Astrophysicists will soon be able to zoom back hundreds of thousands of years through space and time to witness firsthand the processes that created stars in the Milky Way, thanks to an extraordinary simulation to be run on the Pleiades supercomputer at the NASA Advanced Supercomputing (NAS) facility at Ames Research Center.
The program developed by scientists at the University of California, Berkeley and Lawrence Livermore Laboratory combines real observations from the Hubble Space Telescope and other space-based observatories with a state-of-the art 3D code, ORION2.
The simulation takes into account gravity, swirling gases and magnetic waves along with intense radiation to piece together the origin of stars, stellar clusters, and the high-mass stars that form within them.
One of the key results suggests that some star clusters form like pearls in a chain along elongated, dense filaments inside molecular clouds—so-called “stellar nurseries”, says Richard Klein, of UC Berkeley.
The ORION2 radiation-magnetohydrodynamics code developed by Klein and his colleagues follows the formation and evolution of protostellar clusters across 700,000 years.
The science team tested each piece of physics in ORION against known data to demonstrate the code’s accuracy.
“The simulations show the entire evolution of these clusters—starting with a giant molecular cloud that collapses due to gravitational forces, to the formation of multiple turbulent clumps of interstellar gas inside the cloud, which in turn collapse into stellar clusters and cores that ultimately form individual stars,” according to Jill Dunbar of the NASA/Ames Research Center.
“Without NASA’s vast computational resources, it would not have been possible for us to produce these immense and complex simulations,” says Klein.
The ORION2 simulations incorporate a complex mix of gravity, supersonic turbulence, hydrodynamics (motion of molecular gas), radiation, magnetic fields, and highly energetic gas outflows.
In further development of the simulation, the team will incorporate more accurate data derived from physical processes, at larger spatial scales to enable higher resolution results with zoom-in capabilities.
“Higher resolution in the simulations will enable us to study the details of the formation of stellar discs formed around protostars,” says Klein.
“These discs allow mass to transfer onto the protostars as they evolve, and are thought to be the structures within which planets eventually form.”
Each simulation required 1,000 – 4,000 processors on Pleiades. In total, the simulations ran for more than six million processor-hours over several months, and produced about 100 terabytes of data.