Outback to outer space: The world's largest telescope

The Square Kilometre Array will probe the cosmic dark ages – and the Australians say they have just the spot for it

FROM the window of the 10-seater Cessna, all I can see is a vast, semi-arid land dotted with shrub-like mulga trees. The little plane kicks up clouds of dust as it lands. We have arrived at Boolardy, a cattle station in the Western Australian outback some 750 kilometres from Perth, the nearest city and one which is itself isolated from others in Australia.

A blast of engine exhaust hits me as I step out into the hot midday sun. I am standing on land that once was part of a giant supercontinent called Gondwana. About 150 million years ago, Gondwana began breaking up into Australia, Africa and the other southern continents. Today, two of these regions, now separated by the Indian Ocean, are vying to host the Square Kilometre Array, the largest telescope ever conceived. Some day very soon, scientists and government representatives from the handful of countries that fund the SKA will decide whether the winner is Australia or South Africa.

It's appropriate that one of these ancient, weathered lands - which provide such a clear window on Earth's geological past - will be chosen to look billions of years back in time. The SKA will take us to within a few hundred million years after the big bang, and probe the universe's dark ages - an epoch invisible to today's optical telescopes - to glimpse the birth of the first stars and galaxies. It will find billions and billions of galaxies, far more than any telescope has ever managed, and so will help cosmologists understand the evolution of galaxies and unravel the mysteries of dark matter and dark energy, the confounding components of our universe.

Nearer Earth, the SKA will scour the Milky Way for every visible pulsar and use them to detect ripples in the fabric of space-time, providing the most stringent test yet of Einstein's general theory of relativity. In short, it will enable science that is impossible with present-day telescopes, and possibly throw up surprises too strange to imagine.

Such a special telescope requires special land, and I was being driven through one of the candidate locations by Barry Turner, the site manager for the Murchison Radio-Astronomy Observatory. If Australia wins the bid to host the SKA, it will be built here in the Shire of Murchison: 50,000 square kilometres of pastoral land that is home to Australia's indigenous Wajarri Yamatji people.

A gravel road cuts through the dry land. As we drive, Turner muses about the hazards of working in the Australian outback: venomous snakes abound, while the unsealed roads become dangerous after thunderstorms. The rivers, which are normally just a series of puddles - or mere dips in the road like the Roderick river that we drive straight across - are transformed overnight.

As our four-by-four rattles over cattle grids, a large, gleaming white dish comes into view, in stark contrast to the reddish earth. The 12-metre-wide antenna is named Diggiedumble, which in the Wajarri Yamatji language means tabletop, a name for the flat-topped hills that stand tall in the arid land.

By the end of 2012, 36 dishes like Diggiedumble will have been built to test technologies for the SKA. These antennas will form the Australian Square Kilometre Array Pathfinder (ASKAP), a radio telescope that will itself make important astronomical observations. A similar 64-dish radio telescope called MeerKAT is being built in the Karoo region of South Africa.

The SKA will easily dwarf the ASKAP and MeerKAT combined. When it is complete in 2024, it will consist of about 3000 dishes, 2000 of which will be built within a 5-kilometre radius. The rest will spread out in 3000-kilometre-long spiral arms. The distance between the core and the farthest dishes of the SKA will give the telescope its resolution - the ability to distinguish between two nearby sources in the sky. Astronomers can combine signals from them in a way that gives them the resolution of a dish 3000 kilometres across. The sensitivity of the telescope - its ability to collect faint cosmic radiation and therefore see further back in time - comes from the total surface area of SKA's dishes: 1 square kilometre. That is far more than any radio telescope ever built.

The car radio crackles. Astronomers in the vehicle ahead warn of animals on the road and Turner slows down. Normally the road, which cuts north-south through the Shire of Murchison, barely sees any traffic except during the winter, when the cattle need to be mustered. It is this lack of human activity, and consequently low levels of radio interference from human-made sources like television and cellphones, that is Australia's strength in the competition for the SKA.

The shire has a population of about 110, or 0.0022 people per square kilometre. "That's about one big toe per square kilometre," says Peter Quinnof the International Centre for Radio Astronomy Research (ICRAR) at the University of Western Australia in Perth. "We truly believe this is the best place in the world for this telescope." Of course, the South Africans feel exactly the same about their site, and there are reports that it has the competitive edge (see "Duelling deserts").

