Salinity conference

Session discussions

Introduction, background and overview
Session 1: Risk and investment analysis
Session 2: Hydrology and modelling systems
Session 3: Mapping salinity in the near-surface and upland landscapes
Session 4: Mapping deeper salinity in intermediate and regional groundwater flow systems
Discussion, synthesis and conclusion

Introduction, background and overview

Phil McFadden – Thank you very much, Peter and Brian. You were talking there about surgeons. I sometimes wonder, after a surgeon has looked at an MRI, if he thinks the surgery is actually a ground truthing technique. I do think also that in fact we actually have a more complicated situation here than most surgeons have, because most of the bodies that come in to a surgeon have similarities, and what we find is that every time we go up we are looking at a different catchment, a different set of faults, a different structure, and sometimes the problem we have is a lot more difficult.

We now have about 45 minutes for open discussion on the generic aspects of the document. Later on, as we are going through the different specific areas, we will have opportunity to discuss those. As Peter and Brian have emphasised, it really is important today that we do get good feedback, because the purpose of today is not only to get a broader education about what is happening but, most importantly, for Peter and Brian to get specific feedback on where the document needs to be improved.

Having heard that overview and having read the document, do you have any comments as to how it stands at the moment?

Alan Willocks (Geological Survey of Victoria) – I congratulate you on the document as it is at the moment. I think there is some really good information, certainly relating to the risks and hazard assessment.

A few points I'd like to make relate to comments you were making there at the end towards the interpretation requiring experts to assist with interpretation. I think it actually needs to go back a step before that and deal with the acquisition as well. The collection of these data sets that we are talking about really needs to be managed and dealt with by the same sorts of experts and expertise that we have got around the organisations dealing with the interpretation. I think it would be a very poor outcome if we saw catchment management groups commissioning large airborne surveys, for instance, and not dealing with the quality control issues, and aspects from that. I think it is really important that you stress that through the document.

Another point would be: how do we deal with the techniques that you have got, but then how do we deal with experimental techniques that are coming along? What do we do with the techniques where somebody says, 'Oh, this is going to be able to solve all sorts of things for you,' but there is really not enough information to help make that decision as to whether this really is a good technique or not? We have seen a number of techniques over the last couple of years being bandied around, purporting to do all sorts of things, when really the science of them does not necessarily stand up that well in some instances. On the other hand, some new techniques may well come along, and they may well be very useful in the future. But there is no real guidance in the guidelines for how to deal with that.

Peter Woodgate – Thanks, Alan. They are useful comments. In relation to the quality control, it has been raised before in previous studies that there is a need, particularly for the large surveys where significant amounts of information are being collected, to get some overall coordination going. We could let market forces prevail, and that makes it more difficult.

Alan Willocks – I suppose we have seen what happens when market forces prevail. Where market forces have been involved with the mineral industry, for instance, the sort of information that is commonly submitted back to government is quite often lacking in the sorts of detail that it needs to have to make these sorts of surveys useful.

Peter Woodgate – I wanted to say that the other end of the spectrum was the opportunity for some form of review body, whether is has a government auspice or a professional organisation auspice, in order to collectively and independently evaluate emerging technologies. I think somewhere along that spectrum we need to get it right, because there are certainly good lessons from the past.

Brian Spies – I forgot to mention with the case histories that we have seen, looking at case histories over the last decade, that very early, if not much was known about an area, people would just do some very large surveys to find out what was there. We see the use of techniques much more now aimed towards specific issues and problems, with a lot more care in the design of the survey, the parameters, the forward modelling, first of all to work out which techniques have got most chance of working, what do you expect, how are those acquisition parameters set, before the survey is even started – which I think follows along the line of what you are saying is needed. But the experts need to be involved in the design of the surveys and choice of techniques, as well as in the interpretation. Thanks for bringing that up.

Greg Street  (GeoAg Pty Ltd) – I have only had a fairly brief review of the report; I have been rather busy in the last few weeks. I guess my issue really goes back to the basics of what you are trying to get across. I find that the natural resources management 'industry', for want of a better way of describing it, has a poor record in terms of science. I think this is something that is not unique to it; it is gradually spreading across all levels of science. From what my sister, who works in biophysics and so on, tells me, we are seeing a certain amount of fraud starting to appear into science. And there are some old furphies in natural resource management that have sprung up again in this report. So I feel that perhaps there has not been enough rigour in checking some of the issues, and I would retreat to a point, perhaps in Cartesian doubt, that perhaps we should be doubting more of the old chestnuts that we see springing up time and time again.

I take issue, therefore, with something you just said there, that faults have a major impact on salinity. I haven't seen any proof yet that faults have a major impact on salinity. I would suggest that changes in other geology have a far greater input than faults – differences in geological banding and so on.

I can go on at length, and I will try to get in a written submission. But I would like to see in this report an expression of what we do with these spatial data sets. Peter, I think you said that these data sets collect information, and I take issue with that. They collect data. We get information from them and lead to a position of knowledge, and I think that is fairly fundamental. And that is a fundamental fault I find in the report.

In your evaluation of various techniques, for instance, you have air photos seeing in the root zone, and I really don't see how air photos see in the root zone. You make an intuitive interpretation that you are seeing into the root zone. Also, if you are going to say that air photos see into the root zone, then why not airborne EMs seeing in that top five metres? I think we are seeing indications in airborne EM in the top five metres. You are probably a few years out of date now. We have got enough evidence to say that we are seeing reflections of airborne EM in the first five metres.

Just a few points, but I would like to see a little bit more rigour in it. Thank you.

Phil McFadden – Thank you, Greg. Can I just check that? On the one hand, you are concerned that there is not enough information regarding geological controls on salinity, then picking up one that you don't think is necessarily a major control but there are other controls out there and you want to see more of that. And you want more discussion of the techniques that actually see into that root zone. Is that correct?

Greg Street – No, I tried to make a lot of points there, because there are a lot of other points I would like to make and I was just going for those few.

Peter Woodgate – Just a brief response: Brian and I agree with you with the data to information product continuum, and we tried to express it the way you have just expressed it. If we haven't, then we need to have another go at it.

In relation to the 'old furphies', where we are accepting old furphies which you feel are incorrect, we would love to have your evidence to show us that so that we can build that in to the report.

In relation to the philosophical position as to whether air photos actually see into the root zone or not, we will describe where a technique is using a surrogate measure that enables us to infer what is happening at a distance from the actual point at which there is an observation. So with an air photo showing where a tree is actually in serious physiological decline, that serious physiological decline is a result of the roots being affected by salt. If we have to put that in plain English – you're saying no?

Greg Street – I'm saying no, because how do you know it's not just being attacked by insects? We see trees in the West Australian landscape attacked by insects all the time, and they look as if they're affected by salinity.

Peter Woodgate – Yes, we accept that point completely. For techniques where you are using an indirect measure of salinity, you need a point of validation for that. That is why we are suggesting you need to have more than just one technique for any mapping system. With that reservation in mind, we accept your point.

Brian Spies – I think it is very helpful, thank you. I guess we would ask you also to put all this in writing and send it in as soon as you can. The table in the Technical Report goes into a lot more detail about what is actually measured and then what is inferred. They are in different columns. I take your point very well, that with a lot of these techniques we need to be very careful about what data they collect and then the inference or the interpretation as the next step after that.

Greg Street – I'm sorry, I'll take a little bit more time. I guess that there are some things here you need to be very careful about saying – particularly, for instance, a throw-away line that has been appearing for a long time in the use of airborne EM, that ground follow-up is essential. Well, it's not. If you do one catchment and then go to the catchment next door, you have got sufficient knowledge, generally, to interpret the data without having to go to the ground, without having to drill holes. The analogy to MRI scans is very relevant in this case. A doctor going from one patient to the next knows what he is seeing, and he doesn't have to drill holes into that person to know that he has got a tumour – the same analogy.

Phil McFadden – I would just like to test that amongst the audience, as to how people feel about it, because I think it is an important point that we should cover. Has anybody got any comments on that?

Ian Lambert  (Geoscience Australia) – I don't agree with what Greg just said, but I would just like to make one other point before I say something else about that.

The point relates to this report, and the gap between the report produced on salinity mapping methods and the catchment management people who will be making decisions on what they want to have done in their regions. It is all very well to be able to read a report and to cherry-pick what might be seen as a valid approach in their region, but in many cases – for example, if we take airborne EM – the survey produces a lot more valuable information if it is designed on the basis of interpretation of all of the other available information for that region. You can then have an enhanced understanding of the region and design an optimal airborne EM survey on the basis of that. So there seems to be a gap between a report which outlines the various approaches that can be used and designing the actual specifications of the surveys that are picked by catchment management authorities. I think there is a role of government in there to assist in that link.

With respect to airborne EM surveys, we are not measuring directly salt. There can be other conductors, there can be a range of conditions that require that we have to ground truth. Sure, you get pretty smart and pretty good at it, but you can never get away from the fact that this is not a Black Box that gives you an exact answer; it has to be ground truthed.

Mike Lee  (Dept. Agriculture, Fisheries and Forestry) – I would hope that discussions today can address the issue of extrapolation of understanding, as well as the issue of whether one needs to –

Phil McFadden – Is this specifically relating to Greg's comment?

Mike Lee – Yes, it is, Phil, because we can't afford to drill everywhere. We are trying to invest currently in salinity mitigation measures right across the country, and what we need to do is to pick up detailed understandings and extrapolate those understandings. In the absence of being able to drill everywhere, and even fly everywhere, we need to see what can come out of these learnings in a particular spot and how we extrapolate them elsewhere. So there is a dimension in this that goes beyond just applying the technique in one particular, and whether you can use it in the next catchment – there is a whole dimension of how you extrapolate the understandings that are gained.

Phil McFadden – Thank you, Mike. I think that is a very important thing for any science that is trying to learn to understand the system, so that you can then transfer your knowledge from area to another. Any other specific comments on Greg's comment there?

David Robson  (NSW Dept. Mineral Resources) – I am from the Geological Survey of New South Wales. I agree with both Ian and Greg. Greg mentioned that for a property next door or very close to an area that you already have surveyed and got ground knowledge on, you may not have to validate that on the ground with other surveys and with other techniques. But I agree with Ian too. If some of these areas are wide-spaced away from each other in a completely different geological environment, it is so important to make sure the calibration is right and sometimes that calibration has to be from ground surveys.

Phil McFadden – I am hearing that in the document there needs to be a slightly broader discussion of the issue of ground truthing, and what its limits are and what its limits are not. Do you [Peter and Brian] have any comments?

Peter Woodgate – We agree.

Michele Barson  (Bureau of Rural Resources) – I want to come back to this question of data and information, and suggest an approach which I think will make it a lot easier for the average reader of your document – that is, when you are going through each of your 19 approaches, to make it very clear what question the data that is coming out of those approaches is actually answering. In fact, I would suggest that the entire report could be framed around: what do we really need to know? Of course, the first question is: where is the salt? The second question might be: how much salt is there? The third question might be: what is the risk of its being mobilised? And then the next question might go to: how much is this going to cost us? I think framing it around questions – the old 'What do you need to know, and how well do you need to know it?' – which goes to the scale issue – might make this mass of information a little bit easier for the average reader to interpret.

