- Hypernovae are the brightest flashes in the universe, evidence of stellar explosions.
- Blue Waters supercomputer simulation has revealed the cause of those bursts.
- Supercomputer simulations are a team sport.
Exploding stars look cool in the movies. Some of them provide a way to measure the size of the universe, and sure, some of these explosions create the heavy elements necessary for life as we know it. But for all these merits, you’ll not want to be anywhere in the vicinity when they explode.
Hypernovae, the largest of these starbursts, are extremely rare but are the brightest bangs you’ll see in the night sky. Until recently, the cause of these stellar explosions was unknown, but a simulation on the Blue Waters supercomputer at the US National Center for Supercomputing Applications (NCSA) has shone a light on the mechanism responsible.
For most of their lives, these stars are spinning, raging infernos of hydrogen. When a star dies, it runs out of fuel and the nuclear fusion inside its core shuts off. Without this constant combustion, the star loses the outward pressure supporting its gravity and it collapses.
As the star collapses, it falls until it reaches its compacted core, and then rebounds outward through the falling stellar debris. Astrophysicists used to think this bounce caused the explosions seen in hypernovaes, but computer simulations show that this shockwave does not have the energy required to launch the universe’s brightest explosions.
And since the rebound isn’t strong enough, scientists found themselves in a quandary. “We see supernova all the time, so there must be a way this happens in nature,” says Philipp Mösta, a postdoctoral scholar with a NASA Einstein fellowship at the University of California at Berkeley. “We need to understand the details of the mechanism that revives that shockwave and drives the explosion.”
As he worked with 3D simulations over the last few years, each bursting star model required an assumed magnetic field to kick-start the explosion. “And if you do that you can create these beautiful explosions that roughly match what one would see. That works out nicely, but the key question has always been how does the star create the strong magnetic field naturally.”
Until Mösta’s research, this answer remained hidden, awaiting a simulation strong enough to model the collapsing massive star yet with resolution fine enough to resolve the details needed to understand the amplification of the star’s magnetic field.
That’s when Blue Waters came into the picture. Using 130,000 cores in parallel, Mösta and his team ran simulations for 100 million core hours, generating over 500 TB of data in two weeks. Thanks to the team at NCSA, Mösta found the answer that had been eluding scientists for so long.
Beginning with a massive star in collapse, his simulation shows that, if followed at a very high resolution, the turbulence in the spinning, collapsing star is sufficient to create the strong large-scale magnetic field needed to explain the hypernova explosions.
“We have shown how this can happen — you only need the rapid rotation and you get the magnetic field for free.”
How rapid? Our own star rotates at its equator every 25 days or so. The massive stars responsible for the hypernova explosions, with a mass 25 times that of our sun, complete a rotation every few seconds. At that speed, the rebounding stellar material whips around at a blinding speed, transforming the star into a super-sized rotating electric generator. As the star collapses, its shrinking mass actually increases its rotational speed. Because of this velocity, within milliseconds of the rebound the massive star transforms the energy from its rotation into a magnetic field big enough to trigger the explosion known as a hypernova.
Mösta’s discovery is one of those breakthroughs only possible through supercomputing. But even with Blue Waters, “you really need a good team to pull this off,” he says. “Sure, you feel great when you manage to find something like this, but for me this is about much more than the results. It’s about making the simulations work as a team, because this would not have been possible without my collaborators. There’s a unique set of expertise that goes into creating and running these simulations.”