Salinity conference

Mapping deeper salinity in intermediate and regional groundwater flow systems
Mr Richard Lane

Richard Lane obtained a BSc Hons in geology and geophysics from the University of Melbourne in 1983. He worked with CRA Exploration from 1984 to 1997, and was involved in a broad range of mineral and petroleum exploration activities across Australia and Southeast Asia. He joined World Geoscience Corporation (now Fugro Airborne Surveys) in 1997 to work with the Product Development Division. He was Program Leader of the Airborne EM Systems Program of the CRC for Australian Mineral Exploration Technologies during the development of the TEMPEST AEM system from 1997 to 2000. In 2001, he joined Geoscience Australia, where he assists geophysicists working in the Regional Studies and Mineral Systems Research Group.

Good afternoon. I would like to address to you a number of issues that I think need to be covered before we really achieve this goal, which is to make these reports very much required reading. I will be addressing issues that relate to its specificity for airborne EM applied to natural resource management. I will have to move fairly quickly, because I have got 10 points and four recommendations.

Is there too little detail on AEM methods mentioned in the report? As Phil [McFadden] pointed out at the start of the day, this report covers a lot of techniques and each one could be 200 pages. But to my way of thinking, from the number of times questions have been raised using airborne EM as an example of something that causes confusion or misunderstanding, I believe that it deserves a little more attention within this report, and a little more detail.

Brian Spies – Are you saying that yes to that first one? [i.e. Is there too little detail on AEM methods?]

Richard Lane – Yes.

Click on image for a larger version of figure 1

There are, however, too few references out there to draw upon, and I would point to a number of issues that cause this shortage. Many of the applications and the systems are being developed as we speak now, and it takes years to get into the literature. They are often developed by groups that have concerns about fully describing what it is that they doing, because that may affect their income over the years. And there are certainly a number of other concerns whenever you try to relate that to specific examples. It is often held up that that would not be desirable for the community involved, et cetera. These all certainly do contribute to making it difficult.

I had an expectation, from the terms of reference and the objectives, that the Technical Report would contain references to the chain of scientific reasoning behind statements that are made in the User Friendly Guide, and I think that aspect of the report needs to be worked on so that that reasoning is fully referenced.

No matter how good the report is, everybody will still want further information, and when the report is out of date – next week – it would be good to point them to where further information may exist.

Click on image for a larger version of figure 2

My suggestions would be to point them towards several of the professional bodies in geophysics that may be able to help with airborne EM related matters, to the service providers around the countryside, and to the various government funded groups and agencies that deal with these matters and these data. That would be a useful set of pointers.

My first recommendation that goes beyond what the present report is all about is that I think the ASEG [Australian Society of Exploration Geophysicists] needs to explicitly recognise that environmental geophysics is an important part of geophysics in Australia and that they should explicitly either include it in their current activities or exclude it.

Brian Spies – That's outside the scope of this topic.

Richard Lane – It is a recommendation, nonetheless, that would assist in this process.

Also, on the fact that this is clearly a very complicated subject: I think there is a need somewhere out there for a one-day course on airborne EM methods applied to NRM to be put together, that could be delivered to communities who are thinking of using these techniques and also for lots of undergraduate courses around the countryside that consider NRM issues.

A common question that is asked, once a group has decided that airborne EM may be the way to go, is, 'Well, there are 15 different flavours of airborne EM. Which particular flavour would be best suited to my task?' I believe there are a number of documents that have been submitted as part of the process here that touch on this issue – papers which talk about how to characterise the systems, how to then proceed by linking those to the survey objectives so that you can select the best matching system for the objectives. Tim has presented an example of those. So that methodology could perhaps be described in the report.

The million dollar question: how much will the survey cost? Again there is material that has been submitted that indicates that the costs could vary by an order of magnitude, and as it stands at the moment the report gives a single figure for how much it costs. The survey needs to be designed to meet the objectives, and the line spacing is the principal factor that governs the costs. That line spacing could be 200 metres in one survey, it could be two kilometres in another survey. There is some material submitted that covers that.