I would like to make another comment on the extrapolation issue, if I may. I suspect that at the moment we simply don't have enough studies around the countryside to be really confident about the extent to which we can actually extrapolate information. We are starting to get some feelings from the studies that we have done that there are some patterns there. We don't really know how far those patterns might be understood or used yet, and I think that it might take, at least particularly with the AEM technology, a little bit longer, a little bit more information, a few more studies, to understand just how far we can push some of the extrapolations.

I also come back to Ian Lambert's comment: it is extremely important to bring together all the information that we have about a catchment or a region, to design the survey very carefully. Very careful survey design means that we will have to collect a smaller amount of information, using very extensive and very expensive techniques.

Phil McFadden – Thanks, Michele. Certainly I think there has been an attempt in the User Friendly Guide to approach it from a question point of view. I personally think that is a very important thing, because in fact a user on the land does not know what he needs to know, very often. So he needs to be aided down the path of, 'What questions do I need to ask? When I have those questions answered, what is that information going to tell me, and what are the techniques I can use to get those questions answered?' So I think it is a very valid approach.

Greg Chapman  (NSW Dept. Infrastructure, Planning and Natural Resources) – I have got several comments. Basically, there seems to be a big focus in this document on salt store, rather than a focus on salt flux. We have thousands of hectares of land in New South Wales and, presumably, in other states as well where there are enormous salt stores. Those salt stores may not be moving. We have also got a landscape which has gone through major climatic upheavals in the past, where we have had previous salinity fluxes. I think that needs to be taken into account as an understanding of what is going on with salinity. The level of high tide where saline waters have been in the past can be looked at, and if areas have not been previously saline then they are not likely to become saline again.

However, in terms of salt flux, we are really looking at things like water balance, and with water balance there may be areas which are very pertinent for land use or land management change for fixing salinity problems but which don't have any salt store there at all. So if we are going to concentrate on salt store without looking at what is going on in terms of landscape hydrology and water balance, then we are really not going to be addressing the nub of the problem.

The other comment I have is about information management. Information is basically data that has context. It is really important that the data is stored in context and properly looked after. We need custodianship of whatever information sets there are, which are basically purchased using public funds, for them to be stored appropriately. If that does not happen, then it is very difficult to benchmark, to put to future use, or to judge the effectiveness of any programs which are targeted towards amelioration of salinity.

The other thing is too, just focusing back on the salt flux idea, obviously if there is a landscape which has a salt flux issue, then basically we need to look at what sort of parameters are relevant to that. So things like soil water storage, which is not really mentioned in this, can be a major thing which we can actually manipulate. Vegetation and soils are the aspects of the landscape that we can manipulate on a broad acre basis to derive salinity solutions. Now, if we are going to be collecting information on salt stores, why don't we spend a little bit extra to make sure that those parameters are also looked at? Thanks.

Phil McFadden – Thank you, Greg. I think you have a good point there. Hey, in this country there's plenty of salt to keep everybody happy, thanks very much. It's not actually the salt that matters; it is the mobilisation of it. I thought that you [Peter and Brian] had been attempting to make that point in the document, but obviously it has not come across clearly enough; yes, it is the mobilisation that matters and the mechanisms by which that occurs.

I think you also make a very valid point about the data, where we have a real problem in this country. There is a tremendous amount of money being spent on data, by a whole range of different people. It is being gathered without national standards, without anything like that, so we can't draw back and get the sort of systemic information that Greg was talking about, because we don't have a consistent way of bringing those data together. I think that is very important. Any comments there, Brian?

Brian Spies – A very good comment there. I think, based on the feedback we would like, something like that would really help if you could suggest in either of the reports, 'This is what I think you should say. Here is a paragraph or a sentence you could put in such-and-such a location, or change these words, and emphasise it that way.' If you can make specific changes in the report for what you would like to see in there, we won't always agree to put them in but it helps us work out where would be the best place to put it, does it make sense, does it flow, is that a case for peer review, do we want to send it somewhere else? Very specific comments like that would be very useful.

Alan Willocks – I would like to concur with Greg's comments. I think they are really important: the information management component of it is something that is overlooked very often. I would like to take it a little bit further and make the point that actually maximising the use of the existing information is also a really important part of what we need to be doing here. A lot of the emphasis is looking at, 'Well, can we go out and collect some new surveys and use some extra interpretation to deal with that?' but we have currently been dealing with a project in Victoria and it's really hard to get a good water table surface, and this is supposedly an area that has got really good data coverage. Even just getting that basic set of information has proved to be quite difficult.

So the use and management of the data over what has been collected, what has been proposed to be collected – for instance, making sure that a survey is actually located in real-world coordinates – actually provides a really useful bit of information that can then be used many times over for validation of other surveys for other sorts of projects that might not have been envisaged at the time.

Phil McFadden – Thanks, Alan. Yes, I think this management of the information is really critical. It is a very important issue – and I think it is relatively germane to your document, Peter and Brian, isn't it.

Greg Street – I would just like to pick up on that. I saw one of the submissions you received in reply, from Simon Abbott, and helped review his review of your review the other day. It concerns the use of these spatial data sets on which to hang other information. That does not come through in this report very well. We find that the three-dimensional nature of an airborne EM survey is a very good data set on which to hang all the other data sets that you get. It actually builds a very good picture. So you bring your surface information, you hang your drill-hole information, everything else, on it, and you find you build up a much better picture by having that – the same as you would with just an air photomosaic that is rectified. Too much in the past I have seen unrectified data used, but if you have got a rectified photomosaic, we have used those on which to hang data sets in the past. But airborne EM, and other geophysical data, adds this extra dimension.

Colin Pain  (Geoscience Australia) – I would just like to make a comment following up a bit on Greg Street's point about faults. That list, the dot points on page 8 [of the Technical Report], of the factors impacting on salinity, is a bit bald. It is bald to the extent that in some cases I think it is misleading. On the comment on faults, I would say they are simply part of the geological complexity, so really it is geological complexity you are talking about there. On the comments on regolith distribution and so on, that is merely part of the broader geomorphic context of the whole problem. So I think it is probably worth while looking at trying to expand that part, at least, of your report. I am quite happy to provide some words for some of it.

David Robson – I would like to make a couple of points. Firstly, like my co-worker in Victoria I wanted to congratulate both authors. It is a documentation of all, or most, techniques that could assist in the investigation of salinity – in one document. We have not had this up to now. I agree with a lot of the other speakers here this morning that there can be improvements in various aspects of a lot of those techniques that have been discussed, but the most important thing is that it is documented in one document.

The next one: looking at the 3-D picture, that is very important and certainly, on the last couple of conclusions out of the report, that is highlighted. We must look at the 3-D. Some of the other people this morning have talked about, 'Well, what are we trying to do?' We have to identify where salt is at the surface and I think the local farmer has already identified that. The local farmer then wants to know how widespread the problem is and what impacts there are going to be. The report does address hydrology, but there may have to be a bit more enforcement of that, and salt flow. But I have to agree with the authors that the 3-D is very good. Where I would like to stress more is interpretation.

Now, Alan Willocks mentioned a quality survey. If I could just take an example from New South Wales: we had a very good, high-quality airborne EM survey that had a lot of controls, there was help from GA, and it came in as a top-class survey. But what didn't happen was the interpretation of that data. There was a 40- or 50-page report prepared but there was very little, nearly zilch, interpretation in that so-called interpretation report. It did not address where there may have been salinity problems at depth. It was perhaps not written by the right person that could understand the technique. And the correct interpretation, I have to stress that.

As far as case history goes, as I said earlier, it is good to bring all these things together in one report. Quite often the case histories that you have listed, the nine of them, are all in separate papers and the like, and it is somewhat difficult for everyone to get a handle on what those case histories were about. So certainly I would like to see a paragraph or two done on each of those case histories before the report was finalised next week, if that was at all possible.

Tim Munday  (CSIRO Exploration and Mining) – I would just like to make a couple of really basic points that Colin Pain actually touched on, and Greg did too. It relates to things like factors affecting groundwater movement and also direct and indirect indicators of salinity.

You have not really explicitly touched on a very fundamental issue: the geology. You don't actually explicitly mention the geology, which is a critical factor in controlling how groundwater moves, where it goes, et cetera et cetera, and whether it sits. Some very significant aquifers are essentially not regolith, they are rocks, they are sediments. I think that should be explicit in the report, and it is not explicit. So I would commend you to perhaps make a change there, or at least flag that aspect.

Ian Rae  (Australian Academy of Technological Sciences and Engineering) – I would like to go back to the theoretical risk framework that is presented in the Technical Report, and compliment the authors on simplifying down to risk as a product of two factors, which I think is the right way to approach the matter. It leads to the two-dimensional matrix on page 20, with two axes and estimates then of whether the product represents a high, medium or low risk.

But in the preceding couple of pages I think the authors have rather muddied the waters. I would see the two factors as being hazard and likelihood. The salt is the hazard; the extent of it is the size of the salt load. But that is a necessary but a sufficient condition, as Greg from New South Wales said. It is in the other factor, the likelihood, that the groundwater and the movement comes, I think. That is the clear separation between those two things. So if the hazard term, which is the salt – the salt is the hazard, and I disagree with the report there – is very small, then the overall risk is very small. If that term is substantial, then you have to consider the likelihood term, which is: what is the groundwater doing, how is the soil managing this, what is going on at the surface? They are the factors that determine whether there is going to be a substantial risk.

In the Technical Report that clarity, which seems to be there at the beginning, gets muddied as I went through the next few pages. 'Impact' appears out of nowhere, not being very well defined, in my opinion. And then 'consequences' turn up, with hardly any definition at all. Maybe there were too many cooks spoiling the broth, because Mark Tweeddale's triangle appears in one of those pages, and that is a more sophisticated version of the simple two-factor equation. I think the authors ought to have another look at that, and go back to that simple two-factor equation and change the wording of some of the other things to fit into that pattern, because only with the two-factor equation can you get any clarity. Certainly when you get to the User Friendly Guide, starting to introduce hazard, impact, consequence – hey, wait a minute! Who's being simple now?

Phil Dyson  (National Dryland Salinity Program) – I guess I am hearing a number of people saying the same sort of things that I am thinking here. I have just come off three days of working with the Port Phillip Catchment Management Authority, delivering amongst a number of other people an understanding of how groundwater systems and salinity happen in the various geological units across the Port Phillip region. And after three days we identified – boiled it down to – would you believe, some 23 units, 23 groundwater systems that were contributing to salinity in some way.

My important point here, I think, is that that is where you start. You should start the debate with what you know, not what you don't know, and you should basically try and divide your systems up in the first instance on the basis of existing biophysical understanding. What do I mean by that? You simply go in and identify where you have a fracture drop system, where you can put a boundary around that, you know that salinity is coming from that. You go in and identify a deeply weathered groundwater system and you put a boundary around that, where you know and you have had the studies before there to do that. You look at a riverine plain situation, with a buried deep aquifer system that is contributing to salinity – the sorts of things that Phil Macumber and others have described on the Loddon Plains. You look at the deeply weathered granitic environments that you find in Western Australia.

