Artist's impression of the early universe
Credit: NASA/Adolf Schaller
SYDNEY, 24 July 2006 - If current computer simulations are anything to go by, the Big Bang is wrong, according to cosmologists. But the experts aren't throwing away the theory just yet, they're blaming the computers.
According to a report published in U.S. journal Science on Friday, current simulations, which theoretically map the formation of the universe from when the Big Bang 'unfolded' over 13 billion years ago, are inconsistent with actual observations of nearby galaxies.
Cosmologists might have to wait decades before computer simulations can substantiate their theories, as current computers are simply not powerful enough to yield useful data.
For example, the authors, Joss Bland-Hawthorn, of the Anglo-Australian Observatory and Big Bang theory pioneer, P. J. E. Peebles of Princeton University, USA said in their report simulations suggest "our Local Group is expected to have a thousand small mass concentrations [of stars] but we infer the presence of fewer than 50 from the number of visible galaxies".
Such discrepancies are likely due to the limits of modern computers, rather than any particular flaw in the predictions of Big Bang inflationary cosmology.
Gas dynamics that determine how matter settles into galaxies and collapses from there to form much denser stars; and how stellar winds and explosions stir up the remaining gas and control the rate at which new stars form are concepts that are far too complex for current programs, the authors said.
"The simulations invoke many parameters to describe the four per cent of the universe that is made up of baryonic matter [the particles that make up stars, people and so on]," But they said the simulations were not nearly as comprehensive for dark matter and dark energy, which make up the remaining 96 per cent of the universe.
So computer simulations are not about to revolutionise the way we think about the formation of galaxies. "I think we are a long way from achieving this," said Bland-Hawthorn. "There is so much more physics that needs to be included. It will be 10 to 20 years before we get this right."
Regardless, computer simulation is still an invaluable tool. Working with up to 10 billion 'particles', the "computations yield structures that look a good deal like real galaxies and clusters of galaxies," the authors said.
These include spiral galaxies which resemble neighbouring Andromeda and our own Milky Way, "adding to the evidence that our picture for the evolution of the universe is on the right track."
In turn, the reliability of computer simulations will improve as actual observations, such as those of forming stars, are applied to simulation programs.
