Title: From galaxy formation to the Earth formation

Speaker: Dr. Takayuki Saitoh (ELSI)

Abstract:
In these five years, I developed several new numerical schemes and codes in order to carry out studies of the Earth formation and evolution. In my talk, I briefly explain my recent progress and then introduce research projects of next several years using them.

Research project 1. Galaxy formation and galactic habitable zone study.
According to the Big Bang theory, at the beginning of the Universe, there are only hydrogen, helium and a small amount of Li. All elements heavier than helium except the primordial Li are synthesized via stellar evolution, supernovae, and neutron star mergers. The enrichment history of these heavy elements, i.e. metals, is deeply related to the galaxy formation. Hence, only galaxy formation simulations involving chemical evolution is suitable for studying the origin of elements and the metal redistribution manner in a galaxy. I tackle this topic using my N-body/SPH code, ASURA, and the special library for this purpose, CELib (Saitoh 2017).

While there are many interesting topics regarding the chemical evolution of galaxies, here I apply it for the galactic habitable zone (GHZ). In the classical model, the GHZ is the region with sufficiently high metallicity and low SN rate to form planetary systems hosting terrestrial planets. However, the dynamical evolution of a galaxy is not considered in the model. It is known that stars in the galactic disk have complicated motions (i.e., epicyclic motion and radial migration), they occasionally encounter spiral arms where the local star formation rate is high, and their metals depend on the local enrichment history. These are also important to measure the habitability. Using a high-resolution cosmological galaxy formation simulation involving chemical evolution, I reproduce realistic environments where stars exist and improve the GHZ. This information regarding environmental changes would be useful to study the formation and evolution of stellar systems and planets where life can emerge/survive.

Research project 2. Mantle convection simulation.
Mantle convection is the fundamental mechanism of the thermal and geological evolution of the Earth. The numerical treatment is crucial because of its non-linear nature. In general, (extended) Boussinesq and incompressive approximations are applied for mantle simulation. However, in such simulations, the global communication is required and consequently, the performance is very low in massively parallel computers. Multi-dimensional simulations involving a large deformation are difficult in these standard simulations.

In order to carry out high-performance simulations with massively parallel computers, we developed a novel numerical scheme for mantle convection, VIM (Takeyama, Saitoh & Makino 2017), and then we start the VIM++ project last year. The primary aim of our VIM++ project is to solve the whole history of the Earth from the initial condition obtained from GI simulation. This project integrates novel schemes (VIM and CPHSF; Yamamoto & Makino 2017), software framework (FDPS; Iwasawa, Hosono, Makino et al. 2016), and a realistic initial condition (GI simulations with magma ocean; Hosono, Karato, Saitoh, & Makino in prep.). I believe that this is the only project which has a potential to solve the full, complicated history of the Earth with numerical simulation. So far, this is a developing phase. Thus, I explained the structure of the project, the current status, and the project milestones.