Astronomers are excited by what the SKA promises to do. A telescope can typically study only the tiny portion of the sky that it is pointed at. With the SKA, for the first time, researchers will have an instrument that can study the entire sky every day. This will enable them to study transient sources of radio waves that you would miss if you blinked, such as colliding neutron stars or coalescing black holes. "We'd never discover these sorts of objects in traditional astronomy, because we just don't look at the sky enough," says astronomer Steve Tingay of Curtin University in Bentley, Western Australia.

Also for the first time, the telescope's data will allow the construction of a vast map of the cosmos's neutral hydrogen gas. "The universe is mostly made of hydrogen. If you can follow how it is moving and how it has changed over time, you understand a lot about the history of the universe," says Quinn. Neutral hydrogen atoms emit radio waves with a wavelength of 21 centimetres and it is this heartbeat of hydrogen that radio telescopes are after.

Existing radio telescopes, and ones to come like ASKAP and MeerKAT, map the hydrogen content of the universe in tiny bits and pieces. But they either cannot look far enough back in time, or they do so-called pencil-beam surveys that go deep, but can only study a small sliver of space (see illustration). "The SKA blows it all open," says Simon Driver of ICRAR. "It allows us to sample the hydrogen over the entire universe."

This will help us to better understand the properties of dark matter and dark energy, which together are thought to account for 96 per cent of stuff in the universe. Dark matter is the unseen mass that we deduce exists because we can detect its gravitational influence on normal matter. Dark energy is the energy of space-time thought be causing the expansion of the universe to accelerate. Both would influence the gravitational collapse of clouds of hydrogen gas and the formation of galaxies, galactic clusters and superclusters in the universe. By using hydrogen gas to trace the formation of such large-scale structures from the very early universe until now, astronomers hope to get insights into the two types of darkness.

To build up its picture of the sky, the signals from each ASKAP dish will be transmitted through fibre-optic cables to a "correlator" - essentially a supercomputer that combines all the signals. As we walk towards the control room that will house and protect the computers against the baking heat of the outback, I see a huge monitor lizard, almost 2 metres long, lurking beneath one of the makeshift cabins. It is sipping water from a puddle and keeping cool in the shade.

Back to the dark ages

Keeping cool is going to be challenge for computers too, for the ASKAP and particularly for the SKA, which will require far more computing power to combine the signals from its vast array of antennas.

To lower the energy required for cooling, the ASKAP team is drilling 96 boreholes down to 125 metres. The plan is to dump excess heat deep into the ground, where the temperature is a steady 27 °C, rather than into the air, where summer temperatures soar above 40 °C.

It is not just data from the 3000 dishes of the SKA that will heat its supercomputers. An additional deluge will come from thousands of smaller antennas that look more like television aerials than dishes. These will listen for radio waves at much lower frequencies - between 70 and 700 megahertz. Thanks to the expanding universe, some of the most interesting science exists at these frequencies.

For example, if the hydrogen that is emitting radio waves is, say, 13 billion light years from us, then the waves get stretched by the time they reach Earth, and as a result their frequency drops into the hundreds of megahertz range, a phenomenon called red shift. Studying these signals should illuminate the dark ages: an epoch in the history of the universe about which we understand very little.

In fact, these cosmological dark ages are a complete mystery. They represent a time between the formation of primordial neutral hydrogen about 370,000 years after the big bang and the birth of the first stars and galaxies about 500 to 800 million years later.

These first stars were hundreds of times more massive than our sun, and formed from the handful of elements, mainly hydrogen and helium, created during the big bang. The massive stars then synthesised elements as heavy as iron and spewed them into the cosmos as they ended their short lives in spectacular supernova explosions. So the theory goes, at least. No one has seen such stars. The SKA might be able to do just that.

That is because when the first stars ignited, their intense radiation would have stripped electrons from the neutral hydrogen around them. The resulting hydrogen ions do not emit 21-centimetre radio waves. So, pockets of ionised hydrogen would have created "holes" in a background of neutral hydrogen. The SKA will see these holes, helping us to work out when and how the first stars formed. A similar analysis will tell us about the formation of the first galaxies and clusters of galaxies.