You start with that level of framework, and then you dig down into that framework to decide exactly it is that you are looking for in using these techniques. I think it is very, very dangerous if you are actually going to go in and do blanket surveys and you are going to work with people that do not necessarily understand the techniques that they are buying to do these blanket surveys.

You need to also recognise – and I think we are all saying it – that we work within some kind of information supply chain in here (some people have called it a food chain!) and that we are not necessarily just using this document and throwing it at a catchment management authority or somebody like that. It might be that this document goes to a whole bunch of specialised people that are providing advice to a catchment management authority, rather than to that catchment management authority in the first instance.

We are in a dangerous situation, I think, with some of the catchment management authorities and those kinds of groups, where they will just go out and buy the technology without advice. I think we need to make sure that that doesn't happen, and that we provide the very best information in terms of what we already know and on what framework they are actually buying this technology.

Peter Baker  (Bureau of Rural Sciences) – There appears to be one thing missing in the mapping techniques, and maybe it was missing from our submission too so I do apologise for that. There doesn't seem to be stream sampling in there. We are talking about things for natural resource managers, catchment managers. That is the simplest thing that they can do, and one of the most powerful data sets that they can actually gain as part of their understanding the system. It is something that can involve individual landholders, becoming involved, gaining ownership, and that is what we want for natural resource management, that the people on the land have ownership of the process. We seem as scientists, as we are prone to do, to go to the technology all the time and tend to neglect the simplest thing that we can do in understanding the salinity problem, and where the salinity is actually occurring. So I think we also need to look at, as a process of this report, incorporating stream salinity sampling in there, and the best ways of going about it.

Peter Woodgate – That was omitted by design; it was beyond the terms of reference. But you have made the point now and I think that point has been heard by the clients. It is something we can review immediately after this forum, unless the clients would like to comment on it now.

Phil McFadden – Simon, do you have any comment on that?

Simon Veitch (Dept. Agriculture, Fisheries and Forestry) – No, that's okay.

Phil McFadden – I would like to return to the comment that Ian Rae made, because I think you would like to test that –

Peter Woodgate – Very much so, yes.

Phil McFadden – As to how people in the audience felt about the point that Ian raised, about the fact that it started off with real clarity in the one spot, about the risk and hazard, but later on became muddied. Any other comments from the floor on that?

Greg Street – Maybe I will go back a step further, because I think the hazard is actually salt water. We map a lot of salt in the landscape in places where it is not a hazard. We see it in the tops of lateritic hills in Western Australia and it doesn't seem to have any hope of ever being remobilised. The real hazard is salt and water together. I wrote a paper on this, on defining salt hazard maps. It is not referenced here. Perhaps you take issue with the way I did it, but those maps we prepared were located areas we thought were salt hazards and then the maps defined an empirical risk.

Peter Woodgate – Greg, we agree with your point. We have made the point in the study. There are many workers who have also made that point and agree with it. I think there is an issue emerging here with your point, Greg Chapman's point and Ian's point which is fundamental to the definition of hazard and risk in relation to salt and water. This forum could really very usefully get some consensus clarity on those factors and how they interact with each other.

Andy Green – I am worried that I am going to sound like a pedant at this stage, but it does raise the issue of the definition so I suppose one can be a bit pedantic. Risk is 'the chance of something occurring', it starts out in this sentence [on page 18], and to me a chance is a probability. But then you say, 'Risk is classically defined as an impact ... multiplied by its likelihood,' which is also a probability. It seems to me the risk, when you define it in your formula here, is a probability times an amount of money, which is really something that is a weighted probability of expenditure or something like that. So it is not a chance or a likelihood. To me, I find confusion in those definitions there. Sorry to be a pedant.

Greg Chapman – I just want to make the point that there is an Australian Standard that has definitions for risk and hazard. My suggestion is that they be followed, because other agencies would also have various definitions of risk and hazard, and perhaps it is best to use a standard that is applicable for all of Australia.

Peter Woodgate – We have used the Standard.

Phil McFadden – Are there any other comments there? Really, the goal here is to try and help the team change their wording in order to produce a better-quality document. Ian, did you have any specific things that you could suggest for them to be able to put in there?

Ian Rae – I think the Standard is a dog's breakfast as well.

Phil McFadden – So you are suggesting we create a new Standard here, right?

David Robson – We have talked about data storage and where it should be. At the moment, from my understanding, it is in various departments and it is in various catchment management agencies. Representing the Geological Survey of New South Wales, which has distinct links to the other geological surveys in other states, I think maybe, as a proposal, the geological surveys should be the central storage of data that can then be accessed by the other stakeholders.

We have talked about standards. The Geological Surveys of Australia have instigated a number of data standards and they have a number of policies in place to do this. When I say the Geological Surveys, of course GA is part of the Geological Survey group. So that would be my recommendation.

Chris Pigram  (Geoscience Australia) – I am fairly confident – I will check it at teatime – that in Geoscience Australia, where we deal with the geohazards more widely than salinity, the definition of risk is actually the hazard times the likelihood, not the impact times the likelihood. But I will check on that and provide some feedback to the authors.

Nicholas Henry  (NSW Dept. of Infrastructure, Planning and Natural Resources) – I am with DIPNR, in Cowra. We are talking about data and information and collection. One thing I think we might have missed is farmer information, or farmer knowledge, about a very localised area. Farmers live in the countryside that we want to investigate, and they have been living there for generations. So there is a lot of generation knowledge there about, say, wet areas or how the landscape has changed over their generations. That might be a very important layer that we can combine with all these other layers that we are collecting.

Phil McFadden – I am going to draw this session to a close now. It has been a very useful discussion. I can see in my mind this TV series of DIY Rescue, because I see the tension of what we are trying to achieve here. On the one hand, we are desperately trying to make the situation understandable to the guy on the land, to the government decision maker, et cetera, as to what are the questions they need to ask, what are the kinds of technologies that will in fact give them an answer, but at the same time something that has been an ongoing theme through here is the complexity of interpretation, that you can't just take one bit of data and get an answer out of that. You have to have this broad, multidisciplinary approach, and a full range – the last comment there was about what has changed in the past. And how do you interact with your neighbours? There is a deep complexity.

The one thing we don't want to do – and I think I hear this missing in the document – is to convince people out there that because they have read this document and they have had one go at it, they now understand it, they can now think that they can go out and get the one tool, the one technique, and get a golden answer. We have to get across the concept that this has to be fed back in, that the information that people gather has to be fed back in to an overall system so that we can do information management and those kinds of things. I think we have to weave that complexity in, to make certain that we don't actually engender a DIY Rescue situation even worse in the future.

Thank you very much. I think it has been a very good session.

Session 1: Risk and investment analysis

Neil McKenzie – (CSIRO Land and Water) – I would like to follow up on what Andy was raising. I guess the central issue that we have got in providing natural resource information is whether we are reducing the risk or the uncertainty for decision makers. It is a very fundamental and a very simple principle, and I think it would be really good in the report to refer to the work of David Pannell, who was mentioned in the first talk. David has done a terrific job analysing what we often refer to as the indicators industry. In the last 10 to 15 years there has been an enormous amount of investment into indicators of natural resource condition, and his paper in Ecological Economics really puts the burner on whether that has been the logical thing to invest in. I think the basic principle of whether we reduce uncertainty for decision makers guides so much of how we then collect information, and it should almost be the first principle in how we set this up.

The second point is to some extent in response to Stefan's talk, and also Andy's. In the land resource assessment community, we have undertaken some benefit cost assessments of land resource survey work and tried to develop some reasonable methodology with ACIL in doing that. The result is interesting; the methodological problems are enormous, and I think the work that we had done a few years ago is a useful thing to have a look at.

Richard Lane –   (CSIRO Geoscience Australia) – I, like Andy Green, am way out of my technical league here, commenting on these issues. But my question to Stefan is this: is it too simplistic to conclude, from the very small benefit cost ratios that you are seeing out of most of these studies, that one or more of three situations exist? The first is that the problem is being overstated, leading to poor benefit cost ratios. Secondly, are the impacts being underestimated, leading to that same conclusion? And/or are the benefits being understated, leading again to a small ratio of benefit to cost?

Stefan Hajkowicz –   (CSIRO Land and Water) – I will go through those in reverse order. Are the benefits being understated? A lot of the benefits of salinity management are non-market; they are not things we have dollar values for, but they are really valuable – biodiversity, cleaner rivers. These are things that are important and we are purchasing on the behalf of taxpayers. We don't have a very good way of costing those, so quite possibly the benefits are being underestimated.

On the middle one, of whether impact was being underestimated, I think that is a question you have got to put to the biophysical community, the scientists who are assessing the level of that impact. We have tried to use, in all our analyses, the best available. We have tried to make relationships between water use and pipes getting rusty more quickly, between where salinity is emerging and crop yield loss. We are doing our best to estimate that impact as accurately as we can, and hopefully the science on salinity and its impacts will get better.

I reckon, though, a lot of the costing functions for water salinity that are used in economics are not really very good. So is it being overestimated? I don't know. It is probably not very accurately estimated at the moment.

Thirdly, is the problem of salinity being overstated? Well, I think we can say that there are a lot of vested interest in making salinity a big problem, as well. There are people who have to keep it there. (That's controversial but it's probably true.) When I do some of the hard-nosed economic analyses and just look at numbers coming out of a computer, I at least tend to find – and it would be worth bringing in other economists who have looked at this problem, but I think David Pannell would come to a similar conclusion – that it is not a huge problem to Australian agriculture compared with, say, the drought, compared with, say, a change in commodity prices. That is to the best available information we have currently and to what we have been able to gather so far from pretty bad information.

Phil McFadden – Could I just check a couple of things there. I would like to ask some of the AFFA representatives and government representatives how they feel about the statement Stefan has just made, about whether the problem is being overstated.

Mike Lee – David Pannell's work is very interesting. I think it helps put the potential impact on agriculture in perspective with other natural resource problems. But I actually put my hand up, Phil, to raise a question. I would like some discussion about how these economic techniques deal with irreversibility so that future generations really do not have the choice of doing something different. This takes us into a different league of evaluation, I think.

Phil McFadden – Does that come into the terms of reference which you were trying to get out of it, Simon?

Simon Veitch – It is an accompanying issue. Obviously, the salinity mapping methods review is purposely fairly narrow to the methods, but these are useful discussions because this is the real world of how these methods are going to be applied. And to the extent that they indicate management options into the future, ways that we can keep options open for the future and future generations, that is very useful. If we can have passing references in the salinity methods report that indicate those opportunities, that is great. But of course that is not the major focus of the report itself.

Greg Street – I guess I have some fundamental issues with the analysis, in that there are some very basic assumptions being made. The first is that we are going to continue broad acre agriculture through the wheatbelts of Australia. There is no long-term planning that I have seen that takes into account what might happen, what world markets might be in 20, 30, 50 years' time. Are we trying to protect a system that may not be viable economically? That is the first point.

