Seeing Farther than Ever Before - The Square Kilometre Array

05 September 2010

Interview with

Peter Quinn, International Centre for Radio Astronomy Research, Perth, Western Australia.

Peter -   The world - 20 countries and radio astronomers in them - are trying to build the world's biggest radio telescope.  It's called the Square Kilometre Array.  It's a poor choice of name.  It's not square, it's not a kilometre-long.  It's different.

If you look at a radio telescope, say like Parkes [observatory in New South Wales] the famous dish that we seen in the movies, right?  It's an eyeball.  It's 1,000 square metres of metal dish which collects radio waves, focuses them into a point and there, it's turned from analogue to digital, and makes images basically of the sky.  That telescope is really good to look at the local universe, the universe immediately around us.  The galaxies and clusters of galaxies we can see on pictures that most telescopes take.  The biggest radio telescope we have which is about 10,000 square metres, so about 10 times the collecting area of the Parkes dish, pushes this out to kind of the very edge of detectability of galaxies.

Parkes radio telescope viewed from the visitorIf you were to draw a ruler out into the universe from here where we are today - we'll call that zero on the ruler, where on the ruler could you get to with our biggest radio telescope?  The answer is one unit out there.  We'll call the farthest we can see right now one unit.  So, where would we really like to be able to see to into the history of the universe?  And the answer on that scale from zero to one, we would like to be looking at 10.  Effectively 10 times further out into the universe than we can see at the moment.

So what's out there at 10 times further?  Ten times further puts us in a point in the history of the universe when the very first objects formed.  The universe began with a Big Bang, we know that and it was a very hot and sort of a gaseous place, and as the universe expanded, it cooled and starts to form objects, and those objects were the seeds of every other object that's ever been - the seeds of stars, the seeds of galaxies and planets, and what have you.  So, that process of the universe cooling in the very first objects being formed happened at number 10 of this ruler we just talked about.  So we can only get the one - you know, one unit on the ruler.  We need to get on the 10.  So that 10-point is sufficiently far away, sufficiently far back in time.  It's effectively 10 times further than what we can currently see.  

Why do we want to get there? If we can see the beginning of the story, then we know how it all turns out.  It's like having a novel with the first few chapters ripped out.  The Universe is a bit like that for us.  We know the end of the story.  We know we're here, we can see things around us, but as you look out into the universe, you're looking back in time.  So you're looking back to the beginning of the story.  And so, we're trying to fill in those blanks and we don't know how this story began.  We don't know what the first objects were, what they look like, whether they were monster black holes or monster stars, or little tiny galaxies.  You know, how did this process begin and then how did it unravel, how did it actually build up as time went by?

So, it's very simple.  If you want to see 10 times further away, how big a telescope do you need?  Unfortunately, it's 10 times 10.  Ten squared because it goes like the area.  So, 10 squared to 100 times bigger than the biggest thing we have right now.  The biggest thing we have right now as I said is 10,000 square metres, the big telescope in the US - so, 10,000 times a hundred is 1 million square metres  - which equals a square kilometre.  So a square kilometre is exactly how much metal collecting radio waves we would need to amass together to be able to detect this incredibly faint signal from the very beginning of the universe.

Chris -   And practically speaking, what's it going to take to do that because no one's going to build a dish like that?  So how are you going to do this?

Peter -   It's basically dishes.  So the way we do this in radio astronomy is you make individual dishes.  Each of those dishes could be 10 or 12 or 15 metres across and an ensemble of those dishes will then make up a square kilometres worth of collecting area.  So, if you want, say, a square kilometre and you made it out of 15-metre diameter dishes, you'd need about 3,000 of them.  So these 3,000 dishes have to be all built, all equipped with radio receivers and all that information has to be gathered and turned into images of the sky, effectively.  Those dishes also have to be spread out.  You can't just stick them all together inside a little corale.  You've got to basically spread them out.  And in fact, the further you spread them out, the better your picture.  It turns out that the larger the separation of the dishes, the more resolution, the finer the picture, the finer the detail that you can actually make in your picture.  So in fact, we would like to spread the 3,000 dishes across the distance, something of the order of 3,000 kilometres.  So not only is it a large number of dishes, it's also a very large machine for doing science.  It's the world's largest scientific machine almost 3,000 kilometres worth of stuff, all connected together with fibre optics doing fantastic science.

