Robyn Williams: I've always want to do that: music by Richard Strauss -- often linked to space travel. It was used in 2001 A Space Odyssey, the movie, and to introduce many a BBC coverage of the Apollo moon shots. Hello, I'm Robyn Williams, with In Conversation this time with rocket scientist Ken McCracken. Ken grew up in Tasmania where he learned both his science and to be self reliant. So self reliant that the American's snapped him up to help them as the space race took off. Well now Dr McCracken has written all about it in a scintillating memoir called Blast Off. And to start, Ken, do remind us of that choice thought that John Glenn had before he blasted off into space himself and wondered about the tendering system for those rocket parts.
Ken McCracken: Yes, Robyn, well it is said that when John Glenn was sitting up there for hours waiting for the thing to go bang, and he was asked, 'What was it that you were thinking of while you were lying there all that time?' And he said, 'Well I was sitting on top of two million parts, each of which was built by the lowest bidder.' And that rather summarised the space program. I had a bit of a run-in with a very large electronic company that everyone knows the name of. We were buying a special form of photo multiplier which we needed to avoid the satellite being, or the instrument on the satellite that went to Venus, being destroyed by the radiation as it went through the Earth's radiation belt. And so we specified a very fancy photo multiplier with special metals in it.
It was to be delivered in 90 days; we had to have the satellite finished in a year. Ninety days came and went, no photo multiplier; 120 days, no photo multiplier. I'm going bananas. And ultimately they came. My colleague tested them; he ran in and he said, 'Ken, they're four times better than they thought they were.' And I said, 'Four times nothing. They have given us the wrong photo multipliers.' So I phoned this very large company and danced around for about half an hour getting them to admit that this is what they'd done. So I sent a telex to the president, I didn't muck around, to the very top of this very large company, and explaining that they had really done the dirty on us and that since NASA was very interested in knowing the provenance of various things that they would be very interested to know what the company had done.
Robyn Williams: And by the way, did it matter that they'd sent you the standard photo multipliers; couldn't you just have stuck those in and got away with it?
Ken McCracken: Well the way we knew it then, we thought that the photo multipliers, the standard ones, would be ruined as they passed through the Earth's radiation belt.
Robyn Williams: So it really mattered?
Ken McCracken: It really mattered, they knew this when they took the order. Anyway about an hour later there was a great commotion and the chap, the head of my institution I was working at, screaming about people sending irresponsible telegrams to the president of company X, and that the president wanted to talk to me. And the head of our institution thought something very bad should happen to me. Anyway I went down to talk to this president of company X and his first comment was, 'Well, I'm sorry what we did to you, it was totally inexcusable and would it be sufficient if we got the photo multipliers to you in five days?'
Robyn Williams: This is not having delivered in over 120 days, they said they could do the impossible in five days?
Ken McCracken: Exactly. And that's exactly, Robyn, what I said to him, I said, 'You're joking.' And he said, 'Look, I know it seems strange but will you trust me that we will deliver them in five days and they will be the right ones?' And they were delivered and they were the right ones. And all I can say is that either somebody else with a bigger pocket book was going to get them, or they had been told by somebody not to deliver them. But we never will know.
Robyn Williams: Isn't that extraordinary. Now the point of this story, apart from the fact that you need to have really good equipment before you can run a space ship, for God's sake, you know it's very important going all that distance -- is here was an extremely young Australian scientist in America which is far more deferential and it taught you actually to speak your mind having got the facts right.
Ken McCracken: Absolutely, and I should say, Robyn, this was not the first time this happened because NASA first of all had told us we couldn't use these photo multipliers and other things because they'd never been flown in space before. And I had a real barney with NASA headquarters on that occasion and sorted it out ultimately. It taught me, as you say, a young scientist on a two-year post-doc visa which was already two years overdue, to go for it. And with our Australian background, our Australian education was enormously good in those early days because it taught us to think on our feet and it taught us to go for it.
Robyn Williams: And of course your education meant that you were extremely self-reliant because you'd handled this equipment yourself, on your own, in Hobart, or maybe in another lab in Australia. And so being so self-reliant you were far more confident and able to get on with things than these wonderfully trained American scientists from MIT and so on.
Ken McCracken: That's exactly so, I would go to lunch with these PhD students and they would make me feel a complete idiot; they knew all the equations of this, that and the other thing, they knew such and such a theory, I knew nothing. But then they started to talk about working in a laboratory and I listened and I thought what tree are these guys coming down from because they had no laboratory skills at all. I'd been up in New Guinea, a one man band fighting the fungus, fighting the humidity and had learned, as you say, to figure it out, no one to ask -- you do it.
Robyn Williams: So the Australian education was damned good in those days?
Ken McCracken: It was damned good and I have to say from what I can sample of it now through my children and grandchildren it's still very good. We have to fight to keep it that way, but there is no doubt in my mind that I and my peers got enormous training of both the theoretical stuff -- because we had what we needed -- but the practical stuff. And more than anything the thing that I think matters in all of these is the Australian attitude of 'give it a go'. And in all my time it is because I was prepared to give it a go and other wouldn't that has done well for me.
Robyn Williams: Just in case people think you were being something of a picky person, we must remind all those who have heard about the complexity of space missions that if one component is wrong, if one line of instruction is wrong, as with the Hubble Telescope, the whole enterprise can go cactus. Just remind us about what went wrong with the famous Hubble Telescope, which we've seen performing wonderfully so far, but in the beginning was a disaster?
Ken McCracken: Yes, Robyn, well that was an example where NASA decided to save the odd $100 million or so, they knew that there had been mirrors designed for earth observation, or spying purposes. And they decided to use those mirrors, that is the sub-contractor decided to use those mirrors, or argued that they should, and they said we will do test procedures and it will cost you $100 million to do this. And NASA decided that this was not necessary. They had forgotten some of the issues that were relevant to using Hubble. So this perfectly good reflector for use for looking at the earth was flown without the appropriate corrections, and that was the big, big lemon. Not the first lemon but the one we all learned about.
Robyn Williams: And it didn't work did it, they actually had to correct it? It was as if you had some sort of myopia and you couldn't see and you needed glasses put on.
Ken McCracken: That's precisely so, so they sent the astronauts up on the shuttle to clip the new bifocal lens onto the Hubble and make it work.
Robyn Williams: Amazing. Before we get back to Hobart and how you started this story, one more anecdote, a poignant one from America. When you were in Dallas and you're expecting the president of the United States for lunch, what happened?
