Canberra
Astronomical Society

Australian Astronomy and the "Federation Telescope"
- thoughts on the past and on Future Possibilities

by Professor Don Matthewson

On Thursday 19 March 1998 Professor Don Matthewson, our patron and a past Director of Mount Stromlo and Siding Spring Observatories, spoke to the society. He provided an overview of Australian astronomy in recent decades, and suggestions for the way forward. The following text is a lightly edited transcript from the tape recording. A title and subheadings have been added.

~~~

[Tape starts]

It's wonderful to be here, I've spoken to you quite a few times. But it's true, it is 20 years since I became patron, and I remember Ed Simmons was presiding at that point. It's been a tremendous pleasure to be your patron. I haven't done terribly much, but it's been great to watch you grow into a really united body of very enthusiastic astronomers. You're really making a great contribution to astronomy. And so I feel very proud. But I think it's probably time you got yourself another patron, particularly someone with a bit more power. I think that would serve you well in the next 20 years and I hope you think about that.

Tonight I would like to talk to you about an idea that I had about how we could launch Australia very vigorously into the 21st century. It's a pretty challenging and rather daring idea, but I'm sure we can do it. There's a great opportunity, and it would really make an enormous difference to Australian astronomy at this point in time. I think we're getting a little bit run down, not so much in the radio sense or in the space sense, but certainly in optical astronomy. And I know you know of the Gemini project, but that's rather small - 5 percent of time - probably about 10 nights a year, and I feel we need something on our soil. Something right at the very cutting edge of astronomy, and using all of our talents.

Early Radio Days

What I want to do now is take you back a little bit in time to give you a feel for the tradition of Australian astronomy. As you know I joined Radio Physics up in Sydney way back in the early 50's, and this was when the whole show was getting going, and there were some tremendously exciting discoveries. We didn't know a thing about astronomy. It was all electronics and old Army disposal gear that we were working with. I wasn't with it right from the beginning, I came about 5 years later when household names like Bolton, Mills, Slee and Stanley were already discovering exciting things.

[Slide of Yagi aerial]. That was my first day of Radio Physics. I remember walking up to start work at the front door - it was in the University of Sydney in those days - and this Yagi met me, they were doing some solar observations. I've put that up to show the very primitive state of radio astronomy in those day. Nobody knew what was happening, why there was this noise from the Sun or certain parts of the sky. It was all just one big mystery, and that's why I was so excited. Joe Pawsey took me up to Dover Heights that first day to the cliff-top interferometer. It was a very cunning device. As the sun rose above the ocean there was a reflected ray off the ocean, received by the antenna, and then the direct ray; this formed an interferometer - like a Lloyds mirror type thing - this was the array that found that the Crab Nebula was a radio source, and also did much of the early work on Cygnus. It was very famous but unfortunately, six months after I visited, it fell into the sea. The cliff cracked, and in it went. So unfortunately we can't have it in front of an observatory somewhere, like they do in the States sometimes with their old telescopes.

The next thing was this great hole in the ground that they had scooped out and cemented and covered with chicken netting, and there they did some of the first high resolution surveys of the Milky Way in 400 megacycles per second (I'll say megacycles per second because in those days that's all we talked about, not megahertz). This was the primary feed up the top and you held a rope and loosened off the slack, and somebody with a theodolite told you when you'd loosened it long enough, and then you tied it up again with a knot, and did another strip-scan of the sky. But it was quite an ingenious device. Then there’s the old Dapto Rhombyx that discovered the stream of particles moving out from the solar atmosphere, and they could work out the speed at which the particles travelled. They predicted that in two days time this would appear as an aurora on Earth, and sure enough it was. The Rhombyx are very broad-band devices, which is why they use them. That's a very famous antenna, which Paul Wild did a tremendous amount of work on, but now the termites have got it and it's just a pile of dust. But these were the real ground breaking days for astronomy in Australia.

