Home > Research > Supernova > SciDAC

SciDAC Supernova Initiative- Supernovae are nature's grandest explosions and an astrophysical laboratory in which unique conditions exist that are not achievable on Earth. Understanding these explosions and their signals is therefore important, not only because of their central role in astronomy and nucleosynthesis, but because a full understanding of supernovae may lead us to a better understanding of basic physics. Despite decades of research and modeling, no one understands, in detail, how supernovae work. The problem persists largely because, until recently, computer resources have been inadequate to carry out a credible multi-dimensional calculation. We have established a collaboration in computational astrophysics devoted both to solving the ``supernova problem" and to providing sufficient validation checks and community participation that the results will be accepted. Our team includes eight major institutions with participation by a half-dozen more, especially the MPA in Garching, Germany, the DOE/ASCI FLASH Center in Chicago, and the Joint Institute for Nuclear Astrophysics (JINA). Because the project will be challenging even on the largest computers possible today, and because of the difficulties in visualizing and handling large data flows, our team also includes a large fraction of experts in computer science. We will investigate both core-collapse and thermonuclear supernovae and will provide detailed nuclear and spectroscopic diagnostics of all our models. Predictions of neutrino bursts, gravitational radiation, and neutron star kicks will be byproducts. Efficient 3D simulation codes, optimized for massively-parallel computers, will be built, tested, and validated under this initiative.

Convection and the Mechanism of Core-Collapse Supernova Explosions: That in the protoneutron star and supernova contexts there should be hydrodynamic instabitilites (Rayleigh-Taylor, salt-finger, semi-convection) has been known and studied since the work of Epstein (1979). A review of this literature can be found in Burrows, Hayes, and Fryxell (1995, BHF). However, the role of convection and overturn has been controversial and ambiguous from the outset. Many, evoking Ockam's Ravor, have opted to ignore it. This should no longer be possible. There are three classes of instabilities to address: 1) those in the core below the neutrinospheres, 2) overturning and boiling motions due to heating from below between the gain radius and the shock (BHF; Herant et al 1994; Janka and Muller 1996), and 3) Rayleigh--Taylor and Richtmyer--Meshkov instabilities in the outer stellar mantle far beyond the ``iron'' core (Fryxell, Muller, and Arnett 1991). The sequence of stills shown here illustrates the second class and is from BHF. Those authors found that a core that did not explode in 1-D (spherical) did explode in 2-D when the Rayleigh-Taylor (convective) instabilities were allowed to manifest themselves.

Adam Burrows, Evonne Marietta, and Bruce Fryxell completed a study of the hydrodynamic interaction of a supernova blast wave with a secondary stellar companion. Using a 2-dimensional hydrocode, they initiated an investigation of the impulses, the mass stripping, and the post--impact state of the secondary, as well as the distribution in velocity space of the stripped companion debris. Since hydrogen is present on many proposed secondaries, but not in the observed spectra, these calculations will allow them to discriminate between various Type Ia and Type Ib progenitor systems. Shown is a still in the middle of the interaction of a Type Ia supernova explosion with a putative main sequence companion. Go to Type Ia web page for more information.

New data imply that the average velocity of radio pulsars is large. Under the assumption that these data imply that a pulsar is born with an ``intrinsic'' kick, we have investigated whether such kicks can be a consequence of asymmetrical stellar collapse and explosion. Burrows and Hayes (1996) concluded that they can. The neutron star residue recoils in the direction opposite to the direction in which the ejecta preferentially emerge. In addition, they calculated the gravitational wave (GW) signature of such asymmetries due to anisotropic neutrino radiation and mass motions. They predicted that any recoils imparted to the neutron star at birth will result in a gravitational wave strain that does not go to zero with time. Hence, there may be ``memory'' in the gravitational waveform from a protoneutron star that is correlated with its recoil and neutrino emissions. In principle, the recoil, neutrino emissions, and gravitational radiation can all be measured for a galactic supernova.

The neutrino pulse is the best diagnostic of the physics of collapse, explosion, and neutron star/black hole birth. The neutrino spectra, flavor mix, and time evolution are stamped with the various hydrodynamic phases through which the core and shock wave progress (Burrows, Klein, and Gandhi 1992). The detailed characterization of the neutrino burst would provide ground truth for any numerical model and would be a central means for code and theory validation. In 1987, IMB and Kamioka II detected a scant 19 anti-neutrino events from SN1987A in the LMC (Bionta et ali. 1987; Hirata et al. 1987). The duration (~10 seconds) and event number were consistent with basic theory (Burrows and Lattimer 1987), but the SN1987A data were mute on the central issues of the mechanism of explosion and the character of the dynamics. A detailed neutrino signature is a byproduct of each of our current core-collapse simulations. A goal of our future precision 1D, 2D, and 3D simulations will be to obtain a reliable family of neutrino light curves and spectra (for all neutrino species) as a function of progenitor mass and degree of rotation. For instance, Fryer and Heger (2000) have recently suggested that rotation might significantly effect the mu neutrino luminosity, making it a useful index of the angular momentum of the collapsing core. A network of highly sensitive and massive underground neutrino telescopes (including Super-Kamiokande, SNO, LVD, Borexino, KamLand, ICARUS, and a possible new U.S. facility) that could collect thousands of supernova neutrino events from a galactic supernova has been established (SNEWS; Habig et al. 2000).

Updated on: January 17, 2009