Ken McCracken: Yes, sorry I hesitate, this is a very raw memory even 40-odd years of whatever it is after. President Kennedy was in re-election mode, he was visiting Dallas which was a solidly Republican city but he was doing the rounds and so he was told he could only talk on education and space. All the usual suspects were rounded up and my wife and I were space and education, so we were there waiting for him to arrive at the lunch. We were sitting with the Lord Mayor of Dallas, whose company had built some of the semi-conductors that we used in our pioneer space craft. And so we were chatting away and I hear sirens going by and then somebody comes over, leans over to the Lord Mayor and whispers something, he went ashen, he whispered something to another man, we thought he'd had a heart attack. And then the convenor went up and said first of all he had earlier said start eating because the president has been somewhat delayed.
Then he came up and he looked ashen and he said there has been a serious accident, I would suggest that we all leave. And as we were leaving my wife went into the cloakroom and learnt about the shooting. As we were driving home we learnt that the president was dead. Up till then the United States had been a marvellous place.
Robyn Williams: Easy going, friendly and relaxed.
Ken McCracken: Everything, yes, there were some things that were wrong like the segregated bath rooms still in Dallas, but there were many aspects that we from Tasmania having grown up in the austerity of WWII and the post-war years -- it was like Camelot as the saying was. And suddenly in that instant Camelot was dead. It was a shattering experience really to experience something like that. Not absolutely close up but so close up it didn't matter.
Robyn Williams: I want to understand how a young man -- and you were very young when you went to the United States as we said -- from Tasmania, managed to get into such a crucial place in the space race right when it was taking off? How did your schooling that you're so proud of in Tassie equip you to get where you did?
Ken McCracken: Well probably the simple way of answering your question is saying most people would not have done enough to go into it. Space was a very risky business; risky to your sanity, risky to your reputation. But going back to Tasmania, I just happened to start doing something called cosmic ray research which was the only thing that was really going in Tasmania. Unbeknown to me, because back then in 1954 we didn't know that there was going to be a space age, we didn't know there would be satellites and we didn't know that all the people who were going into cosmic ray science would suddenly become space scientists en mass around about 1959. So here I was, a young bloke who liked electronics, who liked bush walking; I was allowed to build a laboratory up on Mount Wellington which would mean going up in the snow to it. Then I took a laboratory up to New Guinea during the International Geophysical Year, my equipment went to Antarctica, I was learning to do science in a pretty rugged environment with no assistance, in fact I had no assistance nor did I have supervisors for much of the time.
It was great, I had fun and then I was invited by one of the very top scientists in cosmic rays, somebody called Bruno Rossi, to work at MIT. And he later told me that he invited me because (a) other people would pay for me of course but (b) anyone who could make equipment work in New Guinea must be a good scientist. And he said this with passion because as a young scientist himself he had taken experiments to Eritrea in 1929 and was the first person to prove that cosmic radiation were particles, protons and not gamma rays. So Bruno was a really top guy and he said come work with me.
Robyn Williams: Even though you didn't have a PhD quite yet.
Ken McCracken: Not quite yet, I think it would have been an interesting experiment to have seen what would have happened if I didn't get it, but I did apply without the PhD and I got it. And I should say of course that I expected to get offered some posting in the mid-west somewhere in a Boonsville type university and I got this letter from MIT and it just about freaked me out because it occurred to me that I was going to make the biggest idiot of myself known to man. Because Tasmania is where Tattersalls coined the phrase 'you've got to be in it to win it' and so I said right, let's go. And it, as we've said already, I didn't know a lot of the theory, it didn't matter one little bit. In fact when I got to MIT Bruno asked what I would like to do -- that was the way it was back then -- and I had a few choices and I said well I don't want to do any of them, I want to do something different.
Because down in the basement there was a very large computer for those days called IBM 704. It's absolutely brain dead by modern terms, but that allowed me to essentially carve my own niche. By 1960 I was very well known just because I did that, with the stimulus of someone like Bruno Rossi who said just you do it.
Robyn Williams: Yes you were in at the beginning of the computer revolution, the beginning of when space was taking off and you had all those contacts in America before it became too kind of weighed down with bureaucracy and all the rest of it, you had free-wheeling stuff. Now I want to know about when you went back to Australia -- and who are the eunuchs?
Ken McCracken: Yes, this is an interesting story, as you say. When I was in Hobart learning to be a scientist I have to say that I tended to do a few things a little irregularly.
Robyn Williams: We're getting that impression.
Ken McCracken: And I was a radio amateur, which allowed me to build transmitters and talk to people all around the world, but I suppose I was doing it in a slightly irregular way so the government department whose job it is to ensure the virginity of the ionosphere and no one does bad things to the ionosphere kept needing to talk to me. And particularly on one occasion, when talking to a cosmic ray physicist on Macquarie Island -- I wasn't it was his wife who was talking to me -- using the equipment of the Australian Corps of Signals, and we managed to interfere with the ABC all around Hobart and interfering with Auntie was not considered very good in those days. Anyway I had a good record with these eunuchs of the ionosphere as I call them, and then much later I ended up being in CSIRO and we were beginning to negotiate to put a satellite receiver to receive LANSAT images as the satellite whizzed back down over Hobart, and I had to go down and talk to the ionospheric people.
Low and behold it was some of the same people I had dealings with 25 years ago, and they looked a bit crestfallen by something, but we pressed on and did the business and did everything we wanted to do. As we were leaving the technical chap who I knew came up said Ken, you know we were so disappointed because we thought we'd start the meeting with your file from the 1950s and 60s but we couldn't find it, and we think maybe it's been asked for by head office in Melbourne. But it was a marvellous, just a little snippet of the way your past can come back to haunt you.
Robyn Williams: Indeed, and they wanted to chastise you, and couldn't quite do so.
Ken McCracken: They thought they were going to have fun.
Robyn Williams: The point of that story really is -- someone I've known, Ken McCracken, for all these years who's been, as we've established, a person who thinks for himself, quite independent and the last one to get on with the bureaucracy by pulling your forelock and genuflecting -- how, when you came back to Australia, did you get on with the big nobs, you know in charge of CSIRO, in charge of science in the government, in the department of science and so on, did you have any brawls?
Ken McCracken: I'm not totally in-your-face Robyn, I know when to be, shall we call it, appropriately interacting. But the point is that CSIRO back then was a different organisation than it is now and the people that were there, and particularly Victor Bergman, could see the value of people who were pushing, who went regularly at things. And that was why I was appointed, to set up an entirely new division. Victor's comment was what we want is somebody who has none of the baggage of the past in geophysics but who has been in the space program and knows the new things, and who knows how to manage difficult programs.
And CSIRO was brilliant in those days. I was given enormous freedom; it was still in the days that the CSIRO was still thinking like it had in the 1930s and 40s when the telephone system was almost non-existent, where the chief as we called the division had almost total power. He simply got the money, did it and told people in Canberra (in Melbourne then) what he'd done and there was no control. The chief had to be sensible, he had to know what he was doing and he had to deliver the goods. Those were good days without all the control and all the other stuff that happens now -- supposedly for economic rectitude but there were other reasons in there as well. But that was not to be the CSIRO I inherited; my CSIRO was very much a gung ho, let's do something and find a new way of doing it organisation.