[Slide] This was up at Potts Hill reservoir in the inner western suburbs. That's a very famous Army disposals radar. Chris and Jim Hindman got a cable from Ewen and Purcell, and they said we've got this 21 centimetre hydrogen line, and we want you to confirm it. So Chris and Jim Hindman rigged up a filter-type receiver on this telescope and in one week they had confirmed that neutral hydrogen did emit a 21 centimetre line to us. This was a very exciting period of time: suddenly an awful lot more of the universe became visible - neutral hydrogen was no longer invisible. This was the little antenna that went on to map the Milky Way, and find out about spiral arms and things like that in a very preliminary way. Just across the reservoir there was old Frank Kerr working away up there on the pole. This antenna was a 36-footer, built in the workshop, and was the first antenna to get neutral hydrogen emission from the Magellanic Clouds. It was a very exciting moment when they first discovered it - it was the very first extragalactic neutral hydrogen.

All of this was happening round about 1953-54. Nobody knew a thing about astronomy. As soon as you'd observed, you then picked up some astronomy text book and tried to make sense of it. Mainly you were just a radio engineer - not a very good one at that. [Slide] This was the thing I started work on. Christiansen designed it, it was a 32-element interferometer. This is still at Potts Hill reservoir, and those other dishes are over the other side. This gave a series of fan beams in the sky, and was used for solar work. On the other side was built a north-south array, the very first demonstration that earth-rotational aperture synthesis would work. While the Cambridge fellows would like to think that they were the first ones - and certainly they were the first ones to actually make it into a very functional type instrument - the first earth-rotational aperture synthesis occurred at this site here round about 1955. Then of course we didn't have computers at all. The first computer came in 1956 - Siliac at the University of Sydney. Everything had to be calculated by the old Whirrer or facet calculator, and it was a long process.

[Slide] Then I moved to Fleurs with Chris, and there's the Chris Cross which Chris designed, you can probably see the 18-foot dishes. Actually one is over at the Exploratory, lying in the dirt somewhere. There were 32 in this arm and 32 in that one. With this array, when you multiplied the two arms together you got a ?? which was later used to scan the Sun. Much later on it became a fully fledged earth-rotational aperture synthesis instrument. The famous Mills Cross is along here. It was only a lot of wooden sticks, half-wave dipoles, running the full length, and probably the cost of the materials was about 500 pounds. We built all of this ourselves, there was no contracting out. We even put in the fence posts. Before they would give us any help from the workshop, Chris and I just started building it, which is one way of getting a project started. We just went out there with a pile of star-posts and started hammering the fence line, and finally they gave us some help.

Those were the string and sealing wax days. We all wandered around very grubby, lived in tents and when we'd finished work at about 8 in the evening, we'd cook our meals on an open fire outside, steak or something like that, and get up at 7 in the morning and continue. Water was very short, and we all washed in the same basin. We'd fill up a clean basin, took in turns who got the first wash and rotated. We were all so excited about building these things and getting some signals from the sky, that it didn't seem to matter very much.

[Slide] Then we moved into the big science game, and this as you know is the famous Parkes radio telescope. I think it's the most beautiful radio telescope I've ever seen. This is the 18-metre, and with the workmen I helped build that. I purposely took this photograph so that the 18-metre almost looked bigger than the 64-metre. It pleases me every time I've ever used it. It looks a lot better now because you've got trees - this was in the very early days when things were very bare. This was our first sully into a big science instrument, and as such it was a very important step forward for Australian astronomy, I think, and one that's very important for our future.

The seventies and later

And then followed a whole series... I'll just click them through: The Anglo Australian telescope. That was a magnificent feat of engineering I think, Freeman and Fox of course, but a lot of the initial thinking for that was by people like Harry Burnett, and they really carried that project through. And although it was of the Kitt Peak design, there were a lot of innovations which made this telescope quite remarkable. In '72 when it first went up, this was really a big science instrument.

This is the dear old 2.3 metre, which has been working away since 1984, and upon which this place relies for a lot of its observing. This was really designed here in-house, and a lot of it was built here. Most of it was built in Australia. The Newcastle dockyards and so forth built parts of it. We had the mirror, that was bought from America and polished by Norman Cole in his garage in Tucson. So this was American, and the azimuth gear was done in Germany, but the rest was Australian and it was certainly completely Australian designed. A very nice telescope, it works well, points very accurately under computer control, etc. In '84 it was quite remarkable.

