The National Institute for Computational Sciences

When Stars Go Boom

Kraken Supercomputer Sheds New Light on Life-giving Stars

by Scott Jones

 


Despite their similarity from our perspective light years away, those little blinking dots in the sky we call stars are anything but alike.

Like people, they are all individuals, each with their own quirky characteristics. Some are big, some small, some bright, some dim. And just like us they will all one day die.

When it comes to core-collapse supernovas (CCSNs), their death is pivotal to our existence. Their demise occurs when they violently implode on themselves, littering the universe with the elements necessary for life to occur. CCSNs provide the building blocks for all of the elements up to iron. Their deaths are, literally, responsible for life as we know it.

The mechanism behind these explosions, however, is still somewhat of a mystery. In fact, it is often cited as one of the greatest unsolved mysteries in all of astrophysics. But perhaps not for long: A team of researchers from Florida Atlantic University, Oak Ridge National Laboratory, the University of Tennessee, and North Carolina State University is using the Kraken supercomputer in an attempt to solve the questions behind these magnificent events once and for all.

Kraken, a Cray XT5 managed by the University of Tennessee and located at Oak Ridge National Laboratory, is used by researchers around the world to investigate numerous phenomena, from weather prediction to the folding of proteins to, well, the explosions of massive, life-spewing stars.

Simulating the Stars

In the case of the CCSNs, the team simulated four different individual stars, each with a unique mass, as each star’s mass greatly affects the nature of its demise. The team’s findings are detailed in a paper in Volume 767 (10 April 2013) of Astrophysical Journal Letters.

While the team did in fact simulate four individual stars, they aren’t trying to reproduce individual explosions, but are trying to determine how CCSNs occur as a class of star, said team member Bronson Messer. The different masses shed light on the different phenomena that collectively contribute to a star’s fatal fireworks.

Messer and his colleagues made some serious headway in their quest to unravel one of the great remaining stellar mysteries. Among their most important findings: while not all of the masses share the same explosion conditions, all of them can be exploded via a combination of neutrino-heating and a phenomenon known as the standing accretion shock instability, or SASI. The SASI occurs when the shock wave expanding from a supernova’s core stalls.

Depending on its mass, several factors contribute to a star’s demise, but all of the simulated stars share the neutrino/SASI link, lending credence to the importance of these two particular forces, said Messer.

Another interesting finding: it’s long been known that most CCSNs, which must be eight solar masses or greater, evolve into neutron stars following their implosion. However, it was thought that CCSNs greater than 25 solar masses are just as likely to become black holes. The team’s simulations reveal that while a 25-solar-mass CCSN could indeed lead to the birth of a black hole, it’s far from a sure thing; it would most likely lead to a neutron star, meaning that most black holes come from much more massive stars, probably around the 40-solar mass range, said Messer.

The More Physics, the Better

While Kraken is indeed a monster computer, you also need a monster application to realistically portray the complex physics involved when a CCSN implodes and explodes.

The team’s monster of choice in this investigation is CHIMERA, a massively parallel multiphysics code especially designed to simulate CCSNs. Named after the Greek mythical creature composed of a goat, a lion, and a serpent, CHIMERA is one of only two three-dimensional codes in the world that account for the differing energies of neutrinos, thought to be a major player in the explosion of CCSNs. CHIMERA was developed with funding from the National Science Foundation’s PetaApps program. “This paper is the culmination of a lot of that work,” said Messer.

What makes these simulations especially important is their ab-initio nature, meaning they incorporate as much established physics as possible; they are literally based only on what we know.

However, there are still some uncertainties. The goal, said Messer, is to reproduce observations of a particular object. Unfortunately, when dealing with objects at such distances, our observations have obvious restrictions. They can, however, tell us a few things.

For instance, researchers know that in nature these stars are likely ovoid in nature. Because light from supernovas tends to be polarized, it wouldn’t make sense for them to be spherical.

And the fact that CCSNs more often than not produce neutron stars is another clue. Neutron stars have the highest proper motion in the galaxy, said Messer, meaning they generally move very rapidly in one direction. Once they are born they are “popped out” and flung in a certain direction, a phenomenon known as a “natal kick” and one that likely requires asymmetry.

However, the inner workings of a star are complicated, and what researchers can’t directly see or infer from the laws of physics must be estimated. “If you neglect any part of the physics, you run a real risk of getting the wrong answer, “said Messer. “Things that we couldn’t compute as well as we would have liked, we understand the approximations.”

Different teams of researchers are pursuing different angles in search of an ultimate answer. In the end, they all have to parameterize one thing or another. What Messer has on his side that many don’t is relativistic gravity, which reveals the potential for the gravitational well, or the potential gravitational field around the CCSN. This, in turn, tells Messer’s team how hard it is to push the exploded material out into space, among other things.

The Essential Leap to 3D

While the team’s recent discoveries are noteworthy, there is much to be done, said Messer, adding that incorporating the full 3-D physics of a CCSN is still impossible, even with a machine the size of Kraken, though the team can simulate individual aspects in 3D.

The leap to full 3D simulations will be crucial, as the SASI acts differently in three dimensions than in two, and since real CCSNs live in our 3D universe, “we know we have to do it,” said Messer. His team’s work on Kraken will no doubt prove indispensable in the final effort to solve the mystery behind one of the night sky’s most important players.

About NICS: The National Institute for Computational Sciences (NICS) operates the University of Tennessee supercomputing center, funded in part by the National Science Foundation. NICS is a major partner in NSF’s Extreme Science and Engineering Discovery Environment, known as XSEDE. The Remote Data Analysis and Visualization Center (RDAV) is a part of NICS.

Links to simulation results:

About NICS: The National Institute for Computational Sciences (NICS) operates the University of Tennessee supercomputing center, funded in part by the National Science Foundation. NICS is a major partner in NSF’s Extreme Science and Engineering Discovery Environment, known as XSEDE. The Remote Data Analysis and Visualization Center (RDAV) is a part of NICS.