Robyn Williams: Of course I remember when Paul Wild became head of the CSIRO, a brilliant solar scientist, not terribly far from your own skills, so you had people of the highest level in science running the organisation.
Ken McCracken: The first paper I ever wrote was with Paul Wild, but that's another story.
Robyn Williams: He died quite recently and it's very sad, I had hoped that there would be recognition of his legacy, because not only was he a solar scientist but he helped invent a landing system that made heaps for CSIRO around the world and he was also very fond of fast trains, which clearly we need in Australia - so on those three lines he was a great man.
Ken McCracken: Oh Paul was the type of man the CSIRO executive was then, farseeing people who took risks and who expected scientists to be good scientists.
Robyn Williams: And did they deliver the goods per buck?
Ken McCracken: Did they -- yes, they did, and look our wine that we have today which came from wine research, look at my own CSIRO division, we provided the initial work in all remote sensing, in exploration under the surface. The CSIRO divisions in those days delivered what we were meant to deliver which was competitive advantage but not products. We could have another interview about this.
Robyn Williams: You also delivered the banknotes didn't you?
Ken McCracken: Well the bank notes -- there was a funny episode where I as a space scientist was asked to be on a committee to look at replacing the old paper banknotes with something new which would not be counterfeited so easily. And there were other attributes that they desired to have in the new bank notes. So there was basically a committee of about ten that barnstormed for about a year under the guidance of Nugget Coombs to come up with concepts which would be much more proof against counterfeiting and the net result of that was the modern plastic banknote which now we see made around the world.
Robyn Williams: We can blame you for that, money in our pockets.
Ken McCracken: Well no some of the people at CSIRO will get more of the blame.
Robyn Williams: Sure. Well what about the future, the future of Ken McCracken as a farmer in the Southern Highlands. You said you've got some remaining involvement in Maryland and a couple of other places still in space research. What are you doing and how do you keep your fingers wet, if you like, or dry, up in space for the next few years?
Ken McCracken: My approach has always been never to go back to what I used to do. So around about 2000 I looked at what we were getting from glaciological evidence and realised there was a lot more there than (a) people were giving it credit and (b) that we could learn a great deal about the inter-planetary magnetic fields, about the sun...and this was a virgin area because all the cosmic ray people thought it was rubbish, and all the people doing glaciology thought anything to do with the sun was rubbish. And sitting in the middle was something that very few people -- I'm being slightly over the top, there were at the top -- there were two people in the world who I joined forces with. And that continues. One of them, a Swiss gentleman Juerg Beer, he and I are writing a book at present on the glaciological study of an isotope of Barillium 10.
Barillium 10 is a weird one, it doesn't exist normally here on Earth, it's only produced when cosmic rays smash into atoms in the upper atmosphere. And that is giving us a record of the manner in which the sun, the cosmic rays and the inter planetary magnetic fields have been varying for thousands and thousands of years. That's my Bronze age space science I'm doing. I work at another institution called the Institute of Space Science in Berne, Switzerland on similar type of things. And in my spare time I look after my farm and I get involved in things like climate change, which you never know when something with my background when something will come along that looks interesting and where you feel you can contribute. Maybe it will be a contribution which not everyone will appreciate, but sometimes that's necessary.
Robyn Williams: Ken McCracken, never a mincer of words; his book Blast Off is a gem. Next week I shall be in conversation with someone who did it the other way round, he grew up in the United States and then came to live and work here in Australia. Professor Phil Long from the University of Queensland -- I'm Robyn Williams.
A History of Satellite-based Remote Sensing in Australia:
Humanity must rise above the Earth, to the top of the atmosphere and beyond, for only then will we fully understand the world in which we live.
-- Socrates, 500 B.C.
Figure 1: Landsat Satellite
(Satellite Images of Australia)
Australia is a vast country, with an area of a little over 7.6 million square kilometres. Its population is concentrated into a small fraction of this, with 83% living within 50 km of the coast. Most of these people are clustered on the eastern seaboard, with a small crescent in the south-west. In addition, Australia is a highly urbanised country with almost 88% of its population living in towns or cities with more than 25 000 people. This leaves huge areas, far bigger than many entire countries, with very sparse population and, at least until very recently, communications and transport infrastructure was confined to a few well-defined routes. Finding out what is happening, or even what exists out there, can be very difficult, expensive and extremely time-consuming. For many thousands of years, every time we wanted to know something about the interior of our country the only thing to do was for somebody to go and look, either with their eyes or some form of data recorder.
This difficulty in obtaining information did not stop people from wanting it, or from going to the, often considerable, effort to capture and pass on information about every part of Australia. Stretching back thousands of years, to the creation of Aboriginal paintings and engravings showing the location of game and water, the mapping and description of the landscape and its resources has been a major preoccupation of the people who live here. The arrival of European settlers led to an upsurge in map-making and formal exploration and description of the landscape and its resources.
Images of the earth from early earth-orbiting and lunar spacecraft had a powerful effect on how we saw the world. Pictures from the Gemini and Apollo missions showed us the world as a single entity, on a scale never before seen. They gave the first inkling of how much could be seen if you went up far enough. In the words of space shuttle pilot Donald Williams:
For those who have seen the Earth from space, and for the hundreds, perhaps thousands more who will, the experience most certainly changes your perspective.
The images from these space missions were, on the whole, incidental to the mission purpose.
With the launch of the first of the TIROS series of meteorological satellites in 1960, another source of images from space became available. As with the images from space missions, their main purpose was not to show images of the Earth’s surface. Despite this, pictures from both sources excited interest in using satellites for observing the earth. Eventually this interest resulted in NASA initiating the Earth Resources Survey Program in 1965 to develop methods for remote sensing of the earth from space.
The launch by NASA (the National Aeronautical and Space Administration of the United States of America) of the American Earth Resources Technology Satellite, ERTS-1 (later renamed to Landsat-1) on 23 July 1972 opened up a new world of data about the surface of the Earth. For Australia, with its huge distances and localised population, the opportunity was too good to miss. For the first time, areas of the Australian landscape could be viewed, mapped and analysed without anyone needing to walk, ride, drive or fly over them.
Every 18 days the satellite recorded the data and transmitted it to earth. Only one satellite was required for the whole planet and, under the terms of the NASA survey program, all countries were to have access to the data.
Starting from a position of considerable expertise in the use of aircraft-mounted remote sensing equipment, Australia quickly developed into a sophisticated user of images, and later digital data, from satellite systems. The uniquely varied nature of our terrain and vegetation, ranging from tropical rainforest to arid semi-desert, led to a series of innovations in processing and interpretive techniques that were taken up all over the world. In the words of Jon Huntington, now a Chief Research Scientist in the CSIRO Division of Exploration and Mining: “If it works in Australia, it will work anywhere”.