There's the Australia Telescope. That is also a great feat, showing Australia's skill not only in being prime contractor - CSIRO - but most of it (apart from the computers) was built here in Australia. Aperture synthesis has been a technique which Australia has really pioneered, and it's one in which we have enormous expertise. You probably know the principle of earth-rotational aperture synthesis: if you had a long radio telescope array, you'd get the same resolution as just observing with that array. Really what you're doing with that array is taking a slice out of the dish, and then - there's the earth rotating, there's that little section of the dish cut out, then it rotates. Its fanned beam - its interferometer beam - covers all position angles of the source in the sky. By doing that, you're actually mapping the Fourier transform of the true radio brightness distribution in the sky. Once you've got all of your components, providing you keep your phase constant, the inverse transform will give you the image. It was really brought to perfection by Cambridge. Then Vesterbilt built their array in Holland, which I worked on for the first year of its life, and that is another very powerful array. Then came the Australia Telescope.

Then there's the very long baseline stuff, we're involved with the Russian RadioAstron, and also Haruka the Japanese one, so Australia has enormous expertise in the synthesis technique. It's a very tricky technique, you need to keep the phase constant, have all sorts of delay lines and a phase closure technique, and also to calibrate quite regularly - sometimes every five minutes - on an unresolved source. There are a lot of tricky techniques which Australia has mastered, it's just part of our skills. That I think is another thing to remember.

Australian expertise

We're very good at building science instruments. There's dear old Starlab, which this place spent about five years of its life working on with NASA and the Canadians. The Canadians were building the telescope, the Americans were building the platform, and we were building the instrument package. A lot of you may not know that altogether we were given $8.7 million by the Government to carry it right through the Phase A and B studies, and the NASA team came out, and in the little room over there they examined this. They said that the Phase B which involved a model and all sorts of things, had been done very well. This is a very big space project. Building the science instruments for a space telescope like that is very difficult. That was working with Australian industry - with Hawker de Havilland, British Aerospace, and Computer Science Australia and so on. We can do very sophisticated Phase B, certainly to the state that NASA appreciates. There's no need to tell you about the wonderful fibre optic work they do at the AAT, the 2D [2 degree field] top end which is taking millions of redshifts to get out the structure of the universe to quite high redshifts.

As well as telescopes, we can build very good instrumentation. So all of these things are what Australian astronomy has been built on. The tremendous expertise, construction, both string and sealing wax sort of stuff in very high tech instruments and also very sophisticated telescopes. And we have that here in the country. It's not something we have to go overseas to get, and that I think is very important.

There's one other instrument which has been recently built by the University of Sydney, which is also important as far as this talk goes, and that's SUSI, the Sydney University Stellar Interferometer. Probably a lot of you have seen it. It's an optical interferometer. They use siderostats which are located in these little boxes, round about 8 inches (20cm) or so. These hold computers and some electronics, and the tops of these slide off when you're observing. This is the evacuated pipe down which the light signals are transmitted to the beam combiner in the central part here. This holds the optical delay lines, I think they call them optical pathlength compensators. It's a north-south array, 640 metres long and it's at Narrabri near the radio telescope. It's not actually an imaging device, because it hasn't got phase coherence, it's more of an amplitude interferometer, but it's good for measuring whether a star is double and things like that. There's a little east-west extension here, and that is in the hope that one day it will become an imaging device. For an imaging device you need three stations to do your phase closure, and to get a good image you really need an east-west arm which would then give you enough position angles to give a very good image of the source.

So what have we got?

1. We've got the 2.3 metre, so we know we can design and build a pretty good telescope. We know we can handle very large projects, we can do the prime contractorship and Australian industry can certainly come up with the goods.
