Computational Astrophysics

首席研究员计划开展一项广泛的计划， 计算天体物理学的基础研究， 星系的动力学和演化 该计划包括一个 一个具有基础应用数学的重大项目 框架. 这个框架由一组非常强大的 分段抛物方法（PPM） 其允许快速且精确地解偏微分 超级计算机上的方程式 整体的一个重要部分 计划是继续发展这些强大的技术 在世界上最大的， 最快的电脑 将应用这些方法的研究领域是 星系的动力学和演化 在这个领域，计算机 模拟提供了处理强大的 高度复杂和非线性流体带来的困难 流动。 星系的星盘被视为流体， 四维相空间 这项工作的目标是获得 对所涉及的物理现象的新见解。 初步 结果表明，所提出的计算的承诺 为实现这一目标而努力。 使用超级计算机的新计算方法将使人们 制作真实的星系模型。 最终目标是 了解星系中发生的物理过程。 伍德沃德博士是一位非常有能力的科学家， 使用超级计算机的专业知识。 他将与 博士后研究员和研究生及本科生 进行研究。 新的数字代码和算法 将被开发。 重要的是，新的科学家 受过超级计算机技术训练

The Principal Investigator plans to carry out a broad program of basic research in computational astrophysics to study the dynamics and evolution of galaxies. The program consists of a major project which has an underlying applied mathematical framework. This framework consists of a group of very powerful numerical techniques called piecewise-parabolic methods (PPM) which permit rapid and accurate solution of partial differential equations on supercomputers. An essential part of the overall program is the continued development of these powerful techniques driven by demanding applications on the world's largest and fastest computers. The research area to which these methods will be applied is the dynamics and evolution of galaxies. In this area, computer simulations offer the hope of dealing with the formidable difficulties presented by highly complex and nonlinear fluid flows. Stellar disks of galaxies are treated as fluids in a four-dimensional phase space. The goal of the work is to obtain new insights into the physical phenomena involved. Preliminary results demonstrate the promise of the proposed computational approach in realizing this goal. The new methods of calculating using supercomputers will let one make realistic models of galaxies. The eventual goal is to understand the physical processes taking place in galaxies. Dr. Woodward is a very capable scientist who has considerable expertise in using supercomputers. He will work closely with postdoctoral associates and graduate and undergraduate students in carrying out the research. New numerical codes and algorithms will be developed. It is important that new scientists be trained in supercomputer techniques.