The history of satellite-based remote sensing in Australia is full of people: space scientists, geologists, ecologists, spectroscopists, chemists, technicians, and engineers. Particularly in the early years they came from an enormous variety of backgrounds, each contributing their own particular set of skills to the development of the field of remote sensing. Some were interested in remote sensing for its own sake, looking for new and interesting applications of the techniques they developed. Others saw it as an important tool for furthering their research within a particular discipline. Some people worked in the field for only a short time, trying out the techniques and then moving on. Those who stayed working in remote sensing were often enthusiasts who constantly sought new and better uses for remotely sensed data, infecting and inspiring others along the way. The core group of people from the early years is quite small, but their enthusiasm is enough for a crowd of hundreds.
Within a few years of joining the world of satellite-based remote sensing, Australia was a major user of satellite data and a leading innovator in the application of remote sensing to a wide range of resource management, exploration, and planning applications. Many of the processing and interpretation techniques developed here were eventually exported all over the world.
The Beginnings of Remote Sensing in Australia
The idea of using remote sensing as a tool for a range of applications was well established in Australia by the late 1960s. Several different types of aircraft-mounted sensors: multi-band photographic, thermal, colour infrared photographic and Side Looking Airborne Radar were all in use, particularly for mineral exploration and geological mapping. Aerial photographs were also being used for more general mapping and surveying. In addition there was some work being done on determining the ecological characteristics of large areas using aerial photographs taken from either balloons or small aircraft.
As far back as the late 1920s geologists and geophysicists had begun using black and white aerial photographs for geological mapping. By the mid-1960s the field of photogeology, as it came to be known, was well established and continuing to evolve rapidly. Using a combination of aerial photographs and ground traverses, the Bureau of Mineral Resources, Geology and Geophysics (BMR) was mapping and analysing the Kimberley region of Western Australia. In 1966 they began experimenting near Brisbane with black and white infrared aerial photography.
By the late 1960s BMR even had some experience with images from space. In 1965 astronauts on NASA’s Gemini V manned mission took some hand-held photographs of central Australia during one of its earth orbits. Some of these were used to evaluate the Amadeus Basin and one, of the country to the west of Alice Springs, showed the circular Gosse Bluff. Analysis of this image gave rise to the theory that it was an impact crater from either of meteor or a comet. Detailed ground-based investigation by the BMR and the US Geological Survey (USGS) investigation, showed it to be formed by a comet impact.
In the late 1960s the CSIRO Division of Land Use Research was starting to test and use a system developed in-house that used colour aerial photographs for land use capability mapping. Unlike the geologists, the land use researchers needed colour images, making this system costly and time-consuming.
This breadth and depth of expertise and experience set the stage for Australia to take full advantage of data from satellite-based systems.
Figure 2: Shark Bay, Western Australia as seen from Gemini 6.
Images from Satellites
Australia’s involvement in satellite-based remote sensing really began in the early 1970s when the National Aeronautics and Space Administration Agency of the United States of America (NASA) issued a worldwide Statement of Opportunity to potential experimenters and users of data from two planned earth-observing satellites: ERTS-1 and 2. These satellites marked the beginning of non-meteorological, high resolution observations of the earth’s surface, with particular emphasis on the mapping of natural resources, terrain and environmental assessment. The satellite family was originally named to reflect its purpose: Earth Resources Technology Satellite. These satellites were renamed “Landsat” in 1975, a name that continues until the present day. To save confusion, the family will be referred to as “Landsat” in this history. NASA was looking for principal investigators from around the world who would be provided with free imagery in return for reports of their significant results.
Very few people in Australia at that time had any direct experiences with satellites or satellite data. One who did was Ken McCracken, chief of the newly-formed CSIRO Division of Mineral Physics, who had worked in the United States at the Massachusetts Institute of Technology with Bruno Rossi designing instruments for spacecraft. From January to September 1970 he convened an interdepartmental committee to co-ordinate Australia’s response to the Statement of Opportunity. The committee included representatives from CSIRO, BMR, the Division of National Mapping, forestry and astronomy.
There were some people, particularly those using sophisticated aircraft-based remote sensing techniques, who had serious doubts about the potential value of the satellite images. Some went so far as to use the phrase “remote non-sensing”. The main perceived drawback was the huge difference in resolution: the aerial photographs being used were usually stereo pairs with a resolution of one metre; in contrast, the satellite images were monoscopic and had a resolution of 80m. In the end there were enough people who could see potential in the new technology to come up with a total of 53 research proposals relating to the use of Landsat images. These were then forwarded to NASA with Norm Fisher, Director of the BMR, as principal investigator.
Initially Australia was cautious about accepting some of NASA’s conditions for the supply of satellite images. The Department of Supply in particular was very concerned about security. The American “open access” policy made it possible for the images to be widely distributed around the world. Eventually the conflict was resolved and in mid-1972 the Australian Committee for (the) Earth Resources Technology Satellite or ACERTS was set up as the umbrella group for the Australian principal investigator and co-investigators. Its membership was essentially the same as that of the original interdepartmental committee and it was chaired by Norm Fisher from the BMR.
Interest in the satellite data was continually growing and in late 1971 the CSIRO Division of Mineral Physics established a remote sensing group as one of its five programs. The division was still very new, having only been set up in the middle of 1970. Ken McCracken’s background in space physics and satellite instrumentation helped the division to become one of the earliest organisations in Australia to become involved in remote sensing.
At around the same time a small company, Technical and Field Surveys Pty Ltd, began positioning themselves to take advantage of the satellite data. The company had been incorporated as a spatial information and remote sensing company by Mike Aubrey in 1970. So far it had specialised in the integration of rock classification and geological mapping for the whole of Australia. As with many organisations in this field, using satellite data was seen as a natural and powerful extension of what they were already doing. An ex-NASA astrophysicist who was then working at the University of Newcastle, Gail Moreton, was recruited in early 1972 to help develop this capability. Later that year Technical and Field Surveys became an observing member of ACERTS.
After years of planning and waiting, Landsat-1 was launched on 23 July 1972. The launch went smoothly and data transmission started soon afterwards. The satellite carried two main instruments: a trio of Return Beam Vidicons (RBV) and a Multispectral Scanner (MSS). The former were basically television cameras that took a series of snapshots of the ground along the path of the satellite. Each of the three cameras had a different filter on it that allowed a different spectral band (essentially ‘colour’) to be recorded. The resulting images each covered an area 185km x 185km. This means that the whole of Australia could be covered by just 500 Landsat images. The MSS was a mechanical scanner that detected reflected sunlight in several different spectral bands. It detected red, green and two different near-infrared wavelengths. Again the image size was 185km x 185km. The effective ground resolution of both instruments was about 80m x 80m. The satellite had a pattern of 14 orbits per day, giving complete coverage of the earth’s surface every 18 days.
