Introduction - ASIAA CASA Development Center (ACDC)

中研院天文所通用天文軟體研發中心(簡稱'通用天文軟體研發中心')

CASA stands for Common Astronomy Software Applications. It is a package developed by an international consortium of scientists based at the National Radio Astronomical Observatory (NRAO), the European Southern Observatory (ESO), the National Astronomical Observatory of Japan (NAOJ), the CSIRO Australia Telescope National Facility (CSIRO/ATNF), and the Netherlands Institute for Radio Astronomy (ASTRON) under the guidance of NRAO. It is designed to support the data post-processing needs of the next generation of radio astronomical telescopes such as ALMA and JVLA. It can later be expanded to be an ultimate package for all radio astronomical telescopes.

On 2016 August 1, we established the ASIAA CASA Development Center (ACDC), in collaboration with NRAO, with the funds of the Ministry of Science and Technology (MoST). The primary goals are to increase our international collaboration and visibility, advance our high-performance computing techniques and software engineering, and raise our science productivity from ALMA and other radio interferometer observations.

ALMA is now close to its full capability and has started to deliver high-fidelity (HIFI) astronomical images, with excellent sensitivity and resolution. Indeed, the resolution has already exceeded that of the Hubble Space Telescope earlier 2015. For example, ALMA images already show unprecedented details of a proto-planetary disk in the star-forming system HL Tau and an Einstein ring in the gravitational lens system SDP.81. The former image shows unexpected concentric rings in the disk in the early phase of star formation, revolutionizing our understanding of planet formation. The latter image allows us to constrain the mass of the central supermassive black hole in the distant lensing galaxy, which could not be done previously with optical telescopes. We expect to see many more of these kind of revolutionary HIFI images in near future.

In addition, ALMA data are extremely rich, many spectral lines from molecular, atomic, and ionized gas can be observed and detected simultaneously with the continuum in one single observation, due to the wide-band receivers and the high-sensitivity of ALMA. It is expected that ALMA will discover new species and complex molecules (e.g., amino acid) that are the building-blocks of life. Moreover, due to its high-sensitivity, ALMA can also map the magnetic fields in various astronomical settings, e.g., the interstellar media, star-forming regions, evolved stars, and galaxies, allowing us to determine the role of magnetic fields in various settings in great detail.

Undoubtedly, such rich and HIFI data from ALMA comes with a very big data size. For comparison, the data size of the existing same-type interferometry telescope, the Submillimeter Array (SMA), is at most a few GB, but the ALMA data size can easily be a few hundred GB and even TB when ALMA reaches its full capability. Therefore, ALMA data really quests for high-performance computing. In addition, ALMA data also quests for a multi-function multi-dimension high-quality visualization and efficient data analysis tools for efficient science production.

The CASA infrastructure consists of a set of C++ tools bundled together under an iPython interface as a set of data reduction tasks. This structure provides flexibility to process the data via a task interface or as a python script. As we know, C++ is a widely used standard programming language in the industries. On the other hand, iPython is a modern script language that has quickly gained its popularity not only in academic research but also in industries. In order to prepare our staff and students for their research and jobs, IAA held a Python crash course earlier in 2015, and more than 100 participants from all over Taiwan attended the course. This indicates the great importance of this scripting language. Joining the CASA center would bring our staff and students up to the international level of software development.

The NRAO was well aware of these programming opportunities, but has not been able to address them owing to lack of staff. As a result, we took this opportunity and established the ACDC, in collaboration with the NRAO. Our long-term (10-year) goal is to establish a world-recognized CASA development center and increase our science productivity. The staff in the center will also be trained to have an international level of software development and system managing skills for high-performance computing and visualization. These skills will be extremely useful in both the industry and the astronomy community in the modern era, in which data size becomes extremely big. The center can also be tied to the ARC (ALMA Regional Center) node here in IAA and our work is expected to enhance the ARC functions in, e.g., data reduction and analysis, and thus increase the science productivity.