Structure-preserving long-time simulations of turbulence in magnetised ideal fluids

Key findings from this study

• The study found that reduced magnetohydrodynamics and Hazeltine's model both generate magnetic dipoles and exhibit inverse magnetic energy cascades over long timescales.
• The researchers demonstrate that vorticity field evolution differs substantially between models despite shared Hamiltonian structure, with reduced magnetohydrodynamics producing sharp filaments while the other models show minimal variation.
• The authors report that kinetic energy inverse cascades appear in Hazeltine's model and the Charney-Hasegawa-Mima equation but not in reduced magnetohydrodynamics.

Overview

The study examines three two-dimensional magnetohydrodynamics models: reduced magnetohydrodynamics, Hazeltine's model, and the Charney-Hasegawa-Mima equation. These models capture essential features of magnetohydrodynamic turbulence and plasma dynamics through non-canonical Hamiltonian formulations. The authors employ structure-preserving discretisations via the matrix hydrodynamics approach to maintain phase space geometry during long-time numerical simulations. This geometric preservation is critical because the infinite conservation laws and symplectic structure inherent to these models govern statistical behavior over extended timescales.

Methods and approach

The researchers implemented structure-preserving discretisations for all three models using the matrix hydrodynamics approach. This method maintains the non-canonical Hamiltonian structure and symplectic properties of the underlying phase flow. Long-time simulations utilized randomised initial data across all three models. The authors performed spectral analysis to examine cascade behavior and compared results systematically between models to identify consistent versus divergent dynamical features.

Results

The study found magnetic potential dynamics converge across reduced magnetohydrodynamics and Hazeltine's model, both producing magnetic dipoles over extended timescales. Spectral scaling diagrams reveal an inverse cascade of magnetic energy and mean-square magnetic potential in both models. Vorticity field evolution diverges significantly: reduced magnetohydrodynamics develops sharp vortex filaments with rapidly intensifying vorticity, whereas Hazeltine's model and the Charney-Hasegawa-Mima equation exhibit only small vorticity variations. The latter two models exhibit spectral signatures of kinetic energy inverse cascade absent in reduced magnetohydrodynamics.

Implications

Structure-preserving discretisations prove essential for capturing qualitative features in long-time magnetohydrodynamics simulations. The phase space geometry maintained by Hamiltonian formulations directly influences turbulent cascade mechanisms and statistical equilibria. Models sharing Hamiltonian structure do not necessarily exhibit identical vorticity dynamics, indicating that other structural features modulate small-scale momentum transfer. The divergence between reduced magnetohydrodynamics and the other models suggests different effective dissipation mechanisms or constraint hierarchies govern kinetic energy evolution.

Scope and limitations

This summary is based on the study abstract and available metadata. It does not include a full analysis of the complete paper, supplementary materials, or underlying datasets unless explicitly stated. Findings should be interpreted in the context of the original publication.