Data from the instruments for North America was transmitted directly to one of the three ground stations in the United States: Fairbanks, Alaska; Goldstone, California; or Greenbelt, Maryland. Data for most of the rest of the world was recorded on one of two tape recorders and transmitted to one of the ground stations when the satellite was within range. The exceptions were countries such as Canada, Brazil and Italy which had built their own ground stations and paid NASA for their data supply.
A measure of the importance of Australia’s involvement in remote sensing was a briefing given to the US Senate in early 1972 by the director of the BMR, Norm Fisher, which emphasised the importance of the Landsat program to mineral exploration.
In July of 1972, just at the time Landsat-1 was being launched, CSIRO Mineral Physics held a briefing on satellite-based remote sensing. This was well-attended with representatives from a wide variety of resource management and land use applications. One of the delegates was a recent PhD graduate working in the CSIRO Rangelands Research Unit, Dean Graetz. Like many others, he had been sent along by his work area to see if this new data source would be interesting or useful. While there was no real data at this stage, some mock-ups based on multi-spectral photography from NASA and CSIRO had been produced to give an idea of what the new images might look like. To quote Dean Graetz: “a summary of our collective reaction was ‘Wow!’” [Graetz, D. (Unknown) MSS: A great beginning with no end in sight]
Ken McCracken remembers:
CSIRO had funded Remote Sensing in my Division by a $5,000 levy on other divisions. Dean started his talk:”Other people have come to see what remote sensing is all about. I’ve come to see what you’ve done with our money!”
Images from Landsat-1
The first real images began arriving in Australia later in 1972. Among others, the photogeology group at the BMR was asked to assess the new technology for usefulness. Colin Simpson was one of those tasked with this:
For the Australian analysts, it was an easy step from aerial photos to the large Landsat images. Because Australians had such as history of using aerial photography, the information about surface geology and vegetation was not of huge interest. In fact it was of far lower resolution than the photographs we were already using. What got the Australians excited was the scale, how much you could see in each image. This allowed for the mapping of major geological structural features such as lineaments, fault lines and tectonic boundaries which allowed increased understanding of subsurface structures [Interview, 11 October 2001.].
For some researchers, their first experience with satellite images has remained impressed on their memory for almost 30 years. In February 1973 Dean Graetz was based in the small NSW town of Deniliquin, working on rangeland management issues in the Broken Hill area:
But it was on that February morning and I was doing research on the diets of sheep and cattle and a letter, a large photographic paper box appeared in my pigeon-hole at work and I opened it up, it was after morning tea, I remember it very well. I walked down into the sun and it was a whole bunch of 9-inch by 9-inch positive film and one of them was the Broken Hill scene. I held it up to the light and I was just, actually overwhelmed. Just the incredible amount of information that I could interpret out of that single-band image. And when I saw the false-colour version I was truly absolutely entranced. I could see a great deal that I could interpret about the nature of the landscape and the effect of land use on it [Interview, 1 November 2001].
Figure 3: Rangelands Around Broken Hill, NSW
(Satellite Images of Australia)
Government organisations were not the only ones to use the new data. Technical and Field Surveys managed to be receiving and interpreting Landsat images within six weeks of the satellite’s launch. The company also negotiated a data supply agreement with the USGS Earth Resources Observation Systems (EROS) data centre that had them selling images within Australia and even back to America.
Characteristics of the Images
The data recorded on the satellite and transmitted to the ground stations was digital and at first the data processing for countries without a ground station was done at the NASA Data Processing Facility of the Goddard Space Flight Centre in the USA. Photographic images were then distributed to the principal investigators. Initially all images for Australia were sent as 70mm photographic negatives to the Division of National Mapping for reproduction and distribution. This led to problems with image quality.
The EROS Data Centre in Sioux Falls, South Dakota was also processing Landsat data and creating photographic images and in 1973 images for Australia began to be sourced from there rather than NASA. These were available as 23cm x 23 cm (9” x 9”) film positives or negatives. At this size, the image scale is 1:1 000 000, in other words 1cm on the image corresponds to 10km on the ground. While the effective resolution on the ground is 80m x 80m, linear features as narrow as 10m can sometimes be seen if they are in sharp contrast to their surroundings.
The images were distributed as black and white photographs for each of the four spectral bands and a false-colour composite based on the standard already in use for infrared aerial photography. For this type of false colour image, colours are assigned to three of the four MSS bands:
· Data recorded as green is displayed as blue;
· Data recorded as red is displayed as green; and
· Data recorded as near infrared is displayed as red
In these standard images, growing vegetation, because it is highly reflective in the near infrared, appears in shades of red; rocks and soils are shades of blue through to yellows and browns; and water will usually range from blue to black depending upon how deep and how clear it is. Man-made structures such as buildings and roads show up in shades of blue and black with characteristic regular patterns. These colours are only a rough guide and can be influenced by many factors such as the angle of the sun, which varies with the season, atmospheric conditions and seasonal changes in vegetation. The main advantages of using a standard mapping of colours to the MSS bands were that it is easy to become familiar with the meanings of the different colours, leading to fast, high-level analysis of the images just by looking at them; and it made full use of the existing interpretive skills developed for aerial photography.
Initially the main users of the false-colour images were those interested in vegetation. The mining and minerals people tended to use black and white images. Only 7% of the images supplied in the early months were false-colour, giving an idea of the dominance of remote sensing by geological applications at that time.
Backgrounds of Early Australian Participants
As satellite images became available, people started working with them for many different reasons and, because there was not yet any formal training in remote sensing available in Australia, they came from a wide range of backgrounds. Some, like Ken McCracken came from a space science background, others like John Perry and Colin Simpson from geology, with others, such as Dean Graetz, coming from ecology, oceanography, mining and rangeland management.
A common pathway into remote sensing in these early months seems to have been for a new member of a work group, often a recent PhD graduate, to be sent to “find out about this remote sensing stuff”. The philosophy was that it was better for someone new and junior to spend time seeing if it was going to be useful rather than wasting the time of someone who was already doing important work.
Two people from the same background who both did receive formal training in remote sensing were Andy Green and Frank Honey. They were both chemists specialising in spectroscopy during their PhDs at the University of Western Australia. In successive years they had CSIRO post-doctoral fellowships at Stanford with Ron Lyon, also an Australian, specialising in remote sensing. When they returned to Australia, Andy Green went to work with Ken McCracken in Mineral Physics at North Ryde in Sydney while Frank Honey started a remote sensing group in the CSIRO Division of Land Resources Management at Floreat Park in Perth.
