Meet Ivy Wong: finding galaxies in the data

By  Rachel Rayner 17 May 2024 5 min read

When Dr Ivy Wong plays 'spot the difference', she does it on a cosmic scale. 

Ivy works with telescopes like our ASKAP radio telescope, one of a new generation of telescopes. ASKAP creates enormous maps of the Universe by scanning – or surveying – the sky again and again. There can be more that 3 million galaxies in one survey! Astronomers play ‘spot the difference’ with these survey maps, analysing how distant stars and galaxies are flashing, changing or moving.

These surveys are getting more and more detailed, which means data volumes are getting bigger and bigger. Researchers are finding traditional methods aren't up to the task of analysing such large volumes of data. Ivy is on a mission to find innovative solutions to this astronomical challenge. 

Asking questions of galaxies beyond ours

Ivy was one of those curious kids who loved to ask ‘Why?’ and wanted to figure out how things worked.

“So, I always knew I wanted a science-related career,” Ivy said. 

During her second year studying at the University of Melbourne, she completed a summer project using Murriyang, our Parkes radio telescope.

As a young physics student, she had little to no experience with the software and computers being used. She had also never been trusted with the safety of 1000 tonnes of world-class radio telescope. It was only after her first week passed without incident that she began to get comfortable. This might be a career worth pursuing, after all.

“I love both science and technology, so radio astronomy was an easy sell to me. I was inspired by the engineering of the telescope and its receiver system.

"So, while the rest of the world celebrated the turn of the millennia, I was looking after a large-scale survey project at the Parkes Observatory.”

A few years later, Ivy’s PhD involved completing a chunk of this very same survey project. It mapped the hydrogen gas in galaxies, which Ivy still thinks is fascinating.

“The idea of detecting hydrogen atoms in other galaxies is pretty cool. Hydrogen is the smallest element and these galaxies are so far away,” she said.  

Now, as a Science Leader in our Space and Astronomy team, Ivy is still looking at hydrogen atoms and mapping the shape of distant galaxies. She works with computer scientists to help her better understand the data from a new generation of telescopes. And the science is still exciting.

New ways for studying the ancient Universe

“As radio astronomers, we’re hoping to ask and answer the impossible questions,” she said.

And there’s always new questions forming, which haven’t been thought of before. There are many phenomena to be discovered because radio astronomy is one of the newest and youngest observational sciences in astronomy itself."

Radio astronomy developed after World War II, when radar equipment was directed at the stars instead of enemy vessels. No one had considered that objects in the Universe may emit radio waves too. 

Now, we have supercomputers gathering and processing data from radio telescopes.  Machine learning analyses that data to identify anything unusual. Developments in AI are assisting too. Ivy is right there to steer it all into a new era. 

Unlike optical telescopes, which capture visible light with mirrors and cameras, radio telescopes gather low-energy radio light and turn it into electrical signals. The more advanced the radio telescopes get, the more signals they can collect from space. The signals must undergo a complex process of data management and refinement before they can be converted into graphs or images. For researchers, the process is akin to sifting through sand to find tiny shells.

Rationing the radio data

Our ASKAP radio telescope generates data at the rate of 100 trillion bits per second – more data at a faster rate than Australia’s entire internet traffic. It is supercomputers, like Setonix at the Pawsey Supercomputing Research Centre, which then stores, calibrates and transforms this data. 

“If we allowed our hardware to capture as much information as we can, we can get up to 100 petabytes or more a year. That’s 100 million gigabytes!" Ivy said. 

“So, we’re not fully recording everything that we could because we simply don’t have that capacity to process such vast volumes of data at once.”

Citizen science is one avenue for collecting the information astronomers need. Ivy worked on an online galaxy classification project where the public were invited to look at images of galaxies and highlight features of interest.

“While this was a lot of fun and a great way to get absolutely anyone involved in radio astronomy, it was still resource intensive.”

Ivy believes that advanced algorithms, such as those used in machine learning, could be crucial in capturing as much information as possible from the Universe. This approach circumvents the need to store massive quantities of data.

“My focus for the next couple of years is trying to apply machine learning methods to the surveys. Can we do it? What will we discover?”

While Ivy is experimenting with machine learning methods for radio astronomy, it is nothing new to a computer scientist.

“It is where computer science meets astronomy – it’s a multidisciplinary field bringing a lot of great and diverse minds together,” Ivy said.  

Lots to love in the Universe – finding others to love it too

“What fascinates me is that to answer these big questions you need great teams that work well together.”

Ivy’s work with students and early career researchers is all part of the process. She knows there are those who will come after her with new and game-changing ideas that will push astronomy research even further.

“We have very curious people. We have very bright kids with the capabilities to see things we don’t. They’ll build on the new methods we have started here.”

“What I like to see is we celebrate scientific talent as much as we celebrate talent in other areas like sporting ability and artistic skill. Not just because these people will discover new and interesting things, but because so much of our world is thanks to science and technology.”

We acknowledge the Wajarri Yamaji as the Traditional Owners and native title holders of Inyarrimanha Ilgari Bundara, the CSIRO Murchison Radio-astronomy Observatory site, where ASKAP is located, and the Wiradjuri people as the Traditional Owners of the Parkes Observatory site, where Murriyang is located.