ASKAP’s first all-sky survey opens a new era of discovery

When ASKAP was first designed over a decade ago, its goal was to conduct large-scale surveys as quickly as possible. Today, we realised that goal with the publication of ASKAP’s first all-sky survey. With ASKAP’s wide field of view, we have mapped 3 million galaxies in 300 hours, providing astronomers with a new atlas of the universe.

The Rapid ASKAP Continuum Survey (RACS) grew out of calculations suggesting that ASKAP’s survey speed was fast enough to match the Sydney University Molonglo Sky Survey (SUMSS) in about a day. We quickly realised how helpful it would be to commission not just the telescope, but the entire end-to-end survey workflow, using a short-duration all-sky survey.

To maximise the scientific potential of this rapid survey, we stretched a day into two weeks and devised a survey plan consisting of 903 tiles covering 83% of the sky. Each tile was observed for 15 minutes at a centre frequency of 888 MHz, reaching a typical noise limit of 250 uJy. RACS was conducted by the observatory as a service to the entire community.

The survey description and first results from RACS were published in PASA today, with about 40% of the data currently available for open access on CASDA and the rest soon to be released. RACS updates our knowledge of the southern radio sky and provides a key starting point for sky models that will improve calibration methods.

Learning more about ASKAP

The first RACS fields were observed in April 2019, with subsequent observations and re-observations spread thinly over several months. Multiple processing passes, extensive quality control and detailed analysis occupied about a year and a half, also an extremely short timescale for an all-sky survey.

Being foremost a commissioning exercise, RACS taught us much about how to best use ASKAP. We developed the ability to schedule short fields close to transit to improve image quality, learnt how to combine multiple tiles efficiently into larger maps and how to convolve multiple beams and tiles to a common resolution. We also learnt a few things about the telescope itself, such as the need for more accurate models of the primary beams near the edges of ASKAP’s extremely large field of view. Many of the resulting improvements have already been deployed on the telescope.

Imaging 83% of the sky

The radio sky contains a wide variety of sources, the density and nature of which depend primarily on whether we observe the plane of the Milky Way, or at high Galactic latitudes. One of our goals for RACS was to provide images across the entire visible sky, with no zones of avoidance for areas with more complex emission. The final map (Figure 1) shows the result. Although some linear artefacts remain along the Galactic plane and around the brightest sources, most of the map is remarkably clean. Imaging was performed independently on each beam without peeling or direction-dependent calibration, providing the foundations on which further improvements can be built. Even so, the result is impressive.

The first RACS data release improves on the sensitivity, resolution and image quality of previous surveys covering a similar area at this frequency and paints a promising future for ASKSP. We have detected about a million previously undiscovered radio sources. In addition, ASKAP’s 36 antennas provide excellent surface brightness sensitivity, revealing extended emission even with 15-minute observations.

Figure 1: RACS sky map shown with ASKAP antennas and the observatory landscape in the foreground. Individual objects of interest are marked and shown at the full resolution of the survey as inset panels. The boundaries of each tile are outlined in white.

Quality control and data availability

As the first all-sky survey with a new telescope, we needed to thoroughly examine all aspects of the RACS data. A quality control team consisting of CASS staff and collaborators at the University of Sydney checked astrometry, flux distributions, source counts and many other statistics to ensure the final images were free from systematic errors and ready for science. The first data release consists of images, calibrated visibilities (with 10 second time resolution and 1 MHz frequency resolution) and single-tile catalogues. In a future publication, we will provide a global catalogue covering the entire survey area.

The future of RACS

While the global catalogue for the 888 MHz RACS pass is under construction, we plan to observe a second pass at a higher frequency, covering roughly 1152-1440 MHz. This will eventually be extended with a third pass going up to 1800 MHz. Future epochs at these frequencies may also be scheduled amidst deeper survey project observations if this can be achieved efficiently.

RACS is the new benchmark atlas of the Southern sky and its publication today demonstrates that ASKAP can achieve its primary mission. ASKAP’s long-term survey science projects will extend our knowledge even further, discovering tens of millions of new radio sources over the next decade.

Resources and links

The RACS survey description paper can be found here: https://doi.org/10.1017/pasa.2020.41

We have constructed a web site describing the survey here: https://research.csiro.au/racs

Science-ready data is available on CASDA: https://data.csiro.au/collections/domain/casdaObservation/results/PRAS110%20-%20The%20Rapid%20ASKAP%20Continuum%20Survey/DAall/

A virtual tour is available here: https://www.atnf.csiro.au/research/RACS/RACStour/index.html