We leave the ASKAP dishes and control room behind and drive to see some antennas that are being tested for the job, though we have to be quick. The sun is low on the horizon and our Cessna needs to be airborne before sunset because there are no runway lights here in the outback.

The antennas look like metallic spiders, arranged in groups of 16. There are 128 such groups in all, collectively known as the Murchison Widefield Array(MWA). When complete, the SKA will incorporate an array at least a hundred times larger, says Tingay, who heads the MWA team.

Tingay wants to use the MWA, and eventually the SKA, to figure out not only what happened 13 billion years ago, but what's happening around us in the Milky Way. One big source of low-frequency radio waves is pulsars: rapidly rotating neutron stars left over after stellar explosions that are beaming radio waves in our direction, much like lighthouse beacons. So far, about 2000 pulsars have been discovered. "The SKA will be able to find all of the pulsars [visible to it] in our galaxy, about 15,000 of them," says Tingay.

This creates exciting possibilities. The ability to map the precise locations of pulsars will open a whole new vista: the universe of gravitational waves. Einstein's general relativity predicts that super-violent gravitational events such as the merger of two black holes will create ripples in the fabric of space-time. No one has yet detected such a wave, but the SKA could change all that.

As a gravitational wave travels, like a ripple expanding through space, it squeezes and stretches space-time, altering the distance between any two reference points it passes through. For instance, a gravitational wave will change, momentarily, the distance between us and a pulsar. This change can be detected by measuring the change in the timing of radio pulses arriving at the SKA from that pulsar. "So, if you had a grid of a couple of hundred of these pulsars all over the sky, and were continuously monitoring them, you would actually be able to map the passage of the gravitational waves across the sky," says Tingay. "That would be as close as you get to a proof in science."

With those thoughts in mind, I find myself back in Boolardy. The sun is setting, and the pilot is anxious to get off the ground. A willy-willy, or dust devil, dances near the airstrip. The Wajarri Yamatji people believe that willy-willies are spirits of their elders watching over the land. Early during the construction of the Murchison Radio-Astronomy Observatory, an incoming plane was held up by willy-willies around the landing strip and had to wait until they cleared to land. These days, there are hardly any. It's as if the elders, convinced that the astronomers are not messing with their land or the sacred sky, have blessed the telescopes. An antique land, an ancient people and a new telescope, all looking back to the beginning of time.

Duelling deserts

The outback of Western Australia, one of two sites where the Square Kilometre Array might be built, may be near-barren land with barely a human to behold, but the competing South African site is no urban jungle, either. I visited it in November 2007.

The Khoisan people of southern Africa call it the Karoo - "the land of thirst". It's a land so arid that sheep farmers need hectares of grazing land for a single animal. The Karoo makes up about 40 per cent of South Africa's land but is home to only about 2 per cent of its population.

Both Australia and South Africa have enacted legislation that creates radio-quiet zones for the Square Kilometre Array (SKA), to prevent pollution by humanity's radio chatter. Given the slightly higher population density in the Karoo compared with Australia's Shire of Murchison, South Africa has also developed bespoke technologies to ensure radio quietness. For instance, cellphone network operators have designed special antennas that will beam signals towards towns, but keep them away from the telescope site.

Adrian Tiplady, project scientist for the South African SKA project, drove me into the Karoo, 10 hours from Cape Town. "It's a fantastic landscape for a radio telescope," says Tiplady. "To the south are mountains that really protect the site from radio-frequency interference from urban centres. To the north there is space for the telescope to expand. It's emptiness, complete emptiness."

If South Africa gets to host the SKA, the plan is to build the SKA's 2000-dish core in the Karoo and spread the rest over eight other African nations: Ghana, Namibia, Mozambique, Kenya, Mauritius, Madagascar, Botswana and Zambia.

In 2007, the site was empty. Today, there are seven dishes of the Karoo Array Telescope (KAT), which will eventually become the 64-dish MeerKAT ("meer" in Afrikaans means "more"). The site also hosts a low-frequency radio telescope similar to the Murchison Widefield Array (see main story). "They have been getting very good radio maps of the sky with very, very little interference," says Bernie Fanaroff, leader of the South African SKA project. "That's by far the most stringent test of the quality of our site."

Anil Ananthaswamy is a consultant for New Scientist based in California and author of The Edge of Physics (Gerald Duckworth and Co)

Info supplied by Jakes ZS1TP