We present self-consistent, axisymmetric core-collapse supernova simulations performed with the Prometheus-Vertex code for 18 pre-supernova models in the range of 11-28 solar masses, including progenitors recently investigated by other groups. All models develop explosions, but depending on the progenitor structure, they can be divided into two classes. With a steep density decline at the Si/Si-O interface, the arrival of this interface at the shock front leads to a sudden drop of the mass-accretion rate, triggering a rapid approach to explosion. With a more gradually decreasing accretion rate, it takes longer for the neutrino heating to overcome the accretion ram pressure and explosions set in later. Early explosions are facilitated by high mass-accretion rates after bounce and correspondingly high neutrino luminosities combined with a pronounced drop of the accretion rate and ram pressure at the Si/Si-O interface. Because of rapidly shrinking neutron star radii and receding shock fronts after the passage through their maxima, our models exhibit short advection time scales, which favor the efficient growth of the standing accretion-shock instability. The latter plays a supportive role at least for the initiation of the re-expansion of the stalled shock before runaway. Taking into account the effects of turbulent pressure in the gain layer, we derive a generalized condition for the critical neutrino luminosity that captures the explosion behavior of all models very well. We validate the robustness of our findings by testing the influence of stochasticity, numerical resolution, and approximations in some aspects of the microphysics.

Interactions with neutrons and protons play a crucial role for the neutrino opacity of matter in the supernova core. Their current implementation in many simulation codes, however, is rather schematic and ignores not only modifications for the correlated nuclear medium of the nascent neutron star, but also free-space corrections from nucleon recoil, weak magnetism or strange quarks, which can easily add up to changes of several 10% for neutrino energies in the spectral peak. In the Garching supernova simulations with the Prometheus-Vertex code, such sophistications have been included for a long time except for the strange-quark contributions to the nucleon spin, which affect neutral-current neutrino scattering. We demonstrate on the basis of a 20 M_sun progenitor star that a moderate strangeness-dependent contribution of g_a^s = -0.2 to the axial-vector coupling constant g_a = 1.26 can turn an unsuccessful three-dimensional (3D) model into a successful explosion. Such a modification is in the direction of current experimental results and reduces the neutral-current scattering opacity of neutrons, which dominate in the medium around and above the neutrinosphere. This leads to increased luminosities and mean energies of all neutrino species and strengthens the neutrino-energy deposition in the heating layer. Higher nonradial kinetic energy in the gain layer signals enhanced buoyancy activity that enables the onset of the explosion at ~300 ms after bounce, in contrast to the model with vanishing strangeness contributions to neutrino-nucleon scattering. Our results demonstrate the close proximity to explosion of the previously published, unsuccessful 3D models of the Garching group.

The neutrino emission characteristics of the first full-scale three-dimensional supernova simulations with sophisticated three-flavor neutrino transport for three models with masses 11.2, 20 and 27 M_sun are evaluated in detail. All the studied progenitors show the expected hydrodynamical instabilities in the form of large-scale convective overturn. In addition, the recently identified LESA phenomenon (lepton-number emission self-sustained asymmetry) is generic for all our cases. Pronounced SASI (standing accretion-shock instability) activity appears in the 20 and 27 M_sun cases, partly in the form of a spiral mode, inducing large but direction and flavor-dependent modulations of neutrino emission. These modulations can be clearly identified in the existing IceCube and future Hyper-Kamiokande detectors, depending on distance and detector location relative to the main SASI sloshing direction.

During the stalled-shock phase of our 3D hydrodynamical core-collapse simulations with energy-dependent, 3-flavor neutrino transport, the lepton-number flux (nue minus antinue) emerges predominantly in one hemisphere. This novel, spherical-symmetry breaking neutrino-hydrodynamical instability is termed LESA for "Lepton-number Emission Self-sustained Asymmetry." While the individual nue and antinue fluxes show a pronounced dipole pattern, the heavy-flavor neutrino fluxes and the overall luminosity are almost spherically symmetric. Initially, LESA seems to develop stochastically from convective fluctuations, it exists for hundreds of milliseconds or more, and it persists during violent shock sloshing associated with the standing accretion shock instability. The nue minus antinue flux asymmetry originates mainly below the neutrinosphere in a region of pronounced proto-neutron star (PNS) convection, which is stronger in the hemisphere of enhanced lepton-number flux. On this side of the PNS, the mass-accretion rate of lepton-rich matter is larger, amplifying the lepton-emission asymmetry, because the spherical stellar infall deflects on a dipolar deformation of the stalled shock. The increased shock radius in the hemisphere of less mass accretion and minimal lepton-number flux (antinue flux maximum) is sustained by stronger convection on this side, which is boosted by stronger neutrino heating because the average antinue energy is higher than the average nue energy. Asymmetric heating thus supports the global deformation despite extremely nonstationary convective overturn behind the shock. While these different elements of LESA form a consistent picture, a full understanding remains elusive at present. There may be important implications for neutrino-flavor oscillations, the neutron-to-proton ratio in the neutrino-heated supernova ejecta, and neutron-star kicks, which remain to be explored.

The VERTEX code is employed for multi-dimensional neutrino-radiation hydrodynamics simulations of core-collapse supernova explosions from first principles. The code is considered state-of-the-art in supernova research and it has been used for modeling for more than a decade, resulting in numerous scientific publications. The computational performance of the code, which is currently deployed on several high-performance computing (HPC) systems up to the Tier-0 class (e.g. in the framework of the European PRACE initiative and the German GAUSS program), however, has so far not been extensively documented. This paper presents a high-level overview of the relevant algorithms and parallelization strategies and outlines the technical challenges and achievements encountered along the evolution of the code from the gigaflops scale with the first, serial simulations in 2000, up to almost petaflops capabilities, as demonstrated lately on the SuperMUC system of the Leibniz Supercomputing Centre (LRZ). In particular, we shall document the parallel scalability and computational efficiency of VERTEX at the large scale and on the major, contemporary HPC platforms. We will outline upcoming scientific requirements and discuss the resulting challenges for the future development and operation of the code.

