PM PT
Scientists have created a computer simulation that not only models the evolution of the universe down to the formation of individual galaxies, but also matches astronomical observations.
Unlike earlier efforts, the simulation depicts ordinary matter in the form of hydrogen gas from the very beginning.
The Illustris project models a cubic section of the universe measuring 106.5 megaparsecs on each side.
"We have improved upon previous work by both solving the equations of gas flows -- hydrodynamics -- more accurately, and including feedback processes from stars and black holes more comprehensively," said Shy Genel, one of the co-authors of the paper on the Illustris simulation published Wednesday in Nature.
More About the Simulation
The Illustris simulation's starting point is the universe at the age of 12 million years. It goes on to simulate the next 13.8 billion years of evolution, up to the present time.
Complex structures such as galaxies began forming when the universe was 12 million years old, and that's when computer models begin to be required, Genel told TechNewsWorld.
The simulation contains 12 billion visual resolution elements. That wealth of data lets researchers create visual pictures for comparison to images of the known universe, verifying its accuracy.
The simulation models 41,416 galaxies out of the billions of galaxies in the universe. That's a small subset of the total, but the simulation closely matches the rate at which certain types of galaxies develop across the universe as a whole.
Still, the size of the subset means "there is still a lot of room for improvement for future generations of simulations," Genel remarked.
Shedding Light on Illustris
Simulations of the combined evolution of dark matter and dark energy, which include only the force of gravity, have been run successfully for decades, but they cannot predict the distribution of galaxies made up of normal matter.
The Illustris project resolves this by directly accounting for the baryonic component, such as gas, stars and supermassive black holes, in addition to gravity. This offers, in principle, a self-consistent and fully predictive methodology.
The project models a comprehensive set of physical processes such as star-formation-driven galactic winds and black hole thermal energy injection throughout cosmic history.
Evaluating the Simulation
"The work looks like a solid step forward in simulating the universe," James Rhoads, a professor at Arizona State University's School of Earth and Space Exploration, told TechNewsWorld.
"The authors of this study have compared their galaxy shapes at the end of the model evolution to the shapes observed in the nearby universe, and say they are getting it about right," Rhoads continued. "They emphasize particularly that their ability to reproduce spiral galaxy shapes and sizes is a step forward."
However, "a careful reading of the 'looking ahead' section in the Nature paper suggests that their mix of galaxy types when the universe was half its current age might match the observations of that time less well," Rhoads suggested.
"We are following the growth rates of galaxies in our model, and we find that those are rather similar to those of observed galaxies but somewhat too fast," Genel said. "Developing improved models that will be able to get these growth rates in even better agreement with observations is one key aspect on which our team is working these days."
The simulation "is part of the process of understanding the universe," Rob Enderle, principal analyst at the Enderle Group, told TechNewsWorld. Its accuracy "is important for discovering dark matter and preparation for sending out manned space probes."