2. We know we've got the synthesis expertise under our belts.
3. We have SUSI so we've got a lot of the skills for optical interferometry.
4. The seeing at Narrabri for about 30 percent of the time is better than an arc second, which is what one would want.
5. We have a funding opportunity, and that's probably the most important. You might say why, they've just funded Gemini at 13 million or whatever it was. But I've been having talks with people down there, and there's the Federation Fund - one billion dollars. They have an office set up there now, and you just submit your project, at least an executive summary.

The Federation Telescope

The project I'm thinking of here is combining all of these together to form what I call loosely the Federation Telescope. It's perfect for the Federation Fund, because it's set up to fund reasonably large projects which enhance or demonstrate the Australian characteristics of innovativeness, ingenuity and all that sort of stuff. Something which will excite the youth of Australia, something which the whole nation would be proud of, and something in which all states could share and which symbolises Federation - that is, all states getting together in the united sense. Now I reckon this sort of telescope would more than do all that, and I think quite happily win the 120 million which I think it would cost. What I am proposing is that we build along this arm here, four 4-metre telescopes of exactly the same design as the 2.3 metre. I've talked with John Hart who was the engineer who designed the 2.3 and there's no problem, you don't have to have anything fancy, you can just do it exactly the same way as the 2.3, and because there are four of them you do it a lot cheaper - and a little bit better now.

So four 4-metre telescopes is no big deal, nowhere near the deal of building four Anglo Australian Telescopes which, bless its heart, is the last of those big equatorially mounted things. These of course would be alt-az, and a lot simpler. So it's not a big problem building four. These wouldn't be spaced all along the north-south arm, there would be two, and two on the east-west arm. Then they'd be multiplied together to form our first optical imaging with probably a resolution of a milli-arcsecond, and that is really getting high resolution. It would do it within a field of about 3 arcseconds. So it's not a big field that you're going to image in. You couldn't do it all the time, you'd have to wait for moments of good seeing. But that's alright, because first of all there are enormous projects you could do with four 4-metre telescopes purely in the stand-alone mode. So the two-thirds of the time when the seeing isn't very good we could use the four 4-metre telescopes standing alone; when the seeing gets a little bit better we could combine the telescopes, not with phase coherence, but just combine them to form an equivalent 8-metre telescope. So you've got three modes in which this thing would work. And then you have your aperture synthesis interferometer with which you have a resolution of a milli-arcsecond. Also you don't always need to use it in the imaging mode, you could use it for astrometry, and then you could get positional accuracy to about 5 micro-arcseconds, and of course the sort of thing one can do with that is fantastic.

So you really have a very powerful thing. The beauty of it is that we already have SUSI more or less going, we certainly know how to build the 4-metres, and so we're not pie-in-the-sky, it's all built on many years of lots of hard work by Australian astronomers. It's quite a feasible project. And the beauty of it is this funding opportunity - bless old Johnny Howard's heart - is there. You've always got to remember, I know everybody's getting frustrated by [the lack of] 10-metres and so forth, but there are tremendous projects to be done with 4-metres. Do you know that no very big discovery has ever been made with a big telescope? Not one of the big breakthroughs in astronomy has been made with big telescopes. I'm not surprised at that, because I've worked on big telescopes here and in Chile and so on, and you're always in a rush; you've got one or two nights, it's such a frantic rush that you've got no time to be a real research sort of person; you're chained to the project that you put forward which got you the time.

Dedicated Telescopes and Major Discoveries

Just take a look at what are the big ones - I've jotted them down here: The redshift distance relation. That was made by a dedicated series of telescopes - dedicated time, the people had a lot of time to get the redshifts from galaxies. You look at Baade's Population 1 and 2 - a tremendous breakthrough, the fact that there were two different populations of stars in galaxies. This was made purely because of the war. Baade was a German, and rather than imprisoning him they let him work at his telescope and he found these different populations. Look at Henrietta Leavitt with the Cepheids, in South Africa. She had a little telescope and the luxury that nobody else wanted to use it, and she discovered Cepheids.

