Relativistic Vlasov-Maxwell modelling using finite volumes and adaptive mesh refinement

TL;DR

This paper introduces a relativistic Vlasov-Maxwell solver with adaptive mesh refinement that significantly improves computational efficiency for high-dimensional plasma simulations.

Contribution

It presents a novel Eulerian solver with block-structured adaptive mesh refinement for relativistic plasmas, enhancing efficiency and accuracy in high-energy tail modeling.

Findings

Achieved a 7x speed-up in laser-plasma interaction simulations.

Implemented high-order finite volume discretization with flux correction.

Demonstrated effectiveness of adaptive mesh in reducing computational cost.

Abstract

The dynamics of collisionless plasmas can be modelled by the Vlasov-Maxwell system of equations. An Eulerian approach is needed to accurately describe processes that are governed by high energy tails in the distribution function, but is of limited efficiency for high dimensional problems. The use of an adaptive mesh can reduce the scaling of the computational cost with the dimension of the problem. Here, we present a relativistic Eulerian Vlasov-Maxwell solver with block-structured adaptive mesh refinement in one spatial and one momentum dimension. The discretization of the Vlasov equation is based on a high-order finite volume method. A flux corrected transport algorithm is applied to limit spurious oscillations and ensure the physical character of the distribution function. We demonstrate a speed-up by a factor of 7x in a typical scenario involving laser-plasma interaction with an underdense plasma due to the use of an adaptive mesh.