By Ahmed Suliman
It’s the technology that could help find missing pieces of physics and answer conundrums like dark matter.
The Square Kilometre Array is a radio telescope project with an Australian component to which the federal government has recently committed $387 million. About $64.4 million of this will go towards establishing a centre in Perth to process and analyse SKA data: a collaboration between the International Centre for Radio Astronomy Research (ICRAR), CSIRO, and the Pawsey Supercomputing Centre.
Pelican sat down with Professor Peter Quinn, ICRAR executive director, to discuss the organisation’s work, his career, and how the technology behind the Square Kilometre Array could change astronomy as we know it.
Ahmed Suliman: Thank you for speaking to Pelican today, Professor Quinn. Firstly, how did ICRAR come about?
Peter Quinn: ICRAR was founded 12 years ago, as a joint venture between UWA and Curtin University. We receive funding from them as well as the state and federal governments. At that time the decision about where to put the SKA had not been decided, and so Australia was keen to invest in things that would attract to bring the project here, and that included a new international research centre.
From the perspective of the WA government, ICRAR was also designed to serve as a landing pad to capture the benefits of the project locally when it arrived in Western Australia. This included jobs, knowledge, research, and opportunities for industry.
AS: What does ICRAR do on a day-to-day basis?
PQ: We are not a traditional academic centre, in the sense that we do more than just astronomy research. Our core skills span across astronomy, engineering, and data science. We have around 120 staff and 90 graduate students working in those three areas across our two research nodes in UWA and Curtin.
Our astronomy research is facilitated through a network of telescopes in Australia and around the world, including not only radio, but optical and space telescopes as well. Astronomy is a very multi-wavelength science, so we look across the spectrum to piece together how the universe works.
In terms of engineering, we are helping build the systems for the SKA, including detectors, receivers, and computing systems. Given we are based in Perth, we are relatively close to the Murchison Radio Astronomy Observatory, which allows us to go to the site, deploy equipment, and test ideas.
The data science side involves working with the large volumes data, especially the data management systems. ICRAR is responsible for designing a fairly big piece of the SKA’s data system. We work with data centres and software engineers around the world, including some of the biggest supercomputers available.
AS: How was it that you came to be involved with ICRAR? Tell us a bit about your career.
PQ: Like all careers in astronomy, maybe in science generally, it is a very zig-zag path. I was born and raised on the east coast of Australia. I was always interested in science, and so I did an undergraduate degree in physics and mathematics at the University of Wollongong, then went on to ANU to do a PhD in astronomy. Like many do at that point in their career, I left Australia to work overseas, starting out in the United States working with organisations like Caltech and Space Telescope Institute in NASA as a research astronomer. My passion was primarily computational astrophysics.
I then had the opportunity to go to Germany to work at the European Southern Observatory. I was asked to start up a new division at this organisation around 1995. This division ended up becoming very large, and designed and ran two huge observatories. This taught me a lot about doing astronomy at a much different scale than before, building and working with big projects.
Australia had not traditionally been involved with big observatory projects. When the Square Kilometre Array began considering Australia as a suitable location, it combined my interest in working on big projects with my desire to return to Australia after 27 years abroad. In 2006, I was granted a WA Premier’s Fellowship to come back to work on the SKA. It took a few years until the need for ICRAR became clear, and I became the first director in 2009. Here we are 12 years and 200 people later.
AS: Many of our readers would have at least tangentially heard about the SKA, but for those who haven’t: can you briefly explain what it is, and what it is aiming to achieve?
PQ: The Square Kilometre Array is a radio telescope, which receives radio waves from the universe. There are two radio telescopes working together to do this. One is in Australia, which receives low frequency radio waves of around 100MHz, what you would expect from FM radio. These radio waves are received by stationary Christmas tree-shaped antennas. Those signals are processed by computers, and converted into images of the low frequency sky.
The other is in South Africa, which works with waves ten times higher in frequency — around 1GHz — which you would typically see in a mobile phone. This uses more traditional dishes which can move around and collect those higher frequencies.