Then there's the quasars. A little telescope found the radio stars. Nobody knew what they were; finally after a lot of observing Maarten Schmidt got the spectrum and found what they were. He discovered the quasars, but as objects. The fact that they were very curious objects was really pinpointed by lots of hard work on little radio telescopes. The pulsars - Jocelyn Bell, looking at the scintillation of radio sources, tried to get some information about the interstellar medium, she worked for years and years, and she was the one who discovered the pulsar. Once again it was a small telescope and she had the luxury of time.

Look at the 3 degree Kelvin background, I think apart from the redshift the greatest discovery of the century, and that was made by two astronomers who couldn't get a job anywhere else apart from Bell Telephone Laboratories. They got the job of observing the very first transmission from France to the States. But in the meantime they tried to do astronomy, and they had this irritating noise. They blamed pigeons, and pulled the thing to pieces and all sorts of things, and finally they discovered that the noise was from the sky and that it was the pale remnant of the big bang. And COBE, a dedicated small space telescope - it found the irregularities in the 3 degree background which gave us the first fingerprint on how the big bang evolved. Look at the MACHO experiment, the 50-inch, it's found MACHOs. So provided you've got enough time, and you're not in a rush, and because you've got four 4-metres you can do an enormous amount of work. That alone is great - combined with your 8-metre anyway in moments of good seeing, and this wonderful prize of interferometry.

You see the electronics is going so fast now, and computer strengths are going so fast that it's all going to be very simple. You won't need to go to the Andes or these fancy places astronomers have to trek off to - La Palma and places like that - because the adaptive optics is getting really sophisticated. Blasting laser beams at the sky, making artificial stars, they can sit in one part of your beam and you can do all your phase coherence measurements and controls. There's so much power now. Back in '56 we were battling without any computers, and in '96 the strength of computers is enormous. So adaptive optics, active optics etc. is going to make all of this even easier than it is now.

The sort of things that are aching to be done, the most beautiful one for 4-metre telescopes, is the Lyman alpha forest. Way back in the early universe when the primordial cloud started condensing, the light shines through these clouds and you get this absorption spectrum. You really can get a hold on the chemical composition of the cloud and all sorts of things. A group of 4-metres would really make a superb job of that, which they can't do now because there just isn't time. There's the supernova patrols that they're doing here, the Abell Cluster supernova patrols, and once you get a lot of supernovae you can use them for distance measurement as you know has been done, and then you get a handle on Hubble Constants and that sort of stuff. A simple one like astro-seismology. This is a really up and coming field. Just looking at a star and getting global oscillations, and from that you get a very good clue to the structure of the star. This is an old field, but it's being rejuvenated and it has tremendous potential. These are the sorts of projects we can start doing once we've got a whole new suite of 4-metres.

In the interferometric mode - a milli-arcsecond. You look into your telescopes when you've got a few arcseconds seeing, imagine if you bump it up to a thousand times that. The mind just boggles at the sorts of things that you can do. You could even image planets. Even though they're close to the star, there's a certain device you can use on these things called central fringe nulling which will just wipe the bright star and you'll see your planet, actually image it. You can look for wobbles. At the distance of Alpha Cen, say a Jupiter sized planet around Alpha Cen would cause a wobble of a few thousandths of an arcsecond. With an interferometer you can detect that. You can actually see if fairly nearby stars have planetary systems around them just by detecting them like that. There's so much excitement.

You might say well, we've got Gemini. But what have we really got? We've got the AAT where a lot of the time is devoted to 2D [2 degree field top-end]; Gemini, sure, we've got a handful of nights, but I think we need something that's on Australian soil that gives us an enormous impetus for the young astronomer and also astronomers generally. I think this is the device, and I'm quite keen on it. It is the 21st Century coming up around the bend at us and we'd really better put on our thinking caps. I won't talk any more because it's nearly an hour, but that's the Federation telescope. Thank you.

Questions

Q: The obvious one that Government's going to ask you I guess is what sort of support facilities are going to be required, and how much is that likely to cost, how much is the ongoing maintenance cost likely to be. You'll need workshops, technicians and all those sorts of things.

