The University’s Center for Astrophysical Thermonuclear Flashes will use one of the world’s fastest supercomputers to model stellar explosions and collapses, natural phenomena that physicists claim hold the key to understanding dark energy.
The Center won 22 million computing hours from Argonne National Laboratory to use the Laboratory’s Blue Gene/P supercomputer. Blue Gene/P’s 160,000 cutting-edge processors can perform 445 trillion calculations per second, according to an Argonne press release. In three days, the supercomputer’s processors can perform a calculation that would take a normal desktop 1,000 years to complete.
Scientists will allocate the Center’s computing hours to simulate Type Ia supernovae, which are formed when white dwarf stars explode. The simulations will model small segments of the star as it begins to explode, building on research done at the Lawrence Berkeley National Lab, another supercomputing facility managed by the University of California.
Precise models of Type Ia supernovae are important because the explosions have a consistent luminosity. Scientists can measure the brightness of Type Ia supernovae to determine how far away they are, using Type Ia supernovae as distance markers for other celestial phenomena.
In 1998, researchers at the Berkeley Lab revealed that Type Ia supernovae were growing increasingly distant from each other, leading scientists to conclude that the universe was expanding at an accelerated rate. The researchers posited a repulsive force, dark energy, to explain this acceleration.
“Gravity always pulls things together,” said Robert Fisher, a research scientist at the Flash Center. “The only way to cause the universe to speed up is if there is some kind of repulsive gravity.”
Since there is no way to measure directly the luminosity that is crucial to determining distance, a supernova has to be understood from the inside out, starting from its creation.
“The question that arises when you understand the fundamental physics is how [stars that become supernovae] explode. Do they go off like combustion, which is called deflagration, or do they explode in a detonation?” Fisher said. “We want to go beyond the models to understand the physics of turbulent combustion.”