Magnetohydrodynamics plays an important role in the study of ionized gas, or plasmas, and in many space and astrophysical processes. The equations are highly nonlinear, require advanced computers to solve, and benefit from well-defined formulism to understand and separate the distinct processes involved with large-scale turbulence. Researchers from Colorado, Hamburg, and Michigan showed recently that such a separation is possible in the case of compressible magnetohydrodynamic turbulence. They report their finding in Physics of Plasmas.
The technique separates the spectrum of turbulence into “shells” representing particular wavelengths and extends a scale-by-scale energy transfer accounting previously used for incompressible magnetohydrodynamics. Lead author Philipp Grete says there were no real mathematical challenges to applying it to the compressible case. However, Grete notes that some prior works claimed that energy transfers within the magnetic energy budget are inseparable from transfers between the velocity and magnetic fields. Working with the newly proposed decomposition, Grete and his coauthors show that both a kinetic and a magnetic cascade exist also in the compressible regime and that transfers from one shell to another are predominately local.
To examine the scale separations, the researchers produced high-resolution simulations in the order of several terabytes per simulations, as well as millions of hours of computational time. The researchers ran the computations on the NASA Pleiades Supercomputer through computational grants, using open-source tools Enzo, Athena and a parallel Fast Fourier Transform library as well as their newly developed analysis framework.
The researchers consider this paper a “proof of concept” and intend to replicate the study in different regimes.
Source: “Energy transfer in compressible magnetohydrodynamic turbulence,” by Philipp Grete, Brian W. O’Shea, Kris Beckwith, Wolfram Schmidt, and Andrew Christlieb, Physics of Plasmas (2017). The article can be accessed at https://doi.org/10.1063/1.4990613.