A: Well generally a telescope costs 10 percent of its capital cost [per year], that's a rough rule of thumb. For instance Gemini, we pay I think it was 13 million, I think they pay 1 million a year to the Gemini project for maintenance. But you see when you've got a going observatory, and we'd build on the Australia Telescope site, the costs are much less than if you start from virgin territory. The Australia Telescope has enormous workshops. So you use an existing infrastructure. But every project has about 10 percent. I think this would be about 120 million and I think it would cost about 12 million a year.

Q: Will it have remote control facilities? If you were at Adelaide University for example could you get on email to the telescope and say look at Eta Carinae between the hours of such and such?

A: From this bench here, with Bob Hawke sitting in the audience, I remember operating the 2.3 metre. The target was displayed on the screen. We had about 100 politicians in this room - a terrible thought - and old Bob was sitting up there with Hazel, and of course Susan Ryan had to ask for Jupiter or some planet, which is always a bit of a worry because you always get them wrong. But what I'm saying is we actually operated the 2.3 metre from here. Generally astronomers don't like doing that because you like to be able to go up to the telescope. You sit over in Baltimore you know, and you watch the results come down from Hubble, and someone's operating it from next door, the controllers. All over the world it's been done, and it's not a big problem. But mostly astronomers like being at the telescope. They're die hards, they don't trust anybody else to make the observation. I feel the same way. I've had people do observations for me and they've always been loused up. The only decent observations are the ones I've made myself. So I'm very suspicious. But then you use Schmidt plates all the time. I think it's a case of attitude.

Q: 120 million is almost an eighth of the billion. I don't know if that's a lot to ask for one project; how many projects are they looking at funding and what selection process? You've got to convince the scientifically challenged bean counters that this one is worth doing.

A: That's right, but there's a bigger hurdle, and that's the [professional] astronomers. They don't even like this. That's why I talk to you. I've talked to astronomers before, so I thought I'd get a little bit of joy talking to you. I think it's a very good idea but I'm the only one who does. I'm puzzled at why. I think they've been brainwashed by going over to Chile and seeing the magnificent seeing. You do get good seeing, a third of an arc second, and it is spectacular for doing globular cluster work and so on, and it is beautiful. We've all observed over there, we all know Tololo etc. and it is impressive. And I think they think, "if we don't get up to the heights of Hawaii or something, we've had it. We must make partnership with somebody - Subaru, Gemini, Magellan, ESO, VLT...". I can understand where they come from. They feel much safer seeing these things being designed and built, but they forget, I think, that 4.7 percent is 18 nights. And generally only 60 percent of those are clear, and that comes down to 11 nights. And 11 nights amongst a nation of 500 astronomers... I think that's where the reality hasn't struck.

Comment: If they took that sort of attitude back in the 60s we wouldn't have the AAT now.

A: Exactly. You wouldn't be sitting here with the strength you have. I've been involved in other sciences too, but we've really got something special internationally with astronomy. I've seen the accelerators and the neutron work and Heliac and so on, but it's not impressive on an international scale. But astronomy is. I was over in Japan at the General Assembly, and you really feel proud with Australians getting up and talking about the 2 degree top end [on the AAT] and all those other good things that come out. People really listen, and I believe that you've got to have that. The trouble is that they haven't built anything recently. Sometimes you think, in Australia could we do it? But with SUSI, John Davis made a wonderful job of SUSI, he's yakka’d away for twenty years with Hanbury-Brown, doing first of all the intensity interferometer and now they've got a Michelson-Morley.

So we know we can do it. We know we can build a 2.3 and I know we can build a 4 without any problems at all. And at Narrabri they've done site testing for 4 years there, and I know the seeing is OK, and a third of the time you can do interferometer work, and can make the change-overs quickly. I have great confidence. I've been involved in building telescopes all my life, since way back when I started in radio physics 43 years ago, and I have great confidence in the Australian instrument maker. Those fibre optics are magnificent, what they've done with the 2D top end. This is all done in Sydney. So I think we really can do this.

[Tape ends]