W. Scales
Due to the profound effects of the earth's ionosphere and magnetophere on radio waves and therefore communication systems using radio waves, the structure and dynamics of these regions of the earth's upper atmosphere have been extensively studied since the 1930's. With the advent of unmanned space vehicles to provide high resolution observational measurements and high-powered computers for simulation and visualization of theoretical models, many of the critical unsolved and controversial issues are within grasp. The work at Virginia Tech involves studying three areas of upper atmospheric science: (1) creation and evolution of artificially produced ionic disturbances, (2) utilizing nonlinear processes produced by 'heating' the ionosphere with high-powered radio waves, (3) studying microscopic magnetospheric processes during highly active geophysical periods.
Large-scale simulation and visualization are currently being extensively used to study each of these problems. Several M.S. and Ph.D. students are involved in this work. The goals are the determination of the likely geophysical conditions for irregularity development, the spatial and temporal scales of the irregularities and ultimately an assessment of the effects of these irregularities on communication systems.
Due to the wide range of space and time scales involved in ionospheric and magnetospheric processes (eight orders of magnitude !), sophisticated numerical models must be developed that are very computationally intensive. Numerical methods at the state-of-the-art are utilized to simulate theoretical models which exhibit highly nonlinear, turbulent and chaotic behavior. The simulation models and computer codes are developed in house at Virginia Tech and include fluid dynamic, magnetohydrodynamic, Particle-in-Cell, and hybrid fluid-particle. Indeed, many of these models may serve as benchmarks for the computational ability of high performance computers. Also, since the trend is to study these processes in 2 and 3 spatial dimensions, visualization has become essential for interpreting the output of these numerical models. In the past, this research has made extensive use of super computers at Corneal University and Los Alamos National Laboratory.
The computer facilities at Virginia Tech have become inadequate at the present to support this research to its fullest capacity. This work will greatly benefit from the use of scalable architecture over a metacenter link with NCSA. Extensive use of visualization both on desktop workstations and in a CAVE environment will be essential for making physical interpretations of the physical processes being studied.