Kengo Tomida

Associate Professor, Graduate School of Science, Tohoku University *Profile is at the time of the award.

2024Inamori Research GrantsScience & Engineering

Research topics
Development of the Athena++ massively-parallel radiation magnetohydrodynamic code
Keyword
Summary
In astronomy, computational simulations play a crucial role because experiments are basically impossible. In order to simulate complex astrophysical phenomena, we need a high-performance simulation code which supports various physical processes such as hydrodynamics, gravity, magnetic fields, chemical reactions, and radiation. In particular, radiation transfer is a fundamental process common to all astrophysical phenomena, but it is difficult to simulate it in star and planet formation processes because the speed of light is very fast. This project aims to address this challenge by developing a high-performance radiation magnetohydrodynamic simulation code.

Message

It is often said that "computer simulations are telescopes for theoretical astrophysics." For good simulations, high-performance supercomputers are obviously necessary, but we need both good hardware and software. In this project, we will develop a high-performance "telescope" that anyone can use and contribute to research not only in astronomy but also in other fields.

Outline of Research Achievements

For astrophysical radiation hydrodynamic simulations, we have implemented a radiation solver based on the flux-limited diffusion approximation and an implicit method in the Athena++ code. We developed a solver that combines the Newton-Raphson method and the Full Multigrid method to solve large-scale nonlinear partial differential equations efficiently. Test calculations have been completed, and we have confirmed that parallel simulations are possible. In parallel, we compare numerical methods for astrophysical magnetohydrodynamic simulations, and also port the code to GPUs.

K. Tomida, et al. (2026) Systematic Comparison between Constrained Transport and Mixed Divergence Cleaning Methods for Astrophysical Magnetohydrodynamic Simulations arXiv 2605.07928


Find other recipients

Science & Engineering

PAGE TOP