The consortium that will build the world’s biggest radio telescope, the Square Kilometer Array (SKA), today announced the final scope of the first phase of the project, which is due to begin construction in 2018 and be completed by 2023. Although it has had to be scaled back to stay within the available funding, the project will still be able to achieve all of its key scientific goals.
“You have to make compromises when you are cutting your cloth to the funding you’ve got,” says SKA Director Philip Diamond. “There is a scaling down, but it is still a highly transformational instrument,” says astrophysicist Philip Best of the University of Edinburgh’s Institute for Astronomy in the United Kingdom.
SKA, funded by 11 countries from around the globe, will be built partly in southern Africa and partly in Australia. The plan is to first build a pilot instrument, which, as well as advancing astronomy itself, will also prove that the principle behind the giant telescope actually works before the construction of phase 2 between 2023 and 2030. The final instrument will have dishes and antennas stretching across most of Africa as well as Australia and will have a total collecting area of a square kilometer.
A baseline design for SKA was drawn up during the planning stage, and teams in South Africa and Australia are in the process of building two prototype arrays to test some of the technology: the 36-dish Australian SKA Pathfinder (ASKAP) and the 64-dish MeerKAT array in South Africa.
In July 2013, the SKA board set a cost cap for phase 1, also called SKA1, of €650 million. Since then there has been a worldwide effort by working groups and advisory boards to see what was possible within that budget. “We pulled it all together last October so that we could understand the costs to deliver the science goals,” Diamond says.
The final plan for SKA1, announced today, has two components. The first is a midfrequency array with about 200 dish antennas in South Africa that will incorporate MeerKAT—down from roughly 250 in the original baseline plan. The second part is a low-frequency array in Australia that will be made up of about 130,000 so-called dipole antennas (similar to a rooftop TV aerial). The original plan called for 250,000 antennas.
The original plan also included a third element, a specialized midfrequency survey telescope in Australia based around ASKAP but with 60 additional dishes. The dishes would be fitted with novel detectors called phased array feeds that can view a wide swath of the sky at once for rapid surveying. That will now have to wait for phase 2 of the project. Best, who served on a science advisory panel to SKA during the reconfiguring process, says that the midfrequency array in South Africa will still be able to do surveys of the radio sky, only more slowly because of its narrower field of view.
Best says it was a “pretty unanimous decision” among the science advisers to follow this path. “Given the cost cap, it was more important to make two groundbreaking instruments,” he says. “It hasn’t lost any capabilities that the original baseline had,” he adds, but some goals may require more observing time and so “a little more give and take” may be required.
SKA1 has two key science goals. The first is to detect the metronomic signals from many pulsars—rapidly spinning neutron stars that send out very regular radio pulses—so that tiny variations in the timing of their pulses can reveal the passage of gravitational waves. The second is to map out a very faint signal from neutral hydrogen gas through the history of the universe right back to the time when the first stars and galaxies were forming. “A massive range of other science is possible,” Diamond says.
SKA is just starting to set the research goals for the second phase, Diamond says, and those will inform decisions about the size of the final instrument, which will likely have about 2000 dishes in Africa and up to a million dipoles in Australia. Most of the receiving hardware is, however, not a radical departure from what exists today. It is the data handling, computing, and software that remains unproven. “This is the biggest challenge we face,” Diamond says.
Scaling back SKA1 is a bit of a disappointment, Best acknowledges. “But it doesn’t change things. The intention for the full SKA is the same, there is just less in the first phase,” he says.