Early Education and Training in Australia
Some of the first training courses in remote sensing techniques held in Australia were given by the BMR through the Australian Mineral Foundation (AMF) in 1972 and 1973. The AMF was formed by a collection of companies in the early 1970s as a vehicle to promote and foster applied training, information services and an industry bookshop for the minerals and petroleum industries. The courses took the form of industry-focussed short courses and were well attended. On one of the first courses were several people from the mining industry who were to go on to become major players in remote sensing. Among them were Kerry O’Sullivan of CRA and Mike Hussey of De Beers Australia. These short courses continued for the next ten years.
Status of the Landsat Program
When the Earth Resources Survey program was initiated by NASA in 1965, it was essentially an experimental program. Landsat-1 was designed as a research and development tool to demonstrate the feasibility of systematic remote sensing from Earth orbit for resource and environmental monitoring. During 1973 and 1974, with use of both images and digital data from the satellite increasing, there was debate in the US government about the administration of the program. The main question was whether it was still an experimental program or whether it should be reclassified as operational. The result was that the program continued as experimental until 1979 with NASA and the US Department of the Interior required to ensure continuity of data.
In 1975 NASA changed the name of the satellite program from ERTS to Landsat. The second satellite in the family, Landsat-2 was launched on 22 January that year.
First Principal Investigators’ Symposium in Australia
In 1973 the first ERTS symposium was held in Australia. This brought together all the members of ACERTS to report on the research they had been doing based on the satellite data. After less than a year of receiving data the range of applications was already very broad. The presentations covered topics such as rangeland management, assessment of flooding and salt damage, geological mapping, mineral exploration, agricultural yield assessment and management, environmental surveying, forestry and mapping (both topographic and bathymetric).
Different applications found different advantages to using the satellite data. For rangeland management, one of the key advantages was the relative cheapness of the data. Typically Australia’s rangelands are of low productivity and hence low commercial value. This does not reduce the need for them to be carefully managed to prevent their degradation from over-grazing. With pictures needed on a regular basis, the cost of aerial photography was too high for it to be perceived as a sensible tool for managing rangeland.
As explained earlier, for geological mapping and mineral exploration, the advantage of the satellite images lay in their scale. With aerial photography it was very difficult to make a seamless mosaic of a large area. Slight differences in exposure and camera angle meant that there had to be careful manipulation to create a mosaic of a large area. Not only did the Landsat images cover comparatively huge areas, they were far more uniform in angle and coverage, making it easier to combine them into mosaics that could cover entire continents.
As far as using satellite images for topographic mapping went, again the advantage lay in the scale and uniformity of the images. The distortions were small enough for the images to be used as the basis for maps at a scale of 1:250 000. Although most of Australia had been mapped at this scale, the satellite images provided valuable cross-checking and validation of work done from aerial photographs.
Limitations of Images
As Australian use of the images became more sophisticated, dissatisfaction with the image quality grew. One of the major problems arose from the unique characteristics of the Australian landscape. In the available image set, Australia was always very bright, much brighter than most other countries. This meant that many surface details were effectively lost in the glare. The problem with the images arose because the data for outside the USA was all processed in the same way before transferring to film, regardless of the underlying characteristics of the landscape. This meant that it was effectively optimised for areas of average brightness.
In addition, the only images supplied by NASA were black and white negatives for each spectral band and a standard false colour composite for each scene. Which spectral bands, or combination of spectral bands would be most useful to a researcher depended greatly on what features of the landscape were being studied. Changing the combination, intensities and colours assigned to the different bands could greatly enhance different features of the scene. While this was possible using the black and white negatives of each band, this was time consuming and often tedious.
The obvious next step for Australian researchers was to go back to the source data and see if the processing could be optimised for our landscape and vegetation.
The final issue was that the usefulness of the images depended very much on the cloud cover when they were recorded. Up to 20% was often considered reasonable. This was not usually a problem, unless of course the cloud covered the part of the scene of most interest to the person who had ordered it. The NASA cataloguing system gave no information about cloud cover, and so with no Australian-based image processing, it could take a frustratingly long time to select and order a scene only to find that the most important area was hidden by clouds.
All these limitations led to the push for magnetic tapes of the digital data to be sent to Australia for processing.
Digital Satellite Data
One of the first people to start looking seriously at the digital Landsat data was Andy Green in CSIRO Mineral Physics. In 1974 he and some colleagues began experimenting with digital Landsat data and found that by reprocessing it he could uncover much of the detail that had been hidden in the previous very bright images. For an example of this, compare Figure 4 with Figure 5. Both were produced from the same digital source data, obtained on 18 December 1972. As there were no desktop computers or high-quality displays at this time, the team from Mineral Physics had to spend time building image processing and display equipment as well as developing algorithms to process the data. The electronics engineer in this team, Guy Roberts, built the first three image processing systems “from the chip up” [Jon Huntington interview, 12 April 2002].
Figure 4: The area around Marble Bar, as processed by NASA
(Mission to Earth Landsat Views the World)
Andy Green’s work became part of a CSIRO Division of Mineral Physics project, investigating the use of digital data for mineral exploration and the mining industry started to take a major interest in the results. An image that captured their imagination and interest was a scene showing an area of the Pilbara around Marble Bar in Western Australia (see Figure 4). As a region rich in iron ore deposits, the mining company people knew this area well: they had driven over it, flown over it, prospected and mapped in the area. The geology was very familiar to them. In this image, they could simply see the geology and, from their existing knowledge, tell that they were seeing it with remarkable accuracy. Kerry O’Sullivan from CRA was one of the first in the mining industry to see the usefulness of remote sensing as a tool for mineral exploration.
Figure 5: The area around Marble Bar, as processed by CSIRO Mineral Physics
(Satellite Images of Australia)
Another key person, Jon Huntington, joined CSIRO Mineral Physics in April 1974, just in time to become involved in the interpretation of the newly-processed data. He and Andy Green began a partnership that was to last for many years, with Andy focussing on the technical and data-processing aspects of various problems while Jon concentrated more on the interpretive and geological dimensions. Initially Jon joined the division on a one-year contract, which became two years and then four… in the late 1970s he finally became a permanent member of CSIRO staff.
Expanding Use of Satellite Data
In 1973 CRA was responsible for a joint venture which discovered diamonds in the remote Kimberley region of Western Australia. Prospecting in this region was extremely expensive and satellite images seemed to have great potential for narrowing the search for more diamonds. Armed with the knowledge of what the area where CRA had already found diamonds looked like, Kerry O’Sullivan was then able to look for similar areas on the satellite images, narrowing the search considerably. For the initial work CRA used tapes and analysis equipment at CSIRO Mineral Physics. As the work progressed, the staff from CRA became more and more concerned about maintaining commercial confidentiality and limiting the number of people who knew what they were doing. The crunch came one day when Ken McCracken, looking for Andy Green, walked in on a private viewing of the results of some data analysis. Ken recalls:
And for a long time the only equipment in Australia that would allow you to produce pictures on screen or nice photographs from tape was in an old army hut up at North Ryde. I went over to see Andy about something, I’m not sure what, and he was in the dark room and there was a number of people in there and they were as embarrassed as hell that I was there.
