ITR/(DMS): A Computational Environment for Multi-scale MHD Turbulence Studies

University Of Maryland, College Park, College Park MD

Investigators

Abstract

Modeling MHD (magneto-hydrodynamic) turbulence has become increasingly important both in terrestrial and engineering applications, such as fusion devices and plasma propulsion, as well as in space and astrophysical environments such as the solar wind and astrophysical jets. The magneto-hydrodynamic nature of the plasma fluid imposes constraints on the numerical procedures involved. Likewise, modeling in the turbulent regime -- for turbulence is ubiquitous -- also imposes constraints and challenges on the numerical procedures employed. The numerical approach and demands of computation time and accuracy translate into algorithmic (or software) advances which can avail the hardware -- itself advancing rapidly. In this project these advances entail: Adaptively refined grids, where fine resolution only in regions of interest reduces overall CPU requirements; MHD shock capturing methods, which improve accuracy of the calculation; parallel computing techniques, which take advantage of the availability of multiple processors and memory and cache hierarchy; and sub-grid models, which allow coarser grids than otherwise for turbulent regimes. The analysis of the resulting large multi-block multi-processor data requires advanced data visualization tools to examine the physics, the raison d'etre for the calculations. The project will develop a computational environment, consisting of a 3D MHD simulation code with adaptive grid refinement capabilities, complex time-dependent boundary conditions and optimized to run under MPI on various parallel computers, and a visualization system that can handle multi-block multi-processor MHD AMR data. In this end-to-end modeling and discovery process, information technology (IT) plays a critical role. From software engineering principles that result in robust codes, and smarter data structures that are required for AMR, to single-node optimization and better communication strategies on parallel architectures, to visualization tool development and implementation, IT makes possible knowledge discovery through simulation in a more efficient and orderly fashion. The requirements for MHD turbulence modeling being so considerable, this efficiency of IT translates into enabling the science or technology. I.e. without the application of these IT principles, physically important and interesting regimes would not be analyzed. The results of this project will significantly enhance the modeling capabilities of diverse flows and will specifically be applied to the problem of heliospheric magnetic field configuration and solar wind turbulence. 3D MHD simulations of the heliospheric fields and flows that include rotation, shear, waves, turbulence, current sheets, pressure-balanced structures, and interaction regions, which thus represent most of the major physical effects of importance, will be performed. The effort will enable future studies on particle propagation and magnetic field configuration for "space weather" effects on Earth. The project will directly train a post-doctoral scholar and a graduate student in relevant physics and computational mathematics; project results will be communicated to the community via conference and journal publications and the web, and through classes taught by the PI.