Oxford University Press's
Academic Insights for the Thinking World

Astronomy’s next big thing: the Square Kilometre Array

When I started research in radio astronomy in 1947, the only known sources of cosmic radio waves were the Sun and the Milky Way. Observing techniques were simple: receivers were insensitive, there was no expectation that other radio sources could be located or even existed. A few years later, a whole vast radio sky was revealed, populated with supernova remnants, galaxies, and quasars. New techniques followed, with sensitive receivers and the big dishes which we now call radio telescopes.

Radio astronomy is due to take another huge leap forward from late 2016, when the construction of the Square Kilometre Array (SKA) begins. Combining the techniques of radio astronomy, telecommunications, and vast computer power, the SKA will in due course provide a completely new level of information about objects such as distant galaxies.

The great leap forward in radio astronomy came in the early 1960s, from a technique peculiar to radio astronomy called aperture synthesis. Radio receivers have a fundamental advantage over optical detectors: they collect radio waves as voltages rather than detecting them as energy. ‘Aperture synthesis’ is the technique by which signals from a large number of different receivers, some separated by large distances, is combined to create the equivalent of a single gigantic radio telescope with the collecting area and sensitivity of the whole combined area.

This is the basis of the SKA – thousands of individual small radio dishes will be combined, making a single telescope with orders of magnitude greater sensitivity than existing radio telescopes.  Furthermore, the individual components will be spread over a large area, which is important since the precision with which maps of the sky can be made depends on the spacing between the components of the array. The signals from the individual elements will be combined to form a signal ‘beam’ that maximizes information from a region of the sky, and with modern data processing a number of independent ‘beams’ can be formed simultaneously within a large area. The result is that many regions of the sky can be observed at the same time.

“Thousands of individual small radio dishes will be combined, making a single telescope with orders of magnitude greater sensitivity than existing radio telescopes (…) many regions of the sky can be observed at the same time.”

Radio, optical, X-ray, and gamma-ray telescopes all have the same task of mapping the multitude of sources of radiation in the sky. The fundamental differences in technique between these different regimes are due to the huge range of wavelengths in the electromagnetic spectrum. Radio telescopes are dealing with wavelengths a hundred thousand times larger than light wavelengths, which accounts for the difference in scale between the SKA and the largest optical telescopes. Forming pictures of the sky, or individual objects, has to be entirely different; the CCD detector arrays universally used in cameras and large telescopes do not work for radio waves. Combining the signals from the elements of a radio telescope array must be done entirely within electronic circuits, and presented as the output of a digital computer.

Aperture synthesis radio telescopes started with only two elements, which could be moved to successive spacings in a long series of observations. This was very early in the digital era, and the recordings, made at Cambridge on punched paper tape, were analysed by EDSAC, the first digital computer. As capabilities of recording, transmitting, and processing data developed, multi-element arrays grew in size. Arrays with large numbers of elements, spread out in various ways, are now in use for different regions of the radio spectrum: the eMERLIN array in the UK uses six fixed telescopes spread over more than 200 km; the GMRT in India has 30 large dishes spaced up to 25 km; the Very Large Array in New Mexico comprises 28 parabolic dish telescopes, spaced over distances of up to 35 kilometres, and ALMA, the millimetre wave array in the Atacama Desert (Chile) has 66 antennas spaced up to 15 kilometres  apart.  Much larger spacings are possible for long radio wavelengths; the LOFAR array based in the Netherlands spreads over 1000 km in six European countries.

Artist’s impression of a 100m diameter low frequency Sparse Aperture Array. CC BY-SA 3.0 via Wikimedia Commons.

Building on the experience of these powerful array telescopes, it became possible to design the international project  of the SKA. This would assemble elements with a combined collecting area hundreds of time larger than existing arrays, spread over large distances, and connected so that multiple beams could be used simultaneously, operating at a large range of  wavelengths from several metres to around one centimetre. A project of this scale could only be achieved as an international collaboration. Selecting a site was critical; there are very few places where the level of radio interference would be low enough, and with an available large flat area.  Two desert areas were chosen, in Australia and South Africa; the SKA will be divided between them. Construction Phase 1 starts this year in both locations; the full SKA will take ten years to complete.

The SKA will look very different at these two locations. The short wavelength array, in South Africa, will use 2000 (200 in Phase 1) steerable dishes, each 15 metres diameter, while the long wavelength array in Australia will use half a million small fixed antennas (130,000 in Phase 1). Prototypes for both already exist, known respectively as MeerKAT  and the MWA. Although the majority of the array elements will be concentrated in areas several kilometers across, there will be outlying elements spreading widely over the African and Australian continents, and both SKAs will extend to intercontinental baselines.

Ten member countries, and over 200 organisations, are contributing to the design of the SKA. The dish and antenna designs already exist. The immense scale of data-processing involved in interconnecting the arrays and forming multiple beams simultaneously at several frequencies is hard to grasp: a supercomputer will be needed working at tens of teraflops (equivalent to 100 million domestic PCs). The length of connecting optical fibres would stretch several times round the Earth, and the total digital traffic will exceed the present-day traffic on the world-wide web.

We already know that the SKA will penetrate to the early years of the Universe, test Einstein’s General Relativity, and tackle the mysteries of neutron stars and black holes. There will be much more; a huge step in observing power is coming as the SKA is built over the next decade. Astronomy will be transformed.

Featured image credit: Artist’s impression of the 5km diameter central core of SKA antennas. CC BY-SA 3.0 via Wikimedia Commons.

Recent Comments

There are currently no comments.

Leave a Comment

Your email address will not be published. Required fields are marked *