I will go to Andy, because I have got some notes here on what he said. I really took issue with Richard George's and Peter Woodgate's paper on airborne EM changing the current trajectory of salinity. I fail to see how a data set can change a physical process, and that comes back to what I was saying this morning. I had issues with that paper when it was published, as editor of the journal. But Andy, I think, clarified that to some degree.

Andy Green – My interpretation of that 'trajectory' word is the trajectory of the management process – in other words, the management practices have not changed. That was my interpretation of it. Maybe the author could comment.

Peter Woodgate – One of the points we were making was that there is a disconnect between the collection of data, the interpretation of the information product and the use of that information product for decision making. For whatever reason, the collection of very good information is not having a significant impact on end user activities. Andy has made the point that it is probably because, in part, the options that we have got at this stage provide us with marginal economic benefit.

Greg Street – I will pick up on that, because one of the particular examples they – you – were referring to was Lake Toolibin. I hate to get specific in this, but Toolibin has been subjected to long-term studies over 20 years prior to the National Airborne Geophysics Project, and yet none of those studies had recognised the palaeochannel. This is something that comes through in the report, that if we have not got a remedial action, then we should not apply the data. But we don't know what the remedial action is until we collect the data. In addition, I would venture the example of your friend David Chadwick's place, where the availability of airborne geophysical data gave David the confidence to apply a land management plan which came out as a cost benefit of better than 1.

Andy Green – I didn't have time to go into it, but I think your highlighting of the palaeochannel at Toolibin is an important example, because it seems to me the places where it is easy to apply and you can find logical answers to the use of these geoscientific data sets are where you are protecting assets. It is very clear that these geoscientific data sets are much more focusable when you are protecting an asset. You have a thing: 'Where do I drill? What do I do?' and your work there I thought was very good. It is when you get into the large area stuff of, 'What do I do over huge areas in recharge reduction?' that it is harder.

Ken Lawrie – I have got a question to raise with regard to a broader message that is going out to a number of communities that we are dealing with. I mean the economists themselves, the broad sphere of the research community, and the end users themselves. I have gone to quite a few workshops, talked to CMAs and gone to conferences over the last year. There is a mixed message out there about what sort of certainty levels we have in our understanding of salinity hazard.

It goes from the extreme of the modelling conference in Townsville I was at earlier this year: 'We don't need any more data. We're not sure if we ever needed the data because our models are so good already.' It ranges from that extreme. And that message not only gets through to the research communities and the modelling community, for example, but also down to individual CMAs, right through to some of our colleagues who work on salinity hazard mapping and who say, 'Well, the best we can do with this has probably got about a 10 to 20 per cent certainty level on our ability to predict the salinity hazard at this point.' So there is a problem there, that there is a real mixed message going out to a broad range of researchers and the broader community. I think we need to have some sort of assessment or statement of that, because there is a very confused research and broader community there.

Greg Chapman – A couple of things. One is that scientists are basically paid and are expected to produce scientific papers which do not naturally relate easily to the people who are farmers, who have the responsibility for looking after their own land and making a living out of it. There is a gap between that and, basically, someone finding the information, getting hold of it, being able to translate it into something that makes sense to them, and then putting it to work on their particular property.

With our soil landscape mapping, we have land management recommendations for each soil landscape which are worked out through a process of dialogue with extension people, and the extension staff are extremely valuable in terms of translating that scientific information and putting it in a format that people can understand and hence make use of. If we do not have those staff – and they seem to be disappearing, their expertise is slowly dwindling away – then that gap will remain. We are not necessarily focusing geophysical information down to farmer level. There is a gap in translating a spectral signature to a particular set of wise or best practice actions.

The other thing is that there is often a huge time lag between a study being done and the adoption of that sort of study. Our land capability mapping in New South Wales was completed in the mid-1980s, and it is only now really starting to be adopted.

Phil McFadden – Greg, do you have anything specific that you think they should be changing in the document that would make the document better?

Greg Chapman – Well, I guess as far as my recommendations go would be to make sure that any information which is delivered goes through a process of dialogue so it can be worked out to geographically based best management practices, or sets of options which might be financially viable, and that there be some sort of a dialogue process with extension people so that they fully understand the whys and wherefores of how that information was collected and can therefore take that message and translate it across to the people who need to make the decisions.

Peter Woodgate – Greg, if I could just clarify that: are you suggesting that we need something up at the front of the document which says not just what this information provides but how this information is to be provided, through a group – in other words, be explicit about having some form of extension or interpretation group that provides a translation from the User Friendly Guide, even, to a farmer who has got no knowledge whatsoever of any mapping technique? Do you want something explicit like that at the front of the report?

Greg Chapman – I guess it depends on how we define where a mapping project starts and when that mapping project actually finishes. It does not finish with a report being lodged somewhere that someone signs off; it finishes with a process of extension and adoption. Otherwise the full benefit of it may take a long time to be realised.

Peter Woodgate – My comment was to the use of the guide. Your comment is also to the use of the mapping product that comes out of the application of the guide. So you would like to see that acknowledged in the report, so that the buyer beware statement acknowledges to the end user that they are going to need to have someone help them interpret the mapping product for the next part?

Greg Chapman – Yes. I think it's a case of not only, 'Don't use this product for these particular purposes,' which is what we seem to focus on, but as more a process of, 'No, this is what it should be used for, and this is how we recommend that it be adopted.'

Michele Barson – I would like to pick up on the comment that options are providing only marginal economic benefit, and I would like to propose something that might be a bit radical. That is that this really depends on your mental model of salinity. At the moment you are talking principally about dryland salinity. I would suggest that if you thought of dryland salinity as being the surface expression of the salt plus water equation – and quite frequently that dryland salinity is just one expression, the salt then moves on into rivers – if we see that as a continuum rather than two separate processes, the economic benefit of doing something about the supposed dryland salinity is really related to stopping water so that it doesn't pick up salt and move it into streams.

Stefan Hajkowicz – Actually, I would just add to that that the problems of salt in the water in our models do seem to be the major ones – like the River Murray getting salty. That is a big economic problem. That is registered that way in our models, too.

Andy Green – Again correct me if I am wrong, but then isn't it a question of what is the most cost effective way of protecting the river? Is it more cost effective to put more salt interception schemes further down the river, or is it more cost effective to revegetate large parts of New South Wales? That is an issue about the government money. We are in economic rationalist mode here, so that is why I am raising this.

Michele Barson – Okay. I will come back to what you mentioned earlier: the connection between geophysical analyses and on-ground management options. The on-ground management options that we have principally relate to controlling the water, so we have got changes in land cover – tree planting, for example – changes in land use, which might, say, move from cropping into pastures, or changes in management practices, for example moving away from long fallow. Those are the kinds of levers that are available to farmers. We also, of course, have engineering options.

The point of the exercise here is to say, 'Where is the salt, how much salt is there, what is the risk of mobilisation, and what can we do to reduce that risk of mobilisation?' We look at that in a number of areas and then we say, 'Which of those options is going to give us the best return on our investment? Is it 1,000 hectares of trees in highland areas of this catchment so that we are reducing movement of salt into streams? Is it changing from annual crops in this particular region? What is the mix of options that is going to give us the best return on our investment?'

John Lovering (Past President, Murray-Darling Basin Commission) – .It has been interesting today for me to hear that the salinity risk and the effect on agricultural production – I think I got this message – may not be as drastic as perhaps people have been told. But I would want to make a plea for the importance, I think absolute importance, of what effect, though, salinity in the future will have on biodiversity protection. That is absolutely crucial. We cannot just say that agriculture is the critical thing – it is, but it is not the only critical thing. Biodiversity protection for the future is something that we have a responsibility for, for future generations.

Phil McFadden – Thanks for that, John. One thing I would like to just raise is that Stefan made a comment that hazard and risk maps are pretty useless to him. That hasn't engendered any discussion at all, and I think that would be of some interest to the authors in the document. Does anybody have any comments on that, as to whether they feel that hazard and risk maps should have value, and where the problem is here?

Brian Spies – I think it is a key area that we do need to look at. It is clear that the hazard is the salt and the water, or a combination of those; risk refers to the asset. There are very, very many types of assets and they are in very many different places, unrelated perhaps to where the land use change is going on. So there are aspects about current land use and what the projections are, there are issues about 'what if' various types of scenarios, and when we get to risk we are talking about a whole range of various types, from agriculture to biodiversity, environmental infrastructure and so forth. I think a lot of the earlier risk or hazard maps were very, very simplistic, and I guess we would like some guidance about how to work through that and approach it in the report.

Ken Lawrie – Just on that issue of salinity hazard mapping: if you look at probably the most sophisticated of all the composite index methods now it is probably G-Sharp(?) in Victoria. [Inaudible name] and Mark Reid have done an excellent job there. That is probably our best effort at salinity hazard mapping, and they are intersecting that with the asset register. But the problem is they still are dealing with conceptual frameworks – the groundwater flow systems is a conceptual framework of what is happening with the groundwater and the salt. We have been in discussion with them about how to get that certainty level up, by getting salinity mapping methods for getting spatially explicit and quantifiable data into these frameworks, because if that is the best we have got, it is still rather conceptual.

Richard Lane (Geoscience Australia, CRCLEME) – On your comment that hazard didn't come into it, as I understood it it was a function of the model that was used to generate those predictions not having any observational input about salt stores and subsurface hydraulic permeability, groundwater flow – any observational input into that, except at the extremely broad range. It was a conceptual, not an observational driven model. Is that correct?

Phil McFadden – Simon Veitch, is this of value to you?

Simon Veitch – I am finding the broad range of discussion around salinity mapping really useful for the range of issues it is bringing out. Akin to that is the question about the cost of salinity to Australia, to add that in context. I think it is useful that this salinity mapping methods review is a means to an end, and that end includes salinity management but it also will have an impact on water management more broadly – potentially acidification, sodicity in the soils, et cetera et cetera – so, though it is unashamedly a device or a means to an end, to allow people to engage more effectively in natural resource management.

Paul Lamble – Just to comment on the economics: perhaps with risk mapping, when you are talking about an impact or a consequence or something like that, that is where the economics and the economists need to become involved so that your risk is delineated in dollar terms or by some other means so that it does actually have a use there, rather than units or tonnes of salt or something like that. You can then get your economist involved in the actual risk process, in determining economic impacts.

Also, we should bear in mind that economics isn't the be-all and end-all. How do you put a price on Murray cod existing in our river systems? You can't, and that needs to be very clear when we do do our economic analysis.

Brian Spies – Just a quick comment there: I think what you say is right. However, I think people are forced to put values on biodiversity and heritage for the future and Murray cod. We can't just say it is too hard. There has got to be some type of numbers somewhere, or some way to sum up what is most important and what management decisions do I make, this versus that? It is difficult to do, but I think it is part of the equation. I guess I would like to hear about whether people are doing that. Is that possible or not?

