Astrophysical plasmas differ from laboratory plasmas in that the system size is invariably >> the ion gyroradius. In astrophysics, these scales are typically modeled using MHD (e.g., using Jim Stone's MHD code ZEUS). There are a number of important problems, however, in which the system is collisionless even on scales >> the ion gyroradius. Examples include the solar corona and solar wind and accretion flows onto black holes and neutron stars. Of particular interest are kinetic modifications to turbulence and angular momentum transport in collisionless accretion flows onto black holes. CMPD will accelerate the first nonlinear studies of kinetic effects on the large-scale dynamics of these plasmas by supporting the development of required multiscale algorithms.

There are also important astrophysics problems involving gyrokinetic turbulence on scales comparable or smaller than the ion gyroradius - a topic that has been intensively studied in the magnetic fusion program. In astrophysics, this turbulence is excited by a nonlinear cascade from larger scale MHD turbulence. CMPD will accelerate the first detailed numerical simulations of particle heating and acceleration by such small-scale gyrokinetic turbulence in an astrophysics context. This will reveal how gyrokinetic turbulence heats particles in the solar wind, in relativistic outflows from compact objects, and in the hot collisionless plasmas ubiquitous around neutron stars and black holes.

Observations and modeling of relativistic outflows from astrophysical compact objects (rotation powered pulsars, jets from active galactic nuclei and gamma ray burst sources) have shown that these magnetohydrodymically driven systems must be dissipative - the observed asymptotic flow velocities are too large to be understood as the result of ideal MHD flow, but can achieve their high asymptotic flow Lorentz factors if the magnetic energy in the flow dissipates and is efficiently converted into flow kinetic energy before the flow end in a termination shock wave. All such flows have embedded current sheets, either in the form of sheet pinches or as force-free electromagnetic structures. Such structures are unstable to reconnection, through formation of relativistic tearing modes or through radiation and dissipation of kinetic Alfven waves. CMPD supports research with Prof. Arons of UC-Berkeley, related to the codes and to the understanding of reconnection physics developed in the Center to the community of astrophysicists interested in these problems.

Finally, Lathrop, et al., have recently observed the magnetorotational instability (MRI) in the laboratory. This is the first real observation of the MRI in any setting. The experiment involves driven rotation in liquid sodium. To understand the details of the MRI in the laboratory setting, one must be able to simulate low-viscosity plasma flows with important boundary-layer physics -- a classic multiscale problem. By eventually making available plasma physics simulation algorithms suitable for the simulation of multiscale dynamics, CMPD will gain the attention of the astrophysical community as the properties of the MRI are explored for the first time in the laboratory.