1991 Fiscal Year Final Research Report Summary
Numerical Simulations of Collisions of Galaxies by Special Purpose Computer
| Project/Area Number |
02640197
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Astronomy
|
|---|
| Research Institution | College of Arts and Sciences, University of Tokyo |
|---|
Principal Investigator |
EBISUZAKI T. College of Arts and Sciences, Univ of Tokyo, Assoc. Prof., 教養学部, 助教授 (10183021)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
OKUMURA S. K. (奥村 幸子) College of Arts and Sciences, Univ. of Tokyo. Assist. Prof., 教養学部, 助手 (20224842)
MAKINO J. College of Arts and Sciences, Univ of Tokyo. Assist. Prof., 教養学部, 助手 (50229340)
SUGIMOTO D. College of Arts and Sciences, Univ of Tokyo. Prof., 教養学部, 教授 (10022592)
|
|---|
| Project Period (FY) |
1990 – 1991
|
| Keywords | Elliptical Galaxies / Special Purpose Computer / Black Holes / Galaxy Formation |
|---|
| Research Abstract |
In 1990, we purchased a workstation(U-300T)made by Sumitomo Electric Industries and connected it with the GRAPE-1, special purpose computer for gravitational N-body simulations.
In 1991, we developed the advanced system, GRAPE-3, in which 48 full-custom LSIs works in parallel. We connected GRAPE-3 with U-300T and achieved an effective speed of 10 Gflops. This speed well exceeds that of the most advanced super computers. Using this GRAPE-3 system- we performed many numerical simulations of collisions of galaxies.
We found the followings.
First, the dynamical properties of elliptical galaxies are well explained in terms of the mergings of galaxies. We found that the results of the previous simulations are not appropriate because of their small number of particles and their biased initial conditions. We also found that the central black holes well explain the positive correlation between core radius and luminosity of elliptical galaxies.
Second, the process in the mergings of galaxies are well explained in terms of waveparticle interactions and phase mixing. We concluded that the so called violent relaxation does not exist, since these two processes are quite different from the relaxation process.
|