Chris -   So, the major remit is a big country with a lot of open space that you can pack in 3,000 dishes.  So Australia is perfect.

Peter -   It just so happens Australia happens to be 3,000 kilometres across just by pure accident.  But yes, you need a lot of flat, wide open space to be able to build this thing.  You also need something else which is incredibly important.  You need no people.  You need to have silence on the radio spectrum front.  The reason is obvious because the signal you're looking for is coming from the very edge of the universe.  It's very, very faint.  Everybody makes radio noise.  Mobile phones, microwave ovens, electric fences, trucks, buses, cars, anything that has a spark plug or any kind of electrical connections makes radio noise.  But we want to be able to be in isolated regions.  

It just so happens that in this western part of Australia, we have, for example there's one place that's called the Shire of Murchison which is in the middle of Western Australia.  It's a region, basically the same size as the Netherlands, so a country-sized shire.  It's inhabitation is listed formally as no more than 120.  So that means you have population densities that are around about 1 milliperson per square kilometre.  Milliperson is like a big toe or two.  So, that's the kind of population density you need, around 10-3 people per square kilometre.  In Europe, you have about 10+3 people per square kilometre, so a million times less dense than Europe.

Chris -   Okay, so we'll go to the idea with geography.  But what about the data?  How much data is each dish going to generate and what sort of computing power do you need to gather that data and then assemble it into a meaningful picture?

Peter -   That's actually one of the most ambitious and challenging aspects of this telescope.  Just to give you some for instances.  Every year, everybody on the Earth generates data and how much data do they generate?  Well every year, the Earth kind of generates around about what's called an Exabyte.  So an Exabyte of data is one and 18 zeros, a billion, billion bytes of information.  The SKA generates that amount of data in a day.  So we're talking about something which outperforms the entire planet in terms of data generation.  The other two sound bytes I like to give about SKA data is that in 2020, when this telescope could be operational, the data traffic inside the telescope, just inside the telescope systems, will be equal to the entire internet traffic in 2020.  So take today's internet traffic expanded up to 2020, the telescope will generate the same amount of data.  And finally, the computer system itself, the computer system we need to run the SKA and transfer all that data hopefully into knowledge will be the world's largest computer system - not today, but in 2020.  So it's an incredible driver for the whole communications and information industry, and they're actually very interested in this project, exactly for that reason.

Artist's impression of the 5km diameter central core of SKA antennas.Chris -   And how much is it going to cost?

Peter -   Cost is an important question too and as scientists we have to be very careful of that and I think in this decade in particular, we have to be very careful in the way we spend public money and actually cost the projects.  We believe at the moment, it'll be approximately 1.5 billion Euros to build it and something in the 100 to 200 million Euros a year to actually operate.  So that's the kind of costings we're working with right now.

Chris -   So that's LHC, Large Hadron Collider level of funding, isn't it?  So it's a lot, but it's not a massive amount when you look at the kind of impact it could make on scientific knowledge.

Peter -   We believe that the impact of this project will be, I think Earth shattering is probably a good word, but - bear in mind the following have already happened.  I mean, the LHC as a project from CERN - CERN, during the process of building that project, invented the worldwide web.  Radio astronomy, even just in building the precursors, building the initial technology for the SKA invented the standard which now defines wifi.  So, everybody uses wifi in the world, that standard came from radio astronomy.  So, I think the commercial returns or the potential technological returns will be enormous.  

Scientifically, this is a project which is truly unparalleled, I think.  This is a project which has not happened before.  If you take Galileo's telescope and you ask how big a step was that for mankind.  To go from the naked eye to Galileo's telescope, that was about a factor of order 10 - 10 times better than your naked eye.  If we look at the impact in the step that SKA will make, it's a factor of 10,000.  Now, that's 1,000 times bigger discovery potential step than any subject has made before in the physical sciences.  It's almost inconceivable what that means.  Given what we've done in building the LHC and building other big experiments and how much advantage with advancement we've had, this is incredible step forward, and it's not so much for the things that we think we'll find, but for the things we don't know we'll find.

Add a comment

CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.