Solving the relativistic magnetohydrodynamics equations with ADER discontinuous Galerkin methods, a posteriori subcell limiting and adaptive mesh refinement

TL;DR

This paper introduces a high-order, adaptive numerical scheme combining ADER discontinuous Galerkin methods, a posteriori subcell limiting, and mesh refinement for solving relativistic MHD equations with high accuracy and stability.

Contribution

It presents a novel numerical framework that integrates high-order ADER-DG schemes with subcell limiting and adaptive mesh refinement for relativistic MHD simulations.

Findings

Achieved convergence up to 5th order in space and time.

Successfully tested on shock tube, rotor, blast wave, and vortex problems.

Demonstrated potential for turbulent relativistic MHD flow simulations.

Abstract

We present a new numerical tool for solving the special relativistic ideal MHD equations that is based on the combination of the following three key features: (i) a one-step ADER discontinuous Galerkin (DG) scheme that allows for an arbitrary order of accuracy in both space and time, (ii) an a posteriori subcell finite volume limiter that is activated to avoid spurious oscillations at discontinuities without destroying the natural subcell resolution capabilities of the DG finite element framework and finally (iii) a space-time adaptive mesh refinement (AMR) framework with time-accurate local time-stepping. The divergence-free character of the magnetic field is instead taken into account through the so-called "divergence-cleaning" approach. The convergence of the new scheme is verified up to 5th order in space and time and the results for a set of significant numerical tests including shock tube problems, the RMHD rotor and blast wave problems, as well as the Orszag-Tang vortex system are shown. We also consider a simple case of the relativistic Kelvin-Helmholtz instability with a magnetic field, emphasizing the potential of the new method for studying turbulent RMHD flows. We discuss the advantages of our new approach when the equations of relativistic MHD need to be solved with high accuracy within various astrophysical systems.