Another interesting article that I found on the International Science Grid This Week (isgtw) is “Astrophysics team simulates key supernova phase at unprecedented resolution,” in which a research team simulated a turbulent flame is a Type IA supernova outburst at unprecedented resolution.
Type Ia supernovae occur only in binary systems that contain a dense white dwarf and a companion star. It is thought that the supernova outbursts are caused by gravity induced transfer of material from the companion to the white dwarf and this material tips the white dwarf’s core into a fusion explosion. Previous research has indicated that this outburst does not take place immediately. A team led by Chris Malone (UC Santa Cruz) have made a lot of progress in understanding this phenomenon of delayed response through use of high performance computational simulations of the evolution of the outbursts and have published their results in the February 10 issue of the Ap J.
In an earlier study, the team modeled the early phases on ignition with the Maestro code developed in collaboration with Lawrence Berkeley National Lab in California, US. They found that the ignition does occurs not at the center of the white dwarf, but is in fact slightly off center and this has a big impact on the evolution of the supernova explosion.
The team subsequently used these earlier calculations as the initial conditions for a high-resolution simulation on the Blue Waters supercomputer, but this time using the Castro hydrodynamics code, written in C++ and Fortran, and based on the BoxLib software framework supernovae. The thermonuclear “flame” takes a second to reach the white dwarf’s surface, but the flow is convective and turbulent and is therefore computationally demanding . This is why the team used the Blue Waters supercomputer (and the Titan system at Oak Ridge National Lab in Tennessee, US). See the image:
The ISGTW article takes up the story:
“We wanted to know, does this background flow from convection affect the explosion as it moves through the star?” Malone says. “With such a large machine at our disposal, we triggered a ‘flame’ that propagates through the star. Prior to this, people triggered an explosion in the star but without a realistic convective flow pattern. We found that for a typical ignition location, the convective roiling doesn’t really affect the flame as it makes its way to the surface.”
The team continues to analyze the terabytes of data derived from their ongoing simulations on Blue Waters, and intends to explore other aspects of the supernovae explosion. For example, when the flame breaks through the surface of the star, it flings out material much like lava from a volcano. Some of the material escapes, but the flame itself continues to burn around the surface of the star.
“As this ‘lava’ of star material is moving very rapidly – almost at the speed of sound – across the surface of the star, there’s a lot of shear and mixing going on between the hot material and the cooler material of the star,” Malone says. “We’re looking at this highly turbulent, highly shear-driven burning to see if that triggers another explosion.”
This post is based closely on an article in the International Science Grid This Week entitled: “Astrophysics team simulates key supernova phase at unprecedented resolution.” The research paper referenced is “THE DEFLAGRATION STAGE OF CHANDRASEKHAR MASS MODELS FOR TYPE Ia SUPERNOVAE. I. EARLY EVOLUTION” byC. M. Malone, A. Nonaka, S. E. Woosley, A. S. Almgren, J.B.Bell, S.Dong and M. Zingale. 2014, Ap J 782, 11. Download from http://m.iopscience.iop.org/0004-637X/782/1/11/pdf/0004-637X_782_1_11.pdf (Note: you will need account to access the paper.)