…And they didn’t want anyone to know, not even the Chief of the Division
… Anyway my point is that CRA immediately decided “we need some equipment like this so that people like McCracken don’t keep walking in on us” [McCracken Interview 3 October 2001].
Within a few years of digital data becoming available there were several organisations capable of processing it. The two main barriers were the requirement for about $250 000 worth of computer equipment, an enormous sum in 1975, and sufficient expertise to use it. The main organisations involved in image processing were CSIRO, BMR, BHP, the Division of National Mapping and two mining companies, CRA and Western Mining.
As mining companies began using satellite data more and more they became increasingly concerned about security. They often asked for far more scenes than just the one they were interested in with the hope of hiding where and what exactly they were looking at. This need for confidentiality also led to the development of image processing facilities at another six mining companies in addition to CRA. As the main source of computer tapes at the time was CSIRO Mineral Physics, much of the division’s work was funded by the fees the mining companies paid for the data.
With tapes of digital data arriving at Mineral Physics on a regular basis, the division soon built up a considerable archive of Landsat data. By 1975 they were the principal repository for these computer tapes in Australia.
Alongside all this activity in the mining sector, there was increasing activity, although on a much smaller scale, in the environmental sector. The timing of the launch of the first Landsat had been lucky for the land use researchers as it occurred just before the strong El Niño event of 1972-3. This meant not only that there was a major drought, but that large areas of Australia were almost cloud-free for long periods. In contrast, 1974-5 was one of the wettest periods on record. Having these two events so close together, and having satellite images showing vegetation changes and floodwater coverage was of immense value. These images are still used, nearly 30 years later, as a basis for comparing more recent data.
Landsat was not the only satellite providing data for earth resource uses. In Western Australia, Frank Honey was using data from an old National Oceanic and Atmospheric Administration (NOAA) satellite for some oceanographic work. As it happened, Frank had written a program to process this data for a fellow student during his time in the United States. At that time it could take up to 12 weeks to get data from NOAA in the United States, if indeed the required time and place was available. Eventually this became deeply frustrating and so Frank looked into building a simple receiver to pick up the satellite data directly.
In the early 1980s, he liaised with Bill Carroll at Curtin University (then the Western Australian Institute of Technology) about building a receiving station for the NOAA data. Within six months of coming up with the specifications for a receiver they had built one and installed it on the roof of a building in Perth. The main limitation of the equipment was that the satellite had to be tracked by hand using an oscilloscope to track the signal strength, automatic tracking being an expensive luxury for this essentially home-grown equipment. The engineering students at Curtin University became very involved in the design and building of the receiver, showing great enthusiasm for the project. When they called for student volunteers to help with the hand-tracking they had so many that they had to draw up a roster, even for the 3am shift. Using this equipment, they got some lovely pictures of volcanic ash clouds from the eruption of the Galunggung volcano in the west of the Indonesian island of Java that caught the public interest at the time.
The Push for Australian Data Reception
As the worldwide use of images from Landsat-1 and 2 increased at a great pace, demand on the American image processing and data distribution facilities also increased. As Australia was only one of the many countries wanting to use Landsat images and data, this led to frustrating delays in the supply of computer tapes and images covering Australia.
Another problem was that between 1972 and 1974 there were intermittent problems with the data recording equipment on the satellite. With limited storage capacity on the tape, delays in transmission could mean that scenes were overwritten before they could be downloaded. This could result in gaps of many weeks rather than 18 days between successive images of the same scene, which was a particular problem for the environmental scientists wanting to look at changes in the landscape.
In addition to these immediate issues, NASA was starting to plan future satellites which would not carry on-board data recorders. Data from them would only be available through direct transmission.
In the last months of the Whitlam government, the Minister for Science, Clyde Cameron, asked the Australian Science, Technology and Engineering Council (ASTEC) to consider a conceptual proposal to establish Landsat reception facilities in Australia. There was resistance from a significant number of ASTEC members to making such a large investment in what was perceived to be primarily outmoded American military technology. ASTEC recommended against the proposal.
In 1976 CSIRO Mineral Physics proposed a remote sensing research program to Australian mining companies through the Australian Mining Industry Research Association (AMIRA). The program ended up attracting ten companies as sponsors, which, considering the size of the industry, was a very high level of support. Over the next ten years this support from the mining community via AMIRA was vital for the developments in the application of remote sensing to mineral exploration and more generally in the acquisition and processing of satellite data. The AMIRA project has been repeatedly funded up to the present, underlining the relevance of the research group and the continuing interest of the industry in pioneering remote sensing research and applications.
By the mid-1970s state governments were becoming increasingly involved with the use of satellite data. Not surprisingly Western Australia quickly became one of the major players. The state, with its vast distances and intensely localised population, was in some ways an extreme illustration of the advantages of remote sensing for Australia. South Australia, another state with large remote areas, also became very involved.
In 1977 ASTEC were again asked to consider a proposal for establishing Landsat reception facilities in Australia. This time the proposal was vigorously supported by the mining industry, which guaranteed to buy enough data to make the project work financially, and the committee was far more sympathetic to this type of technology. The proposal was accepted and a recommendation forwarded to Cabinet that the Australian Landsat Station (ALS) be built. A key figure in this was the chair of the ASTEC subcommittee, Professor Lou Davies, an electrical engineer and scientist of high repute. In the words of Ken McCracken, “he was a technologist with far-sighted vision”.
Someone who had been interested in the Landsat program from the outset was Don Gray, Director of Honeysuckle Creek Tracking Station. His interest and enthusiasm, as well as his background dealing with space technology had led him to give some lectures about the Landsat program to organisations like the Institution of Engineers and the Institution of Radio and Electronics Engineers. He had to strike the difficult balance between arousing interest with his enthusiasm and making a common mistake of the space industry by overselling the project as a panacea for all the world’s ills.
In 1977 Don was heading overseas for the NASA Station Directors’ conference when he was asked to extend the trip and visit the Goddard Space Flight Centre near Washington D.C. to talk about access fees for Landsat data and then go to the Landsat station in Canada to talk about what was needed to set up a Landsat station. During the course of this trip he found out that Honeysuckle Creek was to be closed and decommissioned in 1981. When he returned from his trip he “volunteered” to start setting up the Australian Landsat Station. This was partly because he was going to need a new job and partly because it sounded interesting. As the project progressed Don became even more fascinated by the potential of the Landsat data and the challenge of building and running data reception and processing facilities.
Industry and Professional Associations
By the mid-1970s, expertise in remote sensing was becoming a recognisable field in its own right. This led to the Remote Sensing Association of Australia being formed in 1976. The initial headquarters for the association were at Technical and Field Surveys’ offices in Crows Nest. It was a group that drew people from industry, government and academia. After a number of years a similarity of interests led to the association combining with the photogrammetric sector to become the Remote Sensing and Photogrammetry Association of Australasia (RS&PAA).
