Elsevier Science Home
Computer Physics Communications Program Library
Full text online from Science Direct
Programs in Physics & Physical Chemistry
CPC Home

[Licence| Download | New Version Template] abvu_v1_0.gz(15 Kbytes)
Manuscript Title: A numerical code for the phase-space boundary integration of water bag plasmas.
Authors: S. Cuperman, M. Mond
Catalogue identifier: ABVU_v1_0
Distribution format: gz
Journal reference: Comput. Phys. Commun. 21(1981)397
Programming language: Fortran.
Computer: CDC 6600.
Operating system: SCOPE 3.3.
RAM: 150K words
Word size: 60
Peripherals: magnetic tape.
Keywords: Plasma physics, Non-homogeneous Vlasov systems, Collisionless plasma, Water bag plasmas.
Classification: 19.3.

Nature of problem:
We consider the nonlinear evolution of nonhomogeneous Vlasov plasmas - a problem of great importance in both thermonuclear fusion research and astrophysics. In the past, only limited progress has been achieved in this direction, due to the complexity of the problem.

Solution method:
When the distribution function f(x,v,t) of a collisionless plasma is pictured as the density of an incompressible "phase fluid" moving in the two dimensional (x,v) phase space, by Liouville theorem, it is sufficient to follow the motion of the boundary curve(s) which enclose regions of constant f in phase space in order to know the state of the system. Thus, it is possible to investigate a system consisting of a very large number of particles (enclosed by a boundary curve) without having to treat them explicitly. The method was first used by Roberts and Berk for plasma systems. It has been further developed by Cuperman et al. for gravitational systems. This paper describes the adaptation (trasnformation) of the improved phase-space boundary integration code of Cuperman et al. to the plasma case.

The code may be used only for the investigation of one dimensional (two- dimensional phase space), collisionless plasma systems consisting of regions of constant density (in phase space).

Running time:
Execution times depend on the number of Eulerian strips used and on the the number of mark points required to describe the system with the desired accuracy. Typically, if the number of Eulerian strips used is 215, the execution time per time step for one mark point is about 1.5 X 10**-3 s.