Data-Constrained Simulation of Coronal Mass Ejection Propagation

Overview

Coronal mass ejections (CMEs) are powerful expulsions of plasma from the surface of the sun that propagate towards Earth and interact with Earth's magnetosphere. The resulting "space weather" can produce beautiful auroras, but can also cause failures in power grids, satellite electronics and GPS systems, and danger to astronauts in space. To increase our ability to mitigate these problems, we must learn to predict where, when, and how a CME will originate and how it will change as it propagates through the heliosphere on its way through the solar system. NASA's goal to improve space weather prediction is spurring the development of ever more realistic simulations of CME propagation. The results of such simulations can be validated through comparisons to observations from various NASA spacecraft missions.

Project Details

Using the Space Weather Modeling Framework (SWMF) global magnetohydrodynamics (MHD) code, we run simulations of a CME initiated by a realistic, 3D nonlinear force-free magnetic field model of solar active region NOAA 11060, which erupted on April 8, 2010. A set of initial conditions (ICs) with different amounts of free magnetic energy—which become more and more unstable—is explored. To study the effect of the ICs on the structure and evolution of the CME ejecta, 3D variables of all plasma parameters—such as velocity, density, temperature, and magnetic field—are produced as a function of position and time, in addition to simulated white light and extreme ultraviolet emissions that have been observed by space-based telescopes.

Results and Impact

We determined the effect of eight different ICs on the morphology and plasma characteristics of the CME ejecta, and issued a best-fit model to the coronagraph observations of the event. For example, increasing the axial flux in the magnetic flux rope of the active region increases the speed of the ejecta: our ICs span speeds from 300 to 2500 kilometers per second (km/s).

• The best-fit model has axial flux of 11x1020 maxwells (units of magnetic flux, or Mx) and produces a CME velocity of 740 kilometers per second, which matches observations.
• Velocity and height curves are derived and then matched to those derived from observations.
• Simulated white light images reproduce the deflection of the CME toward the equator, as observed by NASA spacecraft.

This project presents the most realistic CME propagation simulations ever produced, and has the potential to increase the ability of such simulations to help predict space weather.

Why HPC Matters

SWMF is a massively parallel code that requires high-performance computing for production runs. This project would not have been possible without NASA's Pleiades supercomputer. Each simulation ran for at least 16 days on Pleiades using a minimum of 1,000 processors, producing about 35 minutes of CME propagation per run; a total of 1.5 high-frequency propagations were run for all eight ICs. These long runs were necessary to accomplish the high resolution required to resolve any shocks and diagnose any artifacts, as this is the first CME that we have analyzed with this code and these realistic ICs.

What's Next

The next step is to run more simulations of CMEs that have interesting energetic particle or magnetic field orientations to Earth, by propagating the CMEs one astronomical unit through interplanetary space—to Earth—using adaptive mesh.