Alumni Project

Supernova Science Center (SNSC) Accomplishments and Plans

Stan Woosley, UCSC
Adam Burrows, University of Arizona
Chris Fryer, LANL
Rob Hoffman, LLNL (plus 21 others)

During the first 18 months of support, the SNSC has made significant advances in understanding both thermonuclear and core-collapse supernovae. Our immediate goal is a 2D-model for a core-collapse supernova with good neutrino physics and equation of state. Studies of turbulent nuclear combustion and gamma-ray bursters will also be pursued.

In recent developments at the SNSC, the first complete three-dimensional (3D) calculation of the complete evolution of a core-collapse supernovae was carried out by our LANL team last spring (Fig. 1). The resultant energy was typical of observed supernovae in this mass range. However, the calculation can be criticized for its simple treatment of neutrino transport and the equation of state.

figure 1

Figure 1 velocity iso-surfaces in the 3D calculation of a 15 solar mass supernova 60 ms after core bounce (NERSC).

The University of Arizona is presently working on grid-based supernova codes in 1, 2, and 3 dimensions. The 1D version has been completed and will serve as a spherically symmetric benchmark as we move to higher dimensions. Currently, the Arizona team is completing work on a 2D-workhorse -VULCAN/2D. This is a multi-group, multi-angle (SN), moving-grid, arbitrary Lagran-gian-Eulerian (ALE) scheme that also incorporates rotation (Fig. 2).

figure 2

Figure 2 Collapse of a rotating supernova code calculated using the VULCAN-2D code, but without neutrino transport (NERSC).

In anticipation of a successful supernova shock coming out of the core, Zingale and Woosley at UCSC have been studying the hydrodynamical mixing that occurs in the mantle and envelope using a variant of the FLASH code. Two-dimensional studies in Cartesian coordinates with adaptive mesh reveal a cascade of fine structures reminiscent of actual supernova remnants.

Type Ia supernovae are also being studied by the UCSC team using Glatzmaier’s 3D- anelastic code. A fascinating result is that the convective flow in the white dwarf, just prior to ignition is dipolar (Fig. 3). That is, the center of the pre-supernova star is not stagnant, but characterized by high-speed circulation. This suggests the Type Ia supernova will be ignited off center.

figure 3
Figure 3 Convective velocities in a white dwarf near runaway calculated using a 3D anelastic code. The plot is a projection of the entire surface evaluated at constant radius. Yellow is upwards flows; blue down. Note the dipole asymmetry (NERSC).

At LLNL, the nuclear data library essential to studies of nucleosynthesis in massive stars has been assembled and made available to the community. In collaboration with other SNSC team members, these rates have been incorporated into leading edge studies of stellar nucleosynthesis.

Our highest current priority is a 2D model of a core collapse supernova with realistic neutrino physics. This is being pursued on two fronts, at LANL using SPH and at Arizona using the VULCAN code. In the long run (3D), we may want to use the LANL RAGE code or the Chicago FLASH code because of their higher degree of optimization, but we believe that realistic 2D results can be achieved in 2003 with the present apparatus. A major improvement will need to be the equation of state.

In addition we plan to:

  1. Continue the development of 3D models for core collapse – with full neutrino transport. Full up 3D models will not occur until the third year or later.
  2. Compare results on micro-zoned flame propagation studies at LBNL with the group at the MPA. Is turbulence greatly deforming the flame as the white dwarf expands to lower density?
  3. Work with JINA on the nuclear database needed to compute energy generation and element production in massive stars. Carry out a survey of nucleosynthesis in stars of different masses and initial compositions.
  4. Carry out 3D studies of "collapsars," a different sort of supernova, where the explosion is powered by a relativistic jet generated by a rotating black hole at the star's center.

figure 4

Figure 4 3D special-relativistic calculation of a jet passing through and exploding the outer layers of a massive star (NERSC).

We estimate that our computer needs to carry out these projects will be roughly 4 million MPP hours per year. We also obtain appreciable time for our research at LANL and on Beowulf clusters at the University of Arizona and UCSC.

For further information on this subject contact: Stan Woosley
Department of Astrophysics, UCSC
Phone: 831-459-2976
woosley@ucolick.org
http://www.supersci.org/

back to project page

 


Home  |  ASCR  |  Contact Us  |  DOE disclaimer