Stefan Hajkowicz – There are techniques that we are exploring, and I can talk more later about those.

Phil McFadden – Thank you again to both the speakers. I think that's really great, and you engendered a lot of discussion there.

Session 2: Hydrology and modelling systems

Mike Lee – Phil, has there been a false dichotomy going on here: 'You do this, or you do salinity mapping'? Is that the real issue, that there is some sort of sense of competition between this approach and getting out and flying it, whereas the answer is probably in the middle, as recent experience has shown?

Phil Dyson – I hope that is not what came across, because it wasn't meant to. I guess what I am trying to say is that I think there is a point in terms of planning for salinity management where you use all of the information that we have generated, and all the expertise that we have available, to generate this kind of cut [the 10 systems on the final slide] and then you go on from that and you use your mapping techniques, be they geophysics, be they whatever, in a very strategic way to answer very specific questions. I think we should be auditing what we know and what we don't know about each of these systems that we have defined, and then seeing whether the geophysics might be able to answer those questions in a very structured way.

Mike Lee – In relation to the Queensland salinity mapping, which has been quite controversial, the problem seems to me to be a lack of system understanding of how the landscape works. Then you might try and extrapolate it with broader techniques. What is our state of knowledge?

Phil Dyson – What happened in Queensland, I guess, was that the Queenslanders actually developed what we would all call a hazard approach and what they would call a hazard approach, a composite index map, and a lot of people have gone down that path. It says that if you know how wet a landscape is, if you know how much recharge it is going to have, if you know what its salt store is, you can put a weighting on all of those figures and then you can manipulate those figures and give them a score at the end of the day. And you can distribute that score spatially within a GIS environment.

But that process itself, that way of mapping, comes up with a hazard map that is not necessarily informed by a process map in terms of the groundwater side of the story. So where we got to with the Queenslanders was an understanding that we shouldn't use one without the other. In other words, our understanding of whether you would go to a risk situation as opposed to a hazard situation had to be informed by process. Does that make sense?

Mike Lee – Thanks, it does, Phil. The issue is to what extent we can view those maps as useful pictures of hazard, in the absence of their being derived without much idea of process.

Phil Dyson – Are you talking about the composite index maps, as opposed to the groundwater flow systems maps?

Mike Lee – Yes, the published Queensland maps.

Phil Dyson – I am very considered in my response, because most of the people here that know me know that I am not a fan of composite index maps. I think that they give you a broad indication, but if we don't use them without incorporating our understanding of risk, informed by our understanding of process, then we are in an uncontrolled environment.

Glen Walker – I think it depends on the stage of the process. One of the difficulties there is that a lot of the hazard mapping type technologies that people have talked about have been in the very early stages of coming up through that NRM cycle that I guess I was trying to say, and quite often a lot of the information on process understanding wasn't there. But you still need people to be aware that there might be an issue in terms of precautionary principles or in terms of thinking of the sorts of policies or where you actually start putting your effort in.

One of the actions you would think of coming through is actually to get that process understanding under way, and I think whether it is a composite index or whatever method was used on the amount of information that was available at that stage, it was always going to be limited by the amount of data that was there. At least it was a first sieve to say, 'Well, what areas do you actually put your effort in, and what areas don't you put your effort in?' and then start trying to get that process understanding. So I think it is actually at the stage of that NRM process and I think in southern Australia we have tended to go past that stage and I think we now would never go through that, do that, without a process understanding. I think part of Phil's prejudice that he is talking about is reflected: it does not make sense to do that in, say, Victoria, when you have got the level of information that we have now.

Phil McFadden – Mike, could I ask you a question? As a user, reading the documents, does the document give you useful information from this point of view, or do you feel that it needs some alteration?

Mike Lee – No, I think the document is very useful, particularly if it leads us to some common understanding of what a hazard map is or should be. I am sitting next to the Executive Director of the Land and Water Resources Audit here, and the next picture of how we present Australia's salinity situation will possibly be very different from the last one. We will, hopefully, be underpinned by more process understanding.

In relation to one of the things I am seeing in the document, I guess that this suggests the broad brush picture first and then more detail. I think my initial comment about extrapolation of understanding went to the point of having some firm process understanding, to take that more widely by using broader brush methods.

Ken Lawrie – I have a question for Glen. Reading the document, Glen, I felt it was a bit light on in the modelling. One of the problems we seem to face all the time is that we can fly all these wonderful 3-D volume data sets, but we seem to have a limited capacity to turn these into 4-D dynamic models which are spatially explicit. Would you like to comment on that?

Glen Walker – I have to say the authors are probably in a no-win situation there. I think that is probably the fair thing to say. It was almost that first cartoon that I had [on first slide], that the thing we could talk fairly explicitly about is the mapping technologies, but I think people want to ask how you put it into that hole. Not only is it into the models but also into that whole process, and that in itself is something that is probably the subject of another document in itself, as to the actual modelling technologies and things that use the maps. I think it is just a matter of drawing the boundaries but looking to see where the hooks are into those various types of modelling technologies. So, in terms of the balance in the book, I think it would have been very hard to go into modelling technologies and be able to do it justice within the scope.

Chris Pigram – I wanted to go back to Stefan and his analysis, and ask about a question that Mike raised initially, which was about the irreversibility of some of these processes and how that impacts on the economic analysis, and whether it is factored in. Indeed, in the information that you showed us, Stefan, what are the bounds of uncertainty on that information, in terms of the impact on the salinity? The reason I am concerned about that is that you said basically, to paraphrase, if there was not a substantial base you would necessarily have to make it up to do the analysis. That in itself must introduce a whole range of uncertainties. I am just wondering if you could give us some sense of what those ranges are, and also if you would like to comment on Mike's original question about irreversibility.

Stefan Hajkowicz – Okay, on the uncertainty question first: I think we have to say there is quite a lot of uncertainty about our economic analyses of salinity, partly because the economics are still grappling with getting good numbers but also because the science on salinity is still grappling with it all at the moment as well. We don't have great nationwide salinity mapping data sets currently available to do those analyses on. We use the best we have got available. The details are all there in the reports, but there is quite a bit of uncertainty attached to it.

Chris Pigram – Sorry to interrupt, Stefan, but do you try to quantify that? Do you actually do an analysis of that?

Stefan Hajkowicz – I think there are some statements around it. It is very hard to quantify some of that uncertainty because you don't really know very much about the data that has gone into it. We are not even in a really good situation to be able to quantify that uncertainty yet, except to say that it is there and it is probably large. I would like to be able to give you a more precise answer, and maybe we can talk more later, but that is about as far as I think we can get with some of the economic analyses of salinity – the national stuff we have done and the mapping. We are not even really able to quantify the uncertainty very well.

The irreversibility question is a really interesting one, and no, that is not handled in economic analyses very well. If there is a sudden collapse of a river system and all the biodiversity dies permanently and we can never turn it back again, or Adelaide's water supply is no longer drinkable, then those are big consequences that we have not put into the economic analysis yet. We need, as economists, now to start exploring new techniques for handling efficient resource allocation and investment analysis that allows us to move outside of monetising everything inside benefit cost analysis. It is a big concern, and no, it is not factored into most economic analyses currently done. But we are working on it.

Chris Pigram – If I can summarise that: what you are saying is that not only are the science tools, from a point of view of understanding the hazard, inadequate but the economic tools for assessing the impacts are also inadequate?

Stefan Hajkowicz – I would not say they are inadequate. They still tell you a lot, and probably as much as we can tell you. About the science tools, I am not the right person to answer that question, and the science that has been done is fed into our analyses that we have done the economics on. There are other people here who know a lot more about the scientific aspects of analysing salinity; we have done the economic interpretation on it.

But you have also got to remember it is the best available, I think. It is never going to be 100 per cent, and I think we have done what can be done with what we've got.

Phil McFadden – Chris, I think that was a yes.

Ian Lambert – Phil and Glen have emphasised the importance of a strategic approach, and I think that is something that perhaps could receive a little bit more emphasis in the report, to temper what could be seen as a technology driven or reactionary approach.

The question is particularly of Phil. You are doing some very, very good stuff in compiling information and showing us the groundwater systems. The question, though, is: is there anything that government agencies – and here we are talking geoscience, land and water agencies and others – could be doing to further the information base on which this important strategic work is drawing?

Phil Dyson – The answer to that is a definite yes, but the way forward involves perhaps a more strategic institutional approach, because we more often than not don't get our act together because we are always working in this bit, and this bit, and this bit. We are all working on the same system but we are doing it in isolation: this person is talking to that agency, and this person is also talking to that agency, and there are two or three projects doing the same thing. It would make a lot of sense, I think, if we actually did sit down and say, 'Where are the holes in our understanding of these groundwater landscape systems?' and, 'Where are the most strategic places that we can use organisations like GA or CRCLEME or BRS or whoever to resolve those questions?' So the answer to it is an institutional thing, I think, as much as it is a very biophysical, landscape, groundwater issue.

[Unfortunately the last 5 to 10 minutes of discussion were lost due to a technical hitch. The ensuing discussion concerned evapotranspiration and run-off; the need for a strategic approach to vegetation placement for salinity management (ie. sometimes it can make the situation worse); and scale issues regarding salt flux and various data types].

Session 3: Mapping salinity in the near-surface and upland landscapes

Phil McFadden – I guess I heard there two of the tensions that exist here. There is the military intelligence view that one should not show classified images. But, on the other side of the coin, classified images are extremely useful to those who really do know what they are doing with them, and who have all the ancillary information. It is one thing to put information out to the public and it is another thing to educate them well enough to know just how to make use of that information. We also hear the tension once again about ground truthing.

Colin Pain – One is the point I made right at the end, that the data scale has to be equal to or better than your process scale. That is very important, I feel, if we are going to explain any of these things.

Bruce Dickson – I mentioned about the accuracy in the digital elevation models. I think we need something about that.

John Lovering – This time I am wearing a hat as the Chair of the Board of the Australian National Seismic Imaging Resource. It was mentioned just before lunch, and it may not be appropriate to what we have just heard but maybe it will be to the next group of speakers.

What was said, I think by Greg, was that there was no mention in the document on seismic methods. The Australian National Seismic Imaging Resource is a cooperative venture between Geoscience Australia and the ANU – state of the art technology and state of the art instrumentation. It has been used for quite a bit looking at quite deep sounding structures, looking down even into the mantle, but I am wanting to see it used for shallower features, in the areas that I think would interest people concerned with groundwater pathways and so on.

I think it is important that you look at this. I will make sure through Chris Pigram that you get an input from the ANSIR people on this technique. It is an important one, I think.

Phil McFadden – Thanks, John. Yes, I think the critical point you are making there is that seismics can give you information about fairly shallow structural controls which might affect the salt.

Alan Willocks – I would like to make a comment specifically about the report, in terms of regolith mapping. There appear to be some confusions within the report as to what the definition of regoliths really is. The notion of how Colin has been describing it there, using radiometrics to map regolith, is a common approach that is being used in a number of the state governments. However, the report describes regolith as something that starts at two metres. We probably need to get Colin and the group together to have a look at that.

