HR: 1330h
AN: SH13B-04 [Abstracts]
TI: MHD Simulations of Coronal Plasma Driven by Boundary Flows.
AU: * Tokman, M
EM: mayya@math.berkeley.edu
AF: Mayya Tokman, 1091 Evans Hall
Dept of Mathematics
University of California, Berkeley, Berkeley, CA 94720 United States
AU: Hsu, S C
EM: scotthsu@lanl.gov
AF: Scott Hsu, Los Alamos Natl. Lab
ÿP.O. Box 1663, Los Alamos, NM 87545 United States
AB:
A large portion of theoretical work on describing the large scale plasma evolution in solar and laboratory plasmas relies on
Taylor's theory of relaxation which predicts that a plasma configuration will relax to a constant α
profile, i.e., a force-free state with a constant in space coefficient of proportionality
between the magnetic field and current density J = α B. Observations, however, show
that α in the active region is a complex function of space (e.g. Zhang et al., 2001).
In particular, the spatial profile of α consists of adjacent patches of opposite sign.
We present a numerical MHD model of the dynamics of the active region plasma under
photospheric boundary flows which addresses questions associated with the spatial structure
of α. The simulations reveal that the evolution proceeds
through a formation of adjacent, nearly force-free regions with parallel and anti-parallel
magnetic fields and current densities, i.e. with α of opposite signs. We present the
details of the model, explain the structure of the obtained solutions and the reason for the
formation of such configurations. The boundary conditions are modeled by imposing a velocity
profile and deriving consistent conditions on the magnetic field from the MHD equations. In order to
address the numerical difficulties associated with the stiffness of the resistive MHD equations we use
new exponential propagation iterative (EPI) methods which allow accurate time integration with
time steps exceeding the CFL restriction on explicit schemes.
DE: 0644 Numerical methods
DE: 7509 Corona
DE: 7524 Magnetic fields
SC: SPA-Solar and Heliospheric Physics [SH]
MN: 2005 Joint Assembly