Researchers recently tapped into the power of the SuperMUC-NG supercomputer to run the most detailed simulation yet of magnetic turbulence in the Milky Way. Their results offer new insight into the galaxy’s tangled magnetic field, and how it influences star formation and the movement of cosmic rays.1,2

Image Credit: BenSlater/Shutterstock.com
Importance of Galactic Magnetic Fields
Turbulence remains one of the major unresolved challenges in classical mechanics, appearing everywhere from a cup of coffee to the vast expanse of the interstellar medium (ISM). In astrophysical environments, magnetic fields play a key role in shaping turbulent flows. While the ISM contains far fewer particles than even the best Earth-based vacuum experiments, the motion of these particles still generates magnetic fields, much like those produced in Earth’s core.
Although the galactic magnetic field is millions of times weaker than a typical fridge magnet, it has a profound influence on the structure and evolution of the cosmos. In the coldest molecular phase of the ISM (T ≈ 10 K), the interplay between turbulence and magnetic fields impacts several key processes: it affects plasma ionization via cosmic ray diffusion, helps form the filamentary structures that define the conditions for star formation, and contributes to regulating star formation rates by providing support against gravitational collapse.
In this environment, the ISM plasma maintains magnetic fields that are in energy equipartition with turbulent motions, giving rise to organized patterns in both density and magnetic structure. These strong fields are maintained by a turbulent dynamo and, in turn, influence the cascade of turbulence by interacting with shear Alfvén modes and establishing correlations between magnetic fields and fluid velocity.1-3
Download the PDF of the article
Simulating Interstellar Magnetic Turbulence
To quantify cascade physics, turbulence must be simulated with maximum resolved scales on the largest grids. Hydrodynamical ISM-type turbulence has reached Reynolds number (Re) ? 106 on a 10,0483 grid, demonstrating two scale-separated power laws in the kinetic energy spectra, unlike the single Kolmogorov-type law. No such high-Re simulation exists for supersonic, magnetized turbulence. This raises a key question: how do magnetic fields modify the classic Burgers or Kolmogorov turbulence spectra? Much like in supersonic turbulence, there is growing evidence that at high magnetic Reynolds numbers (Rm), the energy cascade in three-dimensional (3D) magnetohydrodynamic (MHD) turbulence behaves differently, suggesting a fundamental shift in how energy is transferred across scales in the presence of strong magnetic fields.1
Large-Rm theories combine magnetic reconnection with turbulence, where reconnection-driven tearing instabilities alter sheet-like eddy cascades. To separate cascade from instability timescales, simulations must reach Rm ? 106, enabling thin current sheets from anisotropic eddies, which are common in ISM simulations, to become unstable via plasmoid instability. This triggers nonlinear fast reconnection when instability growth timescales are shorter than turbulence cascade timescales, resulting in a break scale in the cascade. Most high-resolution models that capture this multiscale magnetized turbulence are for subsonic, incompressible plasmas with uniform background magnetic fields, limiting applicability to the compressible ISM turbulence driven by a dynamo. Understanding spectra, energy flux, and scale hierarchies in highly compressible MHD turbulence remains an important goal. The Galaxy’s ISM is magnetized, compressible, and turbulent, affecting star formation, cosmic ray transport, and metal and phase mixing, yet basic turbulence statistics remain uncertain.1
Methodology
In a recent study, researchers simulated highly compressible, magnetized turbulence characteristic of the interstellar medium (ISM), using grid resolutions of up to 10,080³ cells. The magnetic field was sustained by a small-scale dynamo, and the simulations revealed two coexisting kinetic energy cascades: one involving non-local interactions at supersonic, weakly magnetized scales, and another governed by local interactions at subsonic, strongly magnetized scales.
The team used a modified version of the MHD code FLASH, enhanced with a highly optimized, hybrid-precision, positivity-preserving, second-order MUSCL-Hancock HLL5R Riemann solver to handle the compressible, ideal MHD equations in three dimensions. The simulations were run over a triply periodic domain at resolutions of 2,520³, 5,040³, and 10,080³, marking the largest grids ever applied to this turbulence regime.
A forcing term f was included in the momentum equation, following the standard approach for modeling driven, magnetized turbulence. These large-scale simulations were carried out under project 10391 at the Leibniz Supercomputing Centre in Garching, Germany, using the SuperMUC-NG supercomputer. The 10,080³ simulation, along with its power-spectrum analysis, required around 140,000 compute cores and approximately 80 million compute-core hours.
