The future of space plasma simulation

Space plasma simulation is a subject which is in its infancy, but which is already having an important impact on space science. Its growth is being spurred onward by the continuing increase in capacity (speed and memory) of computers and by advances in the sophistication of numerical models. These advances are making it possible to simulate more realistic situations using more complex models. Already significant three-dimensional MHD calculations of the magnetosphere and its interaction with the solar wind have been carried out. In addition multi-dimensional particle simulations are illuminating many of the microscopic physics processes which go on (instabilities, saturation levels and wave nonlinearity, shock structure, etc.). Notwithstanding these advances, the surface has only been scratched; many challenges and opportunities are provided by simulation both for the space physicist and the model builders (also for computer designers). In MHD models more physics need to be included (Hall effect, gyroviscosity, accurate models of boundaries, how do we put microscopic physics effects into macroscopic codes, etc.). For model builders correct treatments of systems containing a large range of important space and time scales, magnetic field strengths, Alfven wave velocity, etc. present real challenges. What are the best ways to diagnose these complex models and obtain meaningful information? What quantities should be looked at? How should they be displayed? A discussion of the promises, the prospects, and the challenges of the above topics will be given with examples taken from recent work.