Three-Dimensional Core-Collapse Supernova Simulations with Multi-Dimensional Neutrino Transport Compared to the Ray-by-Ray-plus Approximation

TL;DR

This study compares fully multi-dimensional and ray-by-ray-plus neutrino transport in 3D supernova simulations, finding that both methods produce similar averaged results, supporting the use of the simpler RbR+ approximation in 3D modeling.

Contribution

First comprehensive 3D supernova simulations comparing multi-dimensional neutrino transport with the RbR+ approximation, demonstrating their similar outcomes in averaged quantities.

Findings

Good agreement between FMD and RbR+ in 3D simulations.

RbR+ can artificially promote explosions in 2D but not in 3D.

Higher spatial resolution impacts results more than transport differences.

Abstract

Self-consistent, time-dependent supernova (SN) simulations in three spatial dimensions (3D) are conducted with the Aenus-Alcar code, comparing, for the first time, calculations with fully multi-dimensional (FMD) neutrino transport and the ray-by-ray-plus (RbR+) approximation, both based on a two-moment solver with algebraic M1 closure. We find good agreement between 3D results with FMD and RbR+ transport for both tested grid resolutions in the cases of a 20 solar-mass progenitor, which does not explode with the employed simplified set of neutrino opacities, and of an exploding 9 solar-mass model. This is in stark contrast to corresponding axisymmetric (2D) simulations, which confirm previous claims that the RbR+ approximation can foster explosions in 2D in particular in models with powerful axial sloshing of the stalled shock due to the standing accretion shock instability (SASI). However, while local and instantaneous variations of neutrino fluxes and heating rates can still be considerably higher with RbR+ transport in 3D, the time-averaged quantities are very similar to FMD results because of the absence of a fixed, artificial symmetry axis that channels the flow. Therefore, except for stochastic fluctuations, the neutrino signals and the post-bounce evolution of 3D simulations with FMD and RbR+ transport are also very similar, in particular for our calculations with the better grid resolution. Higher spatial resolution has clearly a more important impact than the differences by the two transport treatments. Our results back up the use of the RbR+ approximation for neutrino transport in 3D SN modeling.