In terms of integration of geological data across the area, through to salinity mapping, I would like to relate the way that geological mapping is occurring within some of the state organisations, and particularly within the Geological Survey of Victoria, where we are using geophysical techniques with geological information to ensure that there is an integrated picture being developed for a geological representation of an area.

A few years ago that was something that was difficult to achieve. If the geophysics didn't agree with what the mappers knew, the geophysics must be wrong. I think that in terms of salinity mapping that is where we are at the moment. The geophysics needs to be proving itself each time, on the basis that the hydrogeological models are there and they are sacrosanct models and can't be touched. If the geophysics doesn't agree with the model, therefore the geophysics must be wrong.

There doesn't appear to be a great emphasis or even a great way forward in terms of how all of the data sets actually get integrated together. I think if we could address that disconnect between the scientists so that there is a better integration of that work, there would be a significant step forward in the way that these data sets are actually used.

Phil McFadden – Thanks, Alan. I think I heard some pain there, and it is pain that we hear in all sorts of areas when the scientists themselves can't get their act together and come up with an integrated response. It makes it very difficult for users out there who are trying to make management decisions when the scientists, who are meant to know what they are doing, come at it from different points of view. I think largely that is what this is about.

Alan Willocks – I have also been involved a lot with the catchment management plans that have been going around Victoria at the moment; they are currently going through a strategic set of plans that are being developed. The integration of anything relating to a 3-D understanding of the geological system is basically not there. It is not something that is being recognised within the catchment management plans as even being important.

Chris Waring – (Australian Nuclear Science and Technology Organisation) I would just like to take up Colin's point there and concur. I agree very much that technologies are scale dependent and that there is an omission from the report, in that electrokinetic geophysical technique is not included. It is a very valuable technique to include at that engineering implementation ground scale, at paddock to farm level, or engineering implementation, where you need to know exactly where to drill or place your drainage system, or to find the fine detail for implementation.

Baden Williams – Really, with the salinity mapping what we are doing is looking at the symptoms of a hydrologic imbalance. Salt just happens to be something that gets moved around in the landscape by water. That has become quite an important aspect. There was a question before lunch about where we should plant trees in the landscape. Well, you can do that by exactly the same salinity mapping, with electrical conductivity, and come out with an answer that is useful to the farmer. He may not give a damn about whether it is 1% salt or 2% salt, but if you can tell him that this part of his paddock is letting water move vertically down the profile faster than an area next to it, then he has got a management tool, or he has got the data or the information which allows him to implement any number of management tools for controlling that rate of water movement into the landscape.

Now, some people would immediately say, 'Let's plant trees.' That is a pretty dopy idea in most instances. I don't think there is one instance that I know of where planting trees has made any difference to the hydrologic balance in the subtending catchment, whereas perennial pastures, whether introduced or native species, can make a great deal of difference and be integrated into a farming system.

Let's face it, all of this talk about mapping and scientific techniques and going off to catchment management authorities with that data, or selling it to them, isn't going to make a skerrick of difference unless individual farmers – and all of those individual farmers – do something in their paddocks. And they are only going to do that if they are given information to which they can easily relate and from which they can see that there is an on-ground solution to the problem.

Phil McFadden – Baden, have you read the documents?

Baden Williams – Yes.

Phil McFadden – Do you feel that they contribute anything towards resolving that issue?

Baden Williams – Yes, I think it is an excellent document. I couldn't have been happier when I took it off the Internet and found, 'Yes, at long last here's a sensible document discussing all the different mapping techniques' – which I thought was the topic that we were coming here today to talk about. But we have talked about many, many different things other than mapping techniques, and this is the first part of it.

In my submission I think I made mention that the mapping technique actually has a spin-off in terms of mapping recharge/discharge parts of the landscape – the relative rates of recharge and discharge – and that salt in itself is just incidental to that.

Greg Street – Bruce, I am currently doing two projects with radiometrics, one with airborne and one with ground radiometrics. (Ground radiometrics does not appear in the report, and you [Peter and Brian] might put that in.) The ground radiometric area is close to the Darling Fault, and in an area of 100 hectares, using a totally unsupervised classification, we got 15 classes dropping out of the data, converging very well. We gave those to the client and he said, 'That's too many.' We field checked them and we got down to six. He said, 'That's too many.' We ended up with two. Really, in terms of land management, that is what he wants. The unsupervised classification probably gave too much. I think one of the points in airborne – and ground – radiometrics is that there is more information in it than the client actually requires. The unsupervised classification is not a bad way to go at all, from my experience.

Just going on from that: the interpretation of these data sets is very different from what people have done in the past with them, in terms of mineral exploration. In mineral exploration we have always tended to take the data sets and to look at them and immediately throw out half the data and then gradually zoom in on targets. Target recognition is very important, generally, in mineral exploration, whereas in this sort of work every pixel is important to the end user. And a pixel by pixel interpretation needs to be put into this report, because that is how you do it.

Phil McFadden – Could I respond to that, perhaps, before we go on. I can't resist tweaking you, Greg. Did I hear you say you went out and did some ground truthing? I just wanted to check that!

I think you're dead right, but I would articulate it in a slightly different way. One of the things that are so important with these things is for us as scientific practitioners to get absolutely clear in our heads what question the other guy is wanting answered. So often we come along and answer all of these questions which he doesn't want to know about – they're not relevant to him. We really do have to tie down precisely what it is that the client out there wants answered, and then answer that for them.

Brian Spies – I have a question on that. You said you did the unsupervised classification for the farmer. What do you think those two classes were actually giving information on? Was it soil type, was it moisture, was it weathering?

Greg Street – In the end they were soil associations – that's a better way to put it. The last two classes that we ended up with were a loam through to a loamy sand versus a heavy loam through to a light clay. This was a land capability analysis, and in terms of land management units those soil associations were large enough to be managed on their own. However, the use-by date of these data sets is often very long. You can take an interpretation data set like that and then perhaps with improved irrigation technology (this was an irrigation area) we could go back again and, with sprinkler head designs as they are today – they can tweak sprinkler head design to deliver different volumes of water to different parts of the paddock – we could design a sprinkler system based on radiometric data, for instance, which would deliver the right quantities of water right across that paddock, using these sorts of classifications from radiometrics. And there is not much else that can give you that level of detail.

Brian Spies – The follow-on question for that would be: how does this impact on salinity? While it is at the scope here, what you are probably saying is that down the road, by the use of the right amount of water, you might reduce the amount of salinity somewhere else?

Greg Street – Well, to go back the other way, I suppose: in terms of salinity, we are mapping recharge potential. Simon Cook, of CSIRO Soils, a couple of years ago produced some leakiness maps for the Alaskan catchment, giving some idea of how much water was going through the different soil types, using radiometrics as his basic data set. It becomes a cropping management tool as well, and you can't separate those management tools from salinity when you are working in a commercial enterprise such as cropping.

Bruce Dickson – Greg, I would just comment that I said 'inappropriate use of statistics'. I didn't say statistics should not be used. In fact, the example that Colin showed was an appropriate use. He said he first confined the area he was looking at by geology, and you confined your area by size, a small area. So they were appropriate use of statistics. It is the inappropriate use, of looking at big areas, that worries me.

Greg Chapman – I would say that there needs to be a lot more emphasis on recharge mapping. In some areas, mapping of recharge is probably the most important thing because that is what can be managed. We can't sometimes do much about salt stores; in other places we can't do much about changing the groundwater flow systems.

The other thing, and I guess this is an integrative side of things, is that if we are looking at the landscape, with the landscape going through an evolution that has involved the geology, climate, the soils, groundwater flow systems, salt stores can probably be mapped using a single mapping unit. That mapping unit can be a salinity land management unit which is fairly important for being able to divide up the country.

Alan Willocks – There seems to be a disconnection even in here. We are talking about salinity and salinity mapping, and clearly that is what the terms of reference are. But I wonder whether we really need to be thinking about groundwater management and groundwater as well. Fresh groundwater is just water without any salt in it, so maybe there is a continuum between the freshwater and the salty water. Some of the techniques that we are talking about can assist also with that whole catchment management planning, and finding areas of good groundwater quality and good yields is probably as important as, if not more important than, being able to map where the salt is.

Mike Lee – Can we have some discussion about the overlay of the groundwater flow system approach and the regolith approach that Colin was talking about? Do they illuminate each other? Do they add detail? Does one fit in with the other? Is it extra context, one for the other?

Colin Pain – I think basically the two approaches, if they are indeed two approaches, are pretty complementary. What we heard this morning from Phil [Dyson] was the conceptual frameworks within which the groundwater flow system model works. What I was doing was looking at the finer end of the scale, I guess, and much of what I said could still be regarded as a conceptual framework. The step that you take after that, though, is to use whatever techniques are suitable to come up with spatially explicit distributions of these different systems. I guess that pretty much sums it up.

Richard Lane (Geoscience Australia, CRCLEME) – This is not so much a positive contribution to another mapping technology but more a wish. I am a little concerned – but maybe somebody can allay my concerns – about the reasons why we move closer to getting adequate three-dimensional maps of salt store without having adequate observation based maps of hydraulic permeability to use in our hydrological models. How important is it to know the other half of the deal, how easily groundwater will move through the subsurface? Is there any value in knowing where the salt is, if you have got orders of magnitude uncertainty in how water will move through those particular units?

As an example, this crops up all the time, in that very clay rich units often have very high conductivity and you can relate that to that they hold lots of salt. But then somebody comes along and tells you that it is totally immobile because its permeability is so low. And then you say, 'Well, how many determinations of that permeability have been made?' and the answer is zero.

So is there anything out there that can measure such things? And, as a second question: is it important?

Phil McFadden – Is there any resistance to that – apart from the obvious, that you should never spoil a good story with actual data?

Brian Spies – There is a study from Queensland in one of the submissions – maybe someone here from Queensland can talk about it – about digging great big holes in the ground and going back over several years and looking for how fast water was draining out and the way it was draining. Does someone here know about that study? We will mention it in the report somewhere. But I do think it is very important.

Richard Lane – We may have examples where, say, with an airborne survey we have made several million determinations of the subsurface conductivity, and that will be linked to possibly zero determinations of something that to get the flux of salt water requires a multiplication by some other term that varies by many orders of magnitude.

Phil Dyson – When we have worked through the groundwater flow systems approach for each of the landscapes that comprise those groundwater flow systems, we have done the very best we can in putting the kinds of numbers on that you are actually talking about. So, in describing the regolith and how weathered it is, or the fractured rock, we try to put a hydraulic conductivity, a transmissivity, on all of those areas, and we have spreadsheets that sit behind each one of the regional maps and do that. I am not necessarily saying those numbers are all right. They are the numbers that we collectively agree are appropriate to each of those landscapes.

Glen Walker – I was going to add that the sort of technique that you talked about was the sort of thing where, to help with getting those sorts of numbers, there are a range of physical techniques. I think that sort of technique that you talked about was the sort of technique that has led to these sorts of estimates.