It also touches upon survey design practices. Alan Willocks mentioned earlier that some guidelines and standards need to be developed for airborne EM survey methods so that groups can have a procedure to follow, to make sure that they are getting the most cost effective survey for their objectives.

One thing that I think is covered reasonably well but could be further expanded in the report as it stands is: what does airborne EM map? Clearly, it maps conductivity, and the linkages that are then required to make this into maps of salt hazard.

Click on image for a larger version of figure 3

Conductivity is a complex function of lots of aspects of materials, and salinity – the pore fluid conductivity, related to pore fluid salinity – is just one of those aspects. Then, even when you have got salinity, before you have got hazards you have to introduce hydrology into the deal. So there is a fair chain between mapping conductivity and getting salt hazards.

Ah yes, another good question that comes up: 'Well, can I get my hands on the data myself?' Greg mentioned this one earlier on. I note that you are referring to some guidelines for public distribution of products. I think that is an excellent move. This will also be aided if we do have survey guidelines and standards, in that one of the criteria I am sure that are in the guidelines for the public release is that the data have met some standards somewhere along the line. So they are interrelated issues.

It is possible to generate more than one set of answers from a given data set, and this issue perhaps needs to be addressed in there. The classic statement about geophysical data being 'undersampled, imperfect and highly ambiguous' does apply, and it does mean that you can generate more than one set of conductivity predictions, given different assumptions.

Click on image for a larger version of figure 4

The predictions are the result of a chain of processes, and each has its own effect and its own assumptions. If you use a different measurement system, you will get a different answer. If you apply different processing to the same measurements, you will get a different answer. If you apply a different conductivity transformation process, you will generate different answers. And this is to be expected.

Click on image for a larger version of figure 5

This even happens for relatively simple systems such as the EM31 device. It generates an answer under a particular set of approximations that is relevant only at fairly low conductivities, and it is represented by the black curve on this diagram that relates true conductivity to the predictions of conductivities along the Y axis. If you use a more exact processing method, you will generate the red line on this graph – two different predictions from the one set of data. Why? Because they use different assumptions. The same holds for airborne EM data: if you use different assumptions, you will get different answers.

Click on image for a larger version of figure 6

We should quantify the accuracy of conductivity measurements from airborne EM systems, and the vertical and horizontal resolution.

This is possible. You can calculate the horizontal scale of resolution for different systems, and you will generate different conductivity predictions because you are averaging different volumes of the Earth with different systems.

Click on image for a larger version of figure 7

You can visually show the accuracy of measurements. Here are some examples. The blue line shows some predictions of conductivity, vertical profiles, and the black lines are borehole logs at those prediction sites. That is one method to get across how accurate it is.

Click on image for a larger version of figure 8

Similarly, you can take EM31 readings, grid them up and compare them with shallow conductivity predictions, visually.

Click on image for a larger version of figure 9

You can use statistical analyses to quantify, using methods of correlation and methods of bias and spread. Here is an example of 0-5 m conductivity predictions from airborne EM against ground EM – a pretty fair correlation.

Click on image for a larger version of figure 10

You can quantify that using borehole conductivity logs. Again this is just an example of using independent information to describe the accuracy of the predictions.

Another pet subject that Alan Willocks raised: there have been lots of airborne EM surveys. What happened with all of those previous surveys? Everybody always want to know that.

Click on image for a larger version of figure 11

At the moment you have to scrounge around to find out where those surveys were and what reports resulted.

 

 

 

Click on images for larger versions

The list of all these previous surveys is disturbingly long, and these surveys perhaps deserve some place in our history here. In Western Australia and South Australia, Victoria, New South Wales, Queensland and Tasmania – lots of surveys have been flown. It is very difficult to find out what is happening, and I would recommend that GA and the state surveys do get together and produce a database of these previous surveys and the metadata associated with them, where the data has gone, all the reports associated with those pointers to the ongoing work in those surveys. That would make this job a lot easier. Thank you.