Letters of Intent received in 2016

LoI 2018-1910
Computational Discovery in Astrophysics (GA or Non-GA Symposium)

Topics

* Fundametal physical processes:
- Gravity, radiative transport, hydrodynamcs;
* stars and planets:
- Formation, evolution;
- Star cluster formation, structure and evolution;
- Planetary system formation, evolution and stability;
- Compact object formation, interaction and gravitational waves;
* Computing:
- Data stewardship and code maintenance;
- Machine learning and supercomputing methods;
- Computational complexity and emergent behavior;
- Integration strategies and accuracy;
- Reproduceability.

Rationale

Astrophysics on digital computers is a relatively new discipline
compared to optical telescopes. The objective of the symposium is to
bring together those scientists that use the computer for scientific
discovery. We would like to share our knowledge on approaches, ideas
and methods; to assess the current level, requirements, bottle-necks
and future prospects of numerical simulations; to ensure the needed
level of synergy between various models.

This includes the combined efforts of modeling by solving
computationally demanding problems (gas dynamics, MHD, radiation
transfer, N-body, data processing, etc.), data intensive computing,
machine learning.

We will focus on applying computational methods to automate scientific
activities, such as finding laws from observational data, mine
scientific data, which focuses on building highly predictive models.

Computational astrophysics is not just a specialization of
computational physics, but it opens new windows in the way we perceive
and study the heavens. This rapidly growing (relatively) new
discipline in astronomy combines modern computational methods, novel
hardware design, advanced algorithms, original software
implementations and associated technologies to discover new phenomena,
and to make predictions in astronomy. Computational astrophysics
touches on many topics in astronomy but is not recognized within the
IAU as a separate discipline. This is understandable considering the
lack of a central entity. With the proposal to establish a commission
on Computational Astrophysics we hope to change that.

During the symposium we will bring together experts to discuss a broad
palette of challenging informational and technical developments,
including hardware and software but always focused on the
astronomical applications. Special attention will be given to
numerical integration schemes, data management and emergent behavior.
These three topics are rather novel for astronomy and have been growing
dramatically since the introduction of digital computers in research.