But I guess the point I was trying to make, which I hope I made this morning, was that some of the mapping technologies, particularly some of the geophysical techniques, were on materials to which by surrogate have permeability associated with them – again I am going back to the South Australian geophysics. Most of the things were related to the materials and permeability of materials, rather than salt stores. And whilst we didn't have absolute values coming from it, that told us a lot about where there was retardation to flows or where flows might occur. So I think it does give information on it.

Chris Waring – I would like to take up from Richard's point. EK gives you an interpreted hydraulic conductivity, and it provides a bridge between the salt store, the electrical conductivity from the airborne data... [end of tape recording, but mention was then made of borehole data]

...greater volume of measurements based on a geophysical technique to link between the two.

Robin Greenwood-Smith – In the minerals industry recently the use of ground probing radar, or whatever you might want to call it, has been quite successful in mapping things such as palaeo-channels close to the surface, say looking for diamond carrying buried rivers. I think you could look at ground probing radar as being a technique that is worth investigating, and there are some experts in Australia – probably the experts in the world in this area. I will try and get some input for you on that topic from my contact in that area.

Session 4: Mapping deeper salinity in intermediate and regional groundwater flow systems

Greg Street – Richard, the ASEG met some months ago in Perth for a policy meeting on the future of geophysics in Australia, and recognised that the drop in numbers in mineral geophysics was a danger to the society and they really needed to embrace environmental geophysics, otherwise the society was going to be at risk. It was pretty well embraced by the society that environmental geophysics was a third stream of the society and would expand. I am associate editor for environmental geophysics for Exploration Geophysics, and in addition the ASEG has sponsored a short training course on the lines that you mentioned. It is a three-day course on information from geospatial data for natural resource management. The first one will be held in Perth next month, and you can have a look at it on salinity.org.

Robin Greenwood-Smith – Richard has, I think, gelled for me a problem I have had today with this meeting, that we have got two documents that we are reviewing, one a technical report which everybody wants to have expanded and made more complex, et cetera et cetera, and a users guide which I think we all would want to see made much more simple and probably not even distributed to some people because it might give them the wrong approach.

The analogy that Brian produced this morning with magnetic resonance imaging of the brain for a tumour, though, makes me feel that what we are providing as a users guide is in fact one where the patient would go along to the doctor and say, 'Doctor, I want you to cut this tumour out, it is this size, and I want you to use this particular knife.' I don't think that is what we want. So I think we have got a problem: we are actually trying to solve two particular directions with one set of solutions. I think we need to consider that at this meeting and maybe also in the report.

Brian Spies – Would you have any suggestions?

Peter Woodgate – I think Greg Chapman was suggesting that there needs to be a body in between, to interpret the reports and then deal appropriately with the end user audience.

Greg Chapman – I haven't got any particular hat on here; it is just that I have been involved with catchment management boards and committees and so forth and actually been on a couple of those, and generally there are mostly people from the community who are interested but may not have a technical background, or very rarely have a technical background. So it is going to be quite difficult for them to make sense of things, especially when you are getting into fairly advanced airborne geophysics, which probably a lot of the people who would be doing the translation between land management and the information coming from the sciences might have difficulty dealing with. So it is going to be quite a challenge for those people.

But in general I would also make the comment that the people who are doing the extension are basically a dying breed. For whatever reasons – I guess through managerialism – there seems to have been, at least in some places, a trend towards doing away with those people, and it is going to be very difficult to pick up the skills again.

Phil McFadden – Robin, may I comment while the microphone goes to Simon. I hear what you are saying, very strongly. If I could take the medical analogue: there was a time when you used to go to the GP and he was God, because the patient had no concept whatsoever, was totally ignorant, and had to rely 100% upon the GP. I think most medics today would agree that there is much better health care as a consequence of the public having some education and being able to engage with the medic and actually discuss with the medic what the treatment will be, at some level of education.

One of the problems we face at the moment is that many of the users out there at the moment, the people who have the problem, really do not know enough to be able to engage sensibly with the practitioners in order to actually make some rational decisions for themselves. I hear your problem, and it is a very careful tension that has to be managed, but I think one of the real goals is to raise their level of education to the point where they will actually be able to engage in a sensible manner.

Simon Veitch – Phil, I agree with those comments, thank you. The objective of this study was to remove some of the existing unknowns surrounding salinity mapping methods, at least to get some agreement about the nuts and bolts, the components of how we might investigate salinity in landscapes. By doing so, of course, we do raise all these other questions, and that is very helpful. I think your comment, Phil, is quite right, about raising the level of understanding through looking at mapping methods as a way of getting an introduction of a lot more people to the issue of these methods, that they should not necessarily just pull anything off the shelf when it first comes along.

So, with the shortcomings of just looking at salinity mapping methods acknowledged, it is a good entrée to a whole lot of extra thinking and engagement. It is a device, getting people engaged in a process. I think that is very constructive. It doesn't concern me at all that it is raising a whole lot of other issues to do with interpreting data and information for end user purposes, because it is pretty clear that we do need to bolster that part of our natural resource management.

Mike Hallet (Geological Survey of New South Wales; NSW Dept. of Mineral Resources) – I wanted to hark back to something that Andy Green said, which as far as I can see is underpinning this whole process: trying to relate what we are talking about here, in all this lovely science, to actually facing on-ground management issues.

There was a mention of a paper that suggested that some of these mapping techniques had not actually changed the direction of land management planning. I found that a little bit hard to believe, because I have been personally involved in some of these airborne projects. My field of expertise is specifically airborne EM, but I am sure it relates to other airborne techniques as well, such as Landsat, hyperspectral or whatever.

It seems that there is a lack of flow of data coming back after these things have been done. Richard mentioned the same thing just now, and I think that was an excellent point: what the heck has happened to all of these previous surveys? What has happened to the information flow that could be coming back afterwards to find out what sorts of management initiatives have been made after the work was done? That is some sort of channel that needs to be opened up.

The last point that I wanted to make is that when I was involved in these there were some very good on-the-ground land managers or ag scientists that were guiding the surveys through. I was involved as part of the NAGP, and they were able to, as scientists and geophysicists, relay the geophysical information through to the landholders. If it doesn't exist in any form now, there needs to be some sort of multidisciplinary team that can actually make more available, or in layman's terms explain to the landholders, the whole range of issues and mapping related science that is put forward. There is a lot of very good science, but there is a need to make sure that there is a team that explains it to the landholders and makes sure that the data gets applied properly. That really is, as far as I can see, the very basic point – that the data needs to be applied.

Neil McKenzie – I would just like to sound a note of caution, in that when we are talking about landholders, to speak about landholders as if they are one homogenous group in terms of their technical sophistication is a grave mistake. Our experience has been that so often when we are involved in providing the information that we are gathering on landscapes to landholders they have a capacity to not only understand what we are trying to say but they are generally several steps ahead, and they are asking questions that we can't actually answer. A lot of the questions that landholders need information on are to do with the timing of inputs, to do with having predictions of nutrients or of the potential effect of a crop on water use, and so often our technical capacity is just not there. In a district we often interpret disinterest in our grand sermons on what we are doing as perhaps a degree of ignorance. I think we often go in with a slightly arrogant view, and often we cannot spot when we are actually way behind the eight ball.

So we are dealing with a tremendous range of abilities in our clients, and increasingly, particularly with corporate farming in a lot of the cropping lands of Australia, we are dealing with very, very sophisticated users of natural resource information.

Phil McFadden – Thanks, Neil. I think that is very important. I think to some extent what you are saying is that it is really incumbent upon us not to assume that we know everything, but to go out and listen and find out very carefully where this other person is coming from and what their needs are, before we start getting stuck in, thinking we know everything.

Ian Lambert – I would like to return to the issue of what I also spoke about earlier, the gap between this report and the users, and how we can bridge that gap. We have said it is difficult. We have said the users don't understand a lot of this, and various things. But surely there are very specific suggestions that could be made. There could be independent panels that catchment management authorities could be advised by. These panels could have the relevant state government agencies, maybe federal government and private sector people on them. They are the sort of people that understand what is possible, understand the broad environment in which the catchment management authorities are working, and could give independent advice.

Also, there could be government sanctioned demonstration projects which bring together all the players that can make a contribution in a specific area and involve the users in that area in these studies, to demonstrate how these things can really work.

I think they are two very specific things that can be done. They do require us to break through institutional barriers and think more strategically in some ways, but for goodness' sake, it is very important in the national view that we do get through some of these things and give people the best advice and the best demonstrations of what is possible.

Andy Green – Could I comment on the people who are going to do the uptake. There are quite a lot of people out there who I think will be significant uptakers. They are the people that Tim was referring to in his talk, who were interested in protecting the Murray River. Those people that are engineers, that might be shire engineers, that are concerned particularly with protecting assets, are real good listeners and really sharp, and take things up quickly. So there are almost these two separate potential sets of users for this data, and for one group I don't see nearly as much problem as for the other, more general landholder users. I think that should be borne in mind, because many of these data sets are going to be better focused at these expert users rather than the landholders.

Greg Street – I agree with Andy that there are certain groups who are going to take it up. We called just last week for an airborne EM tender in the Lake Warden catchment, just outside Esperance. It is a biodiversity catchment and there are issues in terms of saline biodiversity to be addressed in that. But the airborne EM survey will probably be used mostly to create the structure for the groundwater model.

I would like to comment on Ian's comment just a moment ago, that we need more case histories, and from perhaps what he was saying I would hate to go round the process again. We have done it twice in airborne EM surveys around Australia. We did it under the National Soil Conservation Program, I think, and then under the National Landcare Program, the National Airborne Geophysics Project. Unfortunately, the results of those surveys have not really hit the light of day. Even for the review of the National Airborne Geophysics Project, the report is still in draft form, some four years later, I think. It would be nice, if you want to do something, to put some money towards finishing off those case histories and getting them into publication, without doing more work at present.

Peter Woodgate – In relation to the National Airborne Geophysics Projects, there are now eight reports in that series and they are going through an international peer review process. They are probably about four to six months away from actually being completed through that process and then published. But they have been on the Web for the four or five years since they were completed, so they are out there.

I don't think it is so much a question of their going to the publication stage. I think it is a question of making widely known the conclusions of those reports, so that people can come back to them and move forward.

Greg Street – Those reports are from the government agencies, and they are not referred to in this document. They should be. The original reports supplied by the contractors as part of that are not published at all and not referred to – and not referred to in the reports that I have seen on the Web. That is unfortunate, because that is where most of the information probably is. Look at the recent special publication of the ASEG. There is one [case history] in there.

Mike Lee – I was really pleased that Tim opened up slightly broader the discussion about AEM, to look at materials and see what is going on below the surface as well as talking about salt stores. But I would just like to talk about the absence of salt stores.

It is going to be really terrible if we invest a lot of money in getting farmers to reduce deep drainage when deep drainage is not relevant to anything. I'm sure that is happening around the country. It is going to be terrible if we invest public funds in planting trees where planting trees is not relevant to recharge and what have you. So, while a lot of the discussion is about maps to identify salinity hazard, it is really important that we are able to do some work to exclude hazard where possible.

