Interview with Andy Green (AG), ex-CSIRO Division of Mineral Physics

by Catherine Rayner (CR)

17 April 2002

(This is not a verbatim transcript as many asides, comments and remarks have been left out but essentially the majority is “as spoken” – Paul Wise)

CR: What I'm interested in is your career in remote sensing, how you became involved, where you worked, and things that happened along the way?

AG: I finished my PhD in Spectroscopy in the early 70s in the Chemistry Department, University of WA. Ken McCracken then offered me a Post-Doctoral position at Stamford University with (unrecognisable). The rationale was that spectroscopy was doing things in the laboratory and remote sensing was doing spectroscopy from space looking back at the earth. I was at Stamford for a couple of years before returning to Ken’s Division in the CSIRO until last year (2000).

In that period I worked on a number of things initially thermal infra-red (IR) remote sensing, following up my work at Stamford on multi-spectral thermal IR, then when digital Landsat data of Australia became available I started working with that. At that time we (CSIRO) were one of the first to be working with digital ERTS/Landsat data. We demonstrated early image processing, I suppose, to give substantial improvement in picture quality and then used that to persuade the mining industry that it (Landsat data) was worthwhile and to invest in further research to use Landsat data to aid mining exploration.

There was a considerable period then when we were developing rudimentary image processing techniques and building image display hardware. At that time you could not display a digital image without first building specialised hardware!

I then got involved with getting the first Australian Landsat reception station, Landsat data reception capability in Alice Springs moving. With Don Gray we refined the contract specifications for the station. In the mid to late 70s I was joined by Jon Huntington. Jon worked on the geological aspect of things while I worked on data processing, technical, spectroscopic end of the work.

Towards the end of that period when we had substantially finished working with Landsat data we became interested in hyper-spectral (high resolution spectral) data. We conducted a number of experiments with our own hardware and Terry Cox came across from Cloud Physics to join the group. Terry bought with him airborne expertise and with his upward looking instruments, which we turned around to look down, and thermal IR instruments, used in the very early days to scan bush-fires and some environmental projects, we developed hardware with more spectral resolution. We also imported various gadgets from NASA where they were applicable. A number of missions were flown where these NASA instruments, in various aircraft, collected data over Australia. The interest was in short-wave IR to map hydroxyl minerals e.g. carbonates, sulphates.

About this same time I was also looking at the wider applicability of image processing; noise reduction was necessary for remote sensing data. I also spread my interest into airborne geophysics and we concentrated on projects to display airborne geophysical data, magnetics, radiometrics, and things like that. I spent an increasing amount of time on that as well as “classical” remote sensing.   

CR: Around what year was that change?

AG: I guess around the start of the 80s I started that work. I spent some time with Hunting Geology/Geophysics in London and watched what they did to see how image processing could have an impact. In the mid-late 80s I was involved in Division administration as Deputy Chief; Jon was left alone to run remote sensing. This continued until the early 90s when I became Director of the CRC in Airborne Geophysics. So that was almost no remote sensing at that point so right through the 90s I have been associated with geophysics rather than remote sensing. Although in a sense it still is still remote sensing, I just moved the frequency range I was interested in from 10¹⁴ Hertz to 10Hertz. So still multispectral remote sensing but not “classical” remote sensing for some people. That is pretty much what I do at the moment.

CR: Going back a bit what interested you to take up the post-doctoral position?

AG: It was a pretty standard spectroscopic PhD and I was looking to do something a bit different. Remote sensing certainly sounded different and interesting. Otherwise it would have been a protracted career “post-docing” from place to place as Chem PhD’s were inclined to do at that time. It was more that it was a change as for a fair bit of the time I have looked at changing every 10yrs or so, and that was the first of them.

CR: Setting up the Australian Landsat Station (ALS) at Alice Springs, you were working with Don (Gray) and the equipment people as well?

AG: In the early stages but most of the work was figuring out what we wanted; in other words the specifications for the station and that was based on the experience we (CSIRO) had processing the data that had come from NASA. Once the contract was let there was some checking on performance and things like that, but the main exercise was getting the specifications right. This was not a formal secondment just through CSIRO.

What I haven’t mentioned before is the Signal Processing Experiment (SPE)/Upgrade to receive Landsat TM data and process it to digits. It was done cheaply, the major expense being a data tape recorder. The antenna upgrade was done by CSIRO Radio Physics who put and extra “mirror” (not really but think of it this way) in the “dish” to receive the X-band transmissions. These were recorded on the expensive tape recorder and the micro-computer put into Alice Springs processed the data to magnetic tapes which were distributed by AMIRA.

CR: What are the things you got most satisfaction from?

AG: I think the biggest “buzz” is seeing the stuff being used. Seeing, in our case, the mining companies picking up the technology and using it and that was a rather difficult period persuading them what to do. I think the geophysical display of data was a big change for them and the SPE to receive Landsat TM was a buzz also.

In a technical sense there were some successes with processing techniques that almost magically pulled noise out of data. There is considerable satisfaction in seeing where Terry Cox has gone with Integrated Spectronics due to Terry’s hard work and having the vision for what was necessary stemming from the work we did. Going on to build an airborne scanner; the best in the world!