As industry involvement in remote sensing continued to expand steadily there was a need for a channel for formal industry representation in the decision-making and policy arenas. In 1977 the industry user committee for remote sensing, INDUSAT was established. Once again Technical and Field Surveys played an important role, both in establishing the committee and by acting as the secretariat for it. The main aims of INDUSAT were to:
· promote the industrial use of remote sensing;
· encourage government to recognise industry’s needs from remote sensing; and
· lobby for a local data reception facility.
There were about 40 companies represented on INDUSAT, the majority of which were from the mining sector with most of the major mining companies participating, and some agribusiness organisations. Several government departments were involved with the committee as observers.
Designing the Australian Landsat Station
With Don Gray formally appointed as the Director of the ALS in February 1979, the design work started in earnest. Don and a colleague, Frank Northey, set off overseas on a detailed fact-finding tour. They went and talked to staff at several different Landsat ground stations, including the one in Brazil and both of those in Canada. Another important visit was to the EROS Data Centre. At each site they were interested in what data reception and processing facilities they had, why that configuration had been chosen and any problems to be avoided along the way. Since the French space agency, the Centre National d'Etudes Spatiales (CNES) was in the throes of designing facilities connected with its planned satellite series, System pour l’Observation de la Terre (SPOT), Don and Frank also visited France to discuss this design process with the staff of CNES.
At the end of their overseas tour, they had gathered plenty of information and heard many opinions about what the specifications needed to cover and the best design for a Landsat ground station.
There was still one problem, in Don’s words:
Frank and I would have no trouble building a tracking station, antennas and receivers and all that stuff, we knew all about that, but what did you do with the data? How did you process it? [Interview, 12 December 2001].
This was where Ken McCracken helped out by arranging for Andy Green to come and work on the detailed specifications for the image processing.
At the time the specifications for the ALS were being drawn up, there were two instruments on the Landsat satellites. When Landsat-1 was launched, the RBV was considered the main instrument, with the MSS providing supplementary data. During the satellite’s first year of operation there were electrical malfunctions and a need for reducing power use which led to the three RBVs and one of the tape recorders being shut down. This left the MSS as the only instrument on board. Over time, the usefulness of MSS data came to be more and more appreciated.
The final budget for the ALS would only stretch to reception and processing equipment for one instrument. Given the problems with the RBV, the MSS was selected. In this at least, the delay in approving funds for the ALS actually worked in Australia’s favour. Brazil had chosen to install all the equipment for the RBV, which meant that they were up for some very expensive equipment replacement soon after their ground station opened.
As time passed the choice of the MSS proved to be a good one, as this instrument was included on Landsat-3, 4, and 5 whereas the RBV changed format for Landsat-3 and was not included on the satellites after that. With MSS data from Landsat 5 available up to the middle of 2001, there is almost 30 years of data available for analysis. Direct acquisition of MSS data for Australia ceased in November 1997.
One of the biggest inhibitors to the uptake of Landsat technology, all over the world, not just in Australia, was the limited knowledge on the part of potential users of what was available. The US-developed cataloguing system was considered difficult to use and gave potential users no information about cloud cover on the day in question for the scenes they were interested in. This could lead to great frustration, as cloud in the wrong part of a scene could render it useless.
Figure 6: The ALS Ground Station Near Alice Springs
(Satellite Images of Australia)
Together with Allan Falconer from the University of Queensland Geography Department, Don Gray developed an image-based catalogue that allowed users to see miniatures of each scene that was available. Scenes were catalogued using a combination of path and row identifiers, date of collection and 1:250 000 sheet zones. The catalogue was produced primarily as colour microfiche, with a printed catalogue available on subscription. This system eventually became the world standard.
There was some debate about who should act as distributors for Australian Landsat data around the country. The Lands Departments in the various states turned out to be a very good first choice, as they had offices in all state capitals and the expertise to help users select the scenes that would be useful to them. A colour microfiche viewer was bought and installed in each state office so that the catalogue could be viewed.
Further work on Image Processing
In the mid- to late 1970s another group within CSIRO began working on image analysis and display. The Image Systems Group, headed by John O’Callaghan, was based in the Division of Computing Research at the Black Mountain site in Canberra. Unlike the group in the Division of Mineral Physics, who built their own hardware, the Image Systems Group worked with off-the-shelf hardware and wrote the image processing and display software. The initial mage processing was done on the Division of Computing Research’s Control Data mainframe computer using software called CSIRO-ORSER which was a variant of the original ORSER package developed at Penn State University by Brian Turner, an Australian, now at ANU. ORSER was the Office of Remote Sensing and Earth Resources.
The image display system, called a Comtal, was the size of a desk, with all the electronics in a sliding drawer under the desktop. The display itself was a colour monitor sitting on top, and it had a resolution of 512 x 512 pixels, each pixel could display 8 levels each of red, green and blue, giving the equivalent of the modern “millions of colours”. The display unit cost more than $100 000 at the time. A colour monitor on an ordinary desktop computer in 2002 would have a resolution of up to 1280 x 1024 pixels, display the same number of colours and cost under $1000. Although the system was expensive, the alternative displays were the equally expensive output to film, or cheap but very crude character overprint approximations on a line printer.
Specific software had to be written to link the results of the image processing with the display unit. This came to be known as DISIMP, Device Independent System for IMage Processing. The software was run on a dedicated PDP-11 linked to the Comtal. The separation of image processing from image display meant that users could run their analysis on the mainframe remotely and then come to Canberra for a more interactive session using DISIMP and the Comtal. As with other successful CSIRO-written software, DISIMP was eventually commercialised and marketed by Clough Engineering Limited.
With the opening of the ALS due in the last quarter of 1979, the advisory committee structure was revised and expanded from just ACERTS, the initial inter-departmental committee that was the NASA principal investigator for Landsat data. ACERTS itself was moved from its home in the BMR to the Department of Science and then effectively replaced by a new committee, the Commonwealth User Committee on Remote Sensing (CUCRS).
A new committee was formed to be the principal advisory body for remote sensing in Australia, the Australian Liaison Committee on remote sensing by satellite (ALCORSS). The first interim meeting was in March 1979, with formal meetings starting in 1980. Membership of ALCORSS consisted of representatives from each state, the Commonwealth, private industry and tertiary education. It provided a forum to promote the efficient and effective use of remote sensing in Australia.
The structure of ALCORSS and the methods for appointing members meant that there were direct links back to remote sensing interests represented. Most members were part of other remote sensing committees:
· The Commonwealth departments were represented by a member of CUCRS;
· The CSIRO was represented by the Chief of the Division of Mineral Physics;