The Australian telescope (called SKA Low) in particular will assist in studying an interesting period of the early universe: the so-called “cosmic dawn”. This was around 0.5 to 1 billion years after the Big Bang, where the universe cooled down sufficiently to condense into matter, forming the seeds of galaxies, stars, et cetera. We are lucky in astronomy because we can look backwards in time, as the further you look away, the longer the light takes to get to you. So if you look very far away, you can see faint objects that date back to those early phases of the universe.
Why a square kilometre? Those extremely old and distant objects are very faint, so you need a big eyeball to see them. If you work out how much collecting area is required, it works out to about a million square metres, or a kilometre squared, of radio astronomy equipment spread out over multiple locations. The radio waves coming from the “cosmic dawn” period in particular are around the FM band, so a little bit of your FM radio feed contains signals from the edge of the universe! These signals will be captured by the Australian section of the SKA and help find the first objects in the universe.
AS: When I think of unexpected interstellar radio waves, the accidental discovery of the Wow! Signal (an intense 72-second radio signal from the Sagittarius constellation) in 1977 comes to mind. Does the SKA make it more likely that we will uncover more signals like that?
PQ: I think that is exactly right. For example, the (relatively) small antennas we have on site at the moment are already finding objects producing huge explosions in the universe that only last for fractions of a second. We had no idea those existed until the antennas started capturing those signals.
One aspect these telescopes will assist with is probing the nature of the “transient sky”, or how objects in the sky change and move over time, through frequent imaging. Most of our current knowledge is centred around static representations of the sky.
AS: From a number of perspectives, the technology behind the SKA sounds incredible. This includes the data flow: 7-10 terabits of data per second, or the equivalent of the entire Netflix back catalogue 10,000 times over. I assume this will be transferred somewhat quicker than your average home broadband speed?
PQ: Just a little bit! In terms of the Australian telescope, in the first phase we will have about 130,000 antennas, each with its own fibre optic connection. This will be one of the largest fibre optic networks in the world. Over the course of the day, exabytes of data are flowing around this network in the desert, comparable to the data volumes on the internet.
To manage this, you need to compress and process this raw data very quickly, and convert it into more compact data types and proto-images of the sky. Even then, you are still storing large image data that needs to be examined by scientists. All this requires dedicated computing, which is provided through a fibre optics cable that links the Murchison site directly with the Pawsey Supercomputer in Perth, 800kms away. In effect, the supercomputer becomes part of the telescope 24/7.
This data then goes on to scientists around the world, who need supercomputers of their own to scan through the output from Pawsey. To assist with this, we are in the process of developing a global network of data and processing hubs, known as the SKA regional centres, that those scientists can then access locally.
The data problem is interesting in two ways: one is how we develop the algorithms to deal with such vast quantities of data. The second is how supercomputers are now active participants in doing astronomy work. This role could expand into working with machine learning to identify objects and write code down the track.
AS: What sorts of questions are you looking forward to answering through the SKA?
PQ: For me, looking at this epoch when the universe was born opens a door to look at how things changed across cosmic time. Given its temporal focus, SKA will act as a great historian of the universe. Understanding these changes could lead to finding missing pieces of physics that could answer conundrums like dark matter and dark energy. Dark matter in particular really annoys me – that we still haven’t found that stuff. However, we do have theoretical candidates, some of which a radio telescope like the SKA would be able to detect.
The one we all talk about over a glass of wine is finding ET, of course! This will be a very sensitive set of ears to listen to the universe. If there are any intelligent signals out there (especially in our galaxy), the SKA has a very good chance of finding them.
AS: How can readers get involved with the ICRAR and the SKA?
PQ: There are multiple paths for getting involved. One of those is citizen science, which we do quite a bit, getting the general public involved in working with scientific data doing things like classifying galaxies. For students, we offer work experience for high schoolers as well as vacation scholarships for undergraduate students to come in for two months over the summer break, where they can work on a project and maybe even write a paper. There are masters and PhD programs available as well. We have some of the very best researchers in the world who act as graduate supervisors, who keep our students engaged and motivated about their work.
We also host an annual 5000 person festival of astronomy that we run in WA called Astrofest (13th November this year). All the information is available on our website.
Artwork by Pauline Wong