The first full-scale three-dimensional (3D) core-collapse supernova (SN) simulations with sophisticated neutrino transport show pronounced effects of the standing accretion shock instability (SASI) for two high-mass progenitors (20 and 27 M_sun). In a low-mass progenitor (11.2 M_sun), large-scale convection is the dominant nonradial hydrodynamic instability in the postshock accretion layer. The SASI-associated modulation of the neutrino signal (80 Hz in our two examples) will be clearly detectable in IceCube or the future Hyper-Kamiokande detector, depending on progenitor properties, distance, and observer location relative to the main SASI sloshing direction. The neutrino signal from the next galactic SN can therefore diagnose the nature of the hydrodynamic instability.

The relevance of the standing accretion shock instability (SASI) compared to neutrino-driven convection in three-dimensional (3D) supernova-core environments is still highly controversial. Studying a 27 Msun progenitor, we demonstrate, for the first time, that violent SASI activity can develop in 3D simulations with detailed neutrino transport despite the presence of convection. This result was obtained with the Prometheus-Vertex code with the same sophisticated neutrino treatment so far used only in 1D and 2D models. While buoyant plumes initially determine the nonradial mass motions in the postshock layer, bipolar shock sloshing with growing amplitude sets in during a phase of shock retraction and turns into a violent spiral mode whose growth is only quenched when the infall of the Si/SiO interface leads to strong shock expansion in response to a dramatic decrease of the mass accretion rate. In the phase of large-amplitude SASI sloshing and spiral motions, the postshock layer exhibits nonradial deformation dominated by the lowest-order spherical harmonics (l=1, m=0,-1,+1) in distinct contrast to the higher multipole structures associated with neutrino-driven convection. We find that the SASI amplitudes, shock asymmetry, and nonradial kinetic energy in 3D can exceed those of the corresponding 2D case during extended periods of the evolution. We also perform parametrized 3D simulations of a 25 Msun progenitor, using a simplified, gray neutrino transport scheme, an axis-free Yin-Yang grid, and different amplitudes of random seed perturbations. They confirm the importance of the SASI for another progenitor, its independence of the choice of spherical grid, and its preferred growth for fast accretion flows connected to small shock radii and compact proto-neutron stars as previously found in 2D setups.

Core-collapse supernovae are among the most fascinating phenomena in astrophysics and provide a formidable challenge for theoretical investigation. They mark the spectacular end of the lives of massive stars and, in an explosive eruption, release as much energy as the sun produces during its whole life. A better understanding of the astrophysical role of supernovae as birth sites of neutron stars, black holes, and heavy chemical elements, and more reliable predictions of the observable signals from stellar death events are tightly linked to the solution of the long-standing puzzle how collapsing stars achieve to explode. In this article our current knowledge of the processes that contribute to the success of the explosion mechanism are concisely reviewed. After a short overview of the sequence of stages of stellar core-collapse events, the general properties of the progenitor-dependent neutrino emission will be briefly described. Applying sophisticated neutrino transport in axisymmetric (2D) simulations with general relativity as well as in simulations with an approximate treatment of relativistic effects, we could find successful neutrino-driven explosions for a growing set of progenitor stars. First results of three-dimensional (3D) models have been obtained, and magnetohydrodynamic simulations demonstrate that strong initial magnetic fields in the pre-collapse core can foster the onset of neutrino-powered supernova explosions even in nonrotating stars. These results are discussed in the context of the present controversy about the value of 2D simulations for exploring the supernova mechanism in realistic 3D environments, and they are interpreted against the background of the current disagreement on the question whether the standing accretion shock instability (SASI) or neutrino-driven convection is the crucial agency that supports the onset of the explosion.

Following a simulation approach of recent publications we explore the viability of the neutrino-heating explosion mechanism in dependence on the spatial dimension. Our results disagree with previous findings. While we also observe that two-dimensional (2D) models explode for lower driving neutrino luminosity than spherically symmetric (1D) models, we do not find that explosions in 3D occur easier and earlier than in 2D. Moreover, we find that the average entropy of matter in the gain layer hardly depends on the dimension and thus is no good diagnostic quantity for the readiness to explode. Instead, mass, integrated entropy, total neutrino-heating rate, and nonradial kinetic energy in the gain layer are higher when models are closer to explosion. Coherent, large-scale mass motions as typically associated with the standing accretion-shock instability (SASI) are observed to be supportive for explosions because they drive strong shock expansion and thus enlarge the gain layer. While 2D models with better angular resolution explode clearly more easily, the opposite trend is seen in 3D. We interpret this as a consequence of the turbulent energy cascade, which transports energy from small to large spatial scales in 2D, thus fostering SASI activity. In contrast, the energy flow in 3D is in the opposite direction, feeding fragmentation and vortex motions on smaller scales and thus making the 3D evolution with finer grid resolution more similar to 1D. More favorable conditions for explosions in 3D may therefore be tightly linked to efficient growth of low-order SASI modes including nonaxisymmetric ones.

The quest for the supernova explosion mechanism has been one of the outstanding challenges in computational astrophysics for several decades. Simulations have now progressed to a stage at which the solution appears close and neutrino and gravitational wave signals from self-consistent explosion models are becoming available. Here we focus one of the recent advances in supernova modeling, the inclusion of general relativity in multi-dimensional neutrino hydrodynamics simulations, and present the latest simulation results for an 11.2 and a 15 solar mass progenitor. We also mention 3D effects as another aspect in supernova physics awaiting further, more thorough investigation.