The radio telescope network will consist of thousands of antennas linked together by fibre and divided into two separate arrays: one made of up to 1 million low frequency antennas in Australia and one made of up to 2500 mid frequency dishes in South Africa, providing continuous coverage of the sky between 50 MHz and 14 GHz. On completion, it is expected to be up to 50 times as sensitive and 1000 times faster than today’s best-performing radio telescopes. An international community of astronomers will be connected to high performance computing facilities with the power of up to 100 million PCs so they can process data from the SKA around the clock.
10 countries are currently backing the SKA, with 100 research institutions and companies in 20 countries representing 500 people helping with the design.
Evolution of the cosmos
“When SKA is up and running, we will be able to create high-resolution spectra of high-redshift radio sources,” explains Evan, whose areas of research include pulsars transients and solar physics. “The SKA can help us discover a great deal about the early evolution of the cosmos, dark energy, the growth of ionised regions, the formation of stars, galaxies, black holes... The more we find out about the universe, the more we can know about matter.”
“When a star explodes into a supernova, the remaining part can become a pulsar - a kind of reanimated zombie star, which can be the size of a city and spin very rapidly. Each time it spins, we see a beacon of light. These pulses are so regular, that we can use them as highly accurate clocks in space – ideal for physics! The SKA can find all the pulsars in the galaxy and observe them in great detail – there should be 200,000 observable pulsars, but today, we can only see about one-fifth of these.”
Bigger Data
“If all goes according to plan, we’ll be able to observe transients, such as small pulses, that are visible for a millisecond and then absent for weeks. We should also be able to find black hole binaries - pulsars circling black holes - for the first time. We can also research extreme gravitational effects– something that can’t be emulated in a lab! This allows us to test theories about, for example, black holes. We can also put some of Einstein’s thought experiments into practice and test some of the ideas behind the general theory of relativity.”
“With the SKA we can observe the sky in great detail and monitor details from microsecond to microsecond. We can look at familiar phenomena in a more detailed manner and get more insights. The telescopes are constantly on, producing stunning amounts of data. We have to analyse as much of this as we can because we’re looking for fast-changing, hard-to-detect phenomena.”
“Each radiotelescope basically contains wires that pick up tiny currents from light signals, and measure their voltages. The SKA consists of tens of thousands of these little sensors. We collect samples of the data they produce every nanosecond and all these samples have to be collated, added up, compared and analysed.”
“Teams all around the world will search through data in real time, looking for pulsars, for example. This searching has to be automated, due to the sheer volume and complexity of the data. The radiotelescopes are always observing – we don’t switch them off now and then so we can sift through the data! Instead, this is analysed ‘live’, with the help of new technologies such as FPGAs (Field Programmable Gate Array). Of course, if you’re searching large parameter spaces you’ll get a lot of false positives, so the searches are refined in smart ways, for example by introducing machine learning.”
5x current global internet traffic
The SKA will include very large digital signal processing and high performance computing facilities. Every single radio telescope antenna in the array will be connected to an operations centre through a monitoring and control network.
The SKA will be constructed in two phases, but already in its first phase, data rates will be substantial: the mid frequency dishes in South Africa are expected to produce in excess of 2 TB/s of raw data. In Australia, the low frequency antennas are expected to produce 150 TB/s of raw data, more than 5 times today’s global internet traffic!
Optical fibre can carry large amounts of data over long distances at high speed, thus increasing the sensitivity of the radio telescope, by maximising the volume of data transmitted from the receptors to correlators located at different sites. The widespread SKA installation will require enough fibre to wrap around the world twice! The environment in which the dishes and fibre cables are installed is particularly harsh, characterized by extremely high temperatures and drought, for example. Furthermore, it’s pretty hard to service the cable in the middle of the desert. Durability and reliability were important reasons for choosing fibre, besides its vast capacity and future-proof nature.
“The dishes are going to remain in the desert for many decades to come. The connection between the radiotelescope to the computers will also need to remain in place for decades. However, the computers we use to analyse the data they produce, will be upgraded every few years. That’s why a fibre optic backbone was chosen. When more powerful computers are available, the fibre is already there to supply the data they’ll need to make even finer measurements. Who know what we’ll learn from that. Personally, I’m hoping we will be making discoveries that we haven’t even thought of today! After all, many great discoveries in physics came from studying observed phenomena and measurements and trying to explain them!”
Watch the SKA ‘teaser’ video here!
Visit the SKA YouTube channel