Turbulence was driven at a turbulent Mach number (M) of about 4, allowing for sufficient resolution across both supersonic and subsonic scales. The total magnetic field comprised a mean component and a turbulent component, the latter evolving self-consistently with the MHD turbulence throughout the simulation.1,4
Key Findings and Implications
This highly detailed, scalable simulation proved to be a powerful tool for studying magnetism and turbulence in the Milky Way’s interstellar medium (ISM). The model (a cubic domain with 10,000 units per side) provided significantly higher resolution than previous efforts. It could represent volumes up to 30 light-years across, while also being scalable down by a factor of 5,000 to explore smaller-scale phenomena.
At its largest scale, the simulation offers new insights into the structure and behavior of the galaxy’s magnetic field. At smaller scales, it supports investigations into localized processes such as the solar wind. Crucially, the model helps quantify how magnetic pressure counteracts gravitational collapse in star-forming regions, offering a clearer picture of the role magnetism plays in star formation.
One of the simulation’s major advancements is its ability to capture the ISM’s dynamic density range, from near-vacuum conditions to the denser environments found in nebulas, something earlier models struggled to represent accurately. By enabling detailed analysis of magnetic turbulence across such varied physical conditions, this simulation advances our understanding of cosmic processes that shape both galactic-scale structures and local space weather.1,2
Researchers demonstrated that the observed 3/2 spectrum in supersonic MHD turbulence can be explained by scale-dependent kinetic energy fluxes and velocity–magnetic field alignment. In highly magnetized modes, the magnetic energy forms a local cascade that deviates from any known ab initio theory. Using simulations at unprecedented grid resolutions up to 10,0803, approaching realistic Reynolds numbers for the cold ISM and exceeding those in warmer phases, two scale-separated cascades were revealed: a Burgers-type spectrum with M > 1, where kinetic-energy-dominated motions non-locally transfer energy across all scales, resembling supersonic hydrodynamical turbulence; and an Iroshnikov-Kraichnan (IK)-type spectrum with M < 1, featuring magnetically dominated, mostly incompressible motions undergoing a local cascade to smaller scales. The kinetic energy flux is scale-dependent across the cascade.1
In the cold ISM, the transition point in the spectrum corresponds approximately to filament widths. Turbulence becomes highly magnetized on scales smaller than a typical filament width, suppressing small-scale cloud fragmentation through added magnetic pressure. The shallower kinetic energy spectrum results in more turbulent support at small scales, thus suppressing star formation. In the warm ionized medium, strong magnetic fields generated by turbulence maintain energy equipartition across the volume, with local forward cascades in both kinetic and magnetic energy, aligned in a scale-dependent manner. With two distinct kinetic energy cascades, scale-dependent energy flux, and a single magnetic energy cascade, this work presents a new paradigm for compressible ISM turbulence maintained by a dynamo.1
These findings guide theoretical studies and can be tested by current and future radio surveys like ASKAP’s Polarisation Sky Survey of the Universe’s Magnetism (POSSUM) and the Square Kilometre Array (SKA), offering insight into ISM magnetism, star formation, and cosmic ray propagation.1,2
Conclusion
These high-resolution simulations provide a powerful new framework for studying compressible, magnetized turbulence in the interstellar medium. They significantly advance our understanding of galactic magnetic fields, star formation, and cosmic ray transport. Future studies could extend this work by incorporating cosmic ray feedback, non-ideal plasma effects, or the dynamics of star clusters.
This research highlights the critical role of high-performance computing (HPC) in space science, enabling the exploration of complex, multiscale processes and helping to bridge theoretical models with increasingly detailed astrophysical observations.
Want to find out what a blazar is? Read on here
References and Further Reading
- Beattie, J. R., Federrath, C., Klessen, R. S., Cielo, S., & Bhattacharjee, A. (2025). The spectrum of magnetized turbulence in the interstellar medium. Nature Astronomy, 9(8), 1195-1205. DOI: 10.1038/s41550-025-02551-5, https://www.nature.com/articles/s41550-025-02551-5
- Astrophysicists explore our galaxy’s magnetic turbulence in unprecedented detail using a new computer model [Online] Available at https://www.eurekalert.org/news-releases/1083449 (Accessed on 09 October 2025)
- Xu, S., Lazarian, A. (2022). Cosmic ray streaming in the turbulent interstellar medium. The Astrophysical Journal, 927(1), 94. DOI: 10.3847/1538-4357/ac4dfd, https://iopscience.iop.org/article/10.3847/1538-4357/ac4dfd/meta
- Federrath, C., Klessen, R. S., Iapichino, L., & Beattie, J. R. (2021). The sonic scale of interstellar turbulence. Nature Astronomy, 5(4), 365-371. DOI:10.1038/s41550-020-01282-z, https://www.nature.com/articles/s41550-020-01282-z
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.