Augmenting a Novel Magnetohydrodynamics Code for Studying Astrophysical Plasmas and Space Weather, 15-R8568
Derek A. Lamb
Craig E. DeForest
Timothy A. Howard
Inclusive Dates: 07/01/15 – Current
Background — Magnetohydrodynamics (MHD) is the study of electrically conductive, magnetized fluids. Much of the Universe, including the solar atmosphere our group seeks to understand, exists in a state in which the equations of MHD apply. Real MHD systems are too complex for analytic solution, and so researchers turn to simulations of these systems. Conventional finite-difference plasma simulations are hundreds to thousands of times more dissipative (electrically resistive) than the real solar plasma, so conventional simulations use far finer grids than are necessary to represent the system under study. This makes simulations computationally expensive, and severely limits the fidelity with which conventional simulations can represent electric-current-bearing systems on the Sun, such as solar flares and the magnetically stressed regions that release coronal mass ejections (CMEs). With prior internal and external funding, we have developed Field Line Universal relaXer (FLUX) code, which reformulates the equations of MHD to take advantage of the analogy between magnetic field lines and the magnetic fields they represent. Our approach discretizes the magnetic field, automatically preserves the magnetic field topology, and results in a thousandfold increase in efficiency compared to conventional codes.
Approach — The objective of this project is to augment FLUX to enable it to solve two new classes of problem: CME eruption and solar wind evolution. All of the previous results obtained with FLUX have been under the approximation that the plasma gas pressure is negligible compared to the magnetic pressure. Enabling these new classes of problem requires three specific augmentations to FLUX: 1) verify and demonstrate the correct treatment of mass by the code in the case of quasi-stationary, quasi-one-dimensional flows; 2) change the modeling engine to allow for non-quasi-stationary flows by adding an inertial term to the force balance equation that FLUX solves; 3) verifying and refining the existing module that implements magnetic reconnection.
Accomplishments — Prior to the start of this project, the FLUX code had been dormant for several years. Our initial efforts have centered around implementing and testing the code on modern hardware, software, and development platforms, updating and improving the documentation, and bringing the existing code in line with modern coding standards. We have performed several relaxations of simple test systems while working to complete the first augmentation. The Figure shows one of these tests, a relaxation of a simple twisted magnetic flux rope.