I am not sure whether it is a complete counterpoint to the other argument. There may be some applications of this technology so that there is very little or no salinity hazard; that would be a very great advance. So my question to all the scientists here today is: are there faster paths to excluding salinity hazard as well as identifying hazard?

Phil McFadden – I think it is a really valid point there, Mike. That is what Alan Turing did in trying to sort out the Enigma problem – he came up with a way of excluding 99% of it so you could go and look at the other 1% and come up with a solution. So I think it is a very valid point. Thanks very much to the speakers this afternoon.

Discussion, synthesis and conclusion

Phil McFadden – This is an open session discussion about all that has gone through today. Each of the speakers is up here and they are forming an expert panel on this. We have had some very good discussion today, and I think we have got an opportunity here to engender some discussion.

We should now move into an open discussion about anything that you wanted to raise that is related to this. From the panel themselves, were there any points that you wanted to raise? If not, let's get started.

Brian Minty – (Geoscience Australia)My comment goes to the overall structure of the report. I think the user friendly version reads quite well, in that it goes from the user needs, with their very specific questions, through an understanding of the problem, getting to know the big picture and so on, followed by the mapping techniques. So the implication there is that you need to understand the big picture if you want to answer the very specific questions that the landholders might have.

The problem is I don't think that is reflected in the main report. Part of the problem is section 5.5, Defining the Questions to be Answered by the Investigation. Very specific questions there are asked, like What is the problem? Will it get worse? Will it stay the same? and so on. These are the questions that the landholders are going to ask in order to select a mapping method to solve their problem.

It doesn't quite work like that, because no, there are no mapping methods that are going to solve those specific problems. You need to understand the whole problem. So I think maybe putting in some of the bigger-picture questions in there would be better, or just not losing that thread of going from the big picture to the mapping methods.

Phil McFadden – Thanks, Brian. I think that is a good point. I think Robin [Greenwood-Smith] actually raised this to some extent. What we don't want is for people to read the document and think, 'Yep, I see which one is my particular silver bullet – because I really did want a silver bullet,' and go running out to use that one silver bullet. We have to be careful that that isn't an outcome of what happens.

Ken Lawrie – I would like to raise a related point. It also relates to Mike [Lee]'s point just before the interval. We are talking about the big-picture stuff, and I think it does get lost, as Brian said in that connection, in the document. But it also comes down to whether we are talking about targeting specific small salinity outbreaks or predicted salinity targets with small-scale surveys, or whether we can look at the cost of acquisition and savings that can be looked at if we are looking at flying large areas, for example, with wider line space data, which can exclude areas where we can actually demonstrate that there is no salt in these areas, versus the areas where we do have known salt.

So it comes down to the submission we made on the cost of acquisition of the airborne geophysics, for example, where we are saying that if you fly at one- or two-kilometre line spacing you can go down to the same sorts of relative costs as radiometrics and magnetics, at the sub-dollar per hectare costs for acquisition, and that gives you the big picture. It gives you the big picture of where the salt may be, where the salt may not be, and it allows you also to go back in time to fly more focused, targeted studies if that is warranted. There is no reflection of that in the actual document.

Greg Chapman – Just in response to that, I would like to disagree a little bit. Just because there is no salt in a particular part of the landscape doesn't mean it does not need to be managed for salinity. In other words, if you have an area which is a recharge zone but it doesn't have any salt, it can still be affecting water table levels and regional water table pressures in areas where there are significant salt stores. So I think we have to be pretty careful of that.

Phil McFadden – I think you have a good point there. Just because you don't have a salinity problem doesn't mean you are bereft of responsibility and that sort of thing. But I think it does change things for people. There was some discussion earlier on – I think you were talking about it, Stefan, as your view – that salinity problems had been oversold to some extent. It might be worth examining that to some extent. But there is a level of fear out there, and if you can say to somebody, 'Just at the moment you yourself are not affected, but you should be putting in place these management practices instead of those management practices,' I think it would be enormously helpful for people.

Peter Woodgate – I think Ken's point was about multiphase and multistage sampling – a smart approach to answering the questions you have got at a number of different levels. We haven't gone to that level of detail. I think we should, and we will.

Brian Spies – I think there is a mention about being able to fly at different scales and changing the costs; we need to make that much more obvious. But to come back to your point [Greg Chapman?], I think the difference relates to what I was hearing Ken saying about doing this airborne and flying to map salinity. You are saying you can map other things which are important to managing salinity; you may not be wanting to map salinity itself. There may be areas with no salinity but it is very important in salinity management to know what is happening in those areas.

Phil Tickle – (Raytheon Australia Pty Ltd)This is just a comment in terms of multistage sampling. Wouldn't a better alternative be to have a very comprehensive stream gauging network to give you a better idea of where the salt stores might be, rather than flying large areas of broadly spaced AEM, for example?

I think that brings us to the importance of people understanding scales. At the moment, through the document we are talking about regional scale, local scale, farm scale operations, and using different technologies. I think for a user it could get quite confusing, in terms of how you can move up and down through those scales. Some of those systems are applicable at multiple scales and I think it would be useful to try and provide here a little bit of a decision support system, if you like, in terms of how the user should move up and down through those scales, and the technologies that are available to do that at multiple scales. It might be the same technology, just at a different intensity or resolution, for example.

Mike Lee – As I said before, I like the broadening of the discussion about AEM to see below the surface, not just to see salt stores and what have you. It seems to me pretty important that, while deep percolation might be an issue and we may be going to a great deal of trouble to get people to change practices, if there are some structural impediments to that recharge getting to the salt, which there appears to be in relation to older plantings in Honeysuckle Creek, we have really wasted a lot of effort. So it is not only seeing the salt stores, it is seeing the mechanisms at work and the barriers to the water meeting the salt.

Greg Street – I have an issue with the regional surveys versus the local, I guess, where there is government decision making and government policy for regional surveys versus landholder for local surveys. I would imagine that most farmers would not be able to implement much on a regional survey, as I see it.

I think also you need to think in terms of the use-by date of these surveys. It needs to be strongly emphasised in the report that these data sets will have a long use time. For instance, I am using radiometric surveys flown in 1979. They are not that good, but hell, they're better than nothing.

Ken Lawrie – While we are on the regional surveys, I just want to emphasise the potential for tension that arises from this. Effectively, do we go down a model where we have the CMAs, quite rightly, involved in designing and implementing plans of work? The problem is that they often, as Greg was saying, have a remit for 10 or five very small areas. If you do the costing of these surveys within that CMA's district, you find that you get coverage of maybe some one-twentieth of the catchment, for the same cost as if you do the broad line spaced acquisition across a couple of catchments or whatever. It is a question of perhaps looking at the people who are purchasing, and needing to have some mechanism there to have some strategic input into that decision making. I don't know if that a role in the Murray-Darling Basin for the commission, or whether it is for the NRM agencies, but it is to have that broad strategic input to assist the CMAs in working together to maximise the potential benefits from these sorts of surveys.

Andy Green – I am an even more extreme case than Ken in the area of taking out line spacing. I think, at least with AEM surveys, we should stop just thinking about these surveys as surveys on regular grids, necessarily. If necessary, think of them as one flight line per drainage basin. In other words, just think of them as a replacement for drilling to sample water tables, for a hydrological model. The depth penetration is getting much better now; the inversions are getting more reliable. You don't actually have to have a whole lot of data lined up side by side. If what you want to do is to constrain a hydrological model, then why not put one flight line across the catchment to give you data to populate a model? There are lots of other ways of thinking about these data sets than the way we have, that has transplanted them from mineral exploration surveys.

Colin Pain – I would like to make just a very quick point. We are not flying blind in all this. We can actually look at the scale of the landscape and we can make decisions about what kind of survey spacing is appropriate on the basis of what we already know from our information that is already available, as Ian pointed out this morning.

Phil McFadden – I think that's a good point. Previously we have just taken it over from the minerals process and we have gone in and given it everything, but now we have got enough information that often we can make sensible choices about costs and that sort of stuff.

Ian Lambert – Just briefly: as we go to the more regional studies, the issue arises of government-good science versus science that should be paid for by an individual catchment out of allocations it gets through the NAP. I think we are entering very difficult territory there, because there is not an awful lot of certain funding around for governments to conduct geophysical surveys. I don't know – maybe Mike Lee could comment – whether there is any opportunity within the NAP framework still for key regional surveys that cut across different catchments, that really are enhancing our understanding of regional systems but not specifically designed for an individual catchment that might have to fund it.

Mike Lee – Under the National Action Plan there have been funds put aside in Queensland for a substantial body of work conducted at state level on salinity mapping and groundwater investigations. Some of the people in this room are involved directly in that. We will be talking again with New South Wales to coordinate some effort under the new arrangements for some salinity mapping and related work in New South Wales. So there is some potential, but if we move beyond the regional level we are really relying heavily on the states to provide some leadership to draw this together in a coherent way for the particular state. I know it has been happening in Victoria, but as I understand it we are yet to see a program emerge. There may be others in the room who are able to comment on a state level.

Greg Street – I guess the precedent is that there have been large-scale geophysics surveys flown pretty much in every state for mineral exploration, and there seems to be this dilemma of whether they should be flown for private good or for natural resources management. I really can't see the difference.

Ian Rae – That is a perfect opening, thanks, Greg. When I read the report I had my environmentalist hat on, and it twitched every time I came to the word 'assets', because I don't think that was treated very well. Maybe this goes close to the edge of the terms of reference for Peter and Brian's report, but there seemed to me to be three sorts of asset in question. One was the farming land itself and its fertility or fecundity. A second one was a group of services, like roads or people who want to drink the river downstream. And the third one was biodiversity, which did get specifically mentioned but was just lumped in with the others. I think they are very different. They have very different users. So I looked back into the questions that it was suggested different groups of users might ask.

I think the biodiversity one, in particular, is unlikely to come up for an individual farmer. I'm not suggesting that farmers are red-necked and ignorant about these things, but their main thrust is to make sure that their land works and that they comply with regulations about contaminating other people's land or rivers. They are not really big on biodiversity, in the main part. But that is the sort of thing you are likely to get from an environmental representative in a community or maybe on a catchment management board.

Certainly on a bigger scale that is the sort of thing governments get interested in, and Stefan talked about feeding that in to the economic models and postulating a purchase of that good on behalf of the taxpayer. (I think those were the words he used.) But I think that is a difficult issue. Maybe it doesn't need to be handled in the report; maybe it ought to be more in the User Friendly Guide. Can I perhaps get some comment from the authors, and maybe from the panel, about that, Phil?

Phil McFadden – Yes, certainly. It was a plea that John Lovering made earlier on in the day, that in going through this whole issue biodiversity was a critical aspect for the nation. Have you got any comments there [Peter and Brian]?

Brian Spies – There were a couple of submissions that emphasised that and said this is something that is often overlooked. We have mentioned it in a couple of places. You [Ian Rae] are saying you don't think it is strong enough or it is not clearly tied in to other aspects. Perhaps we can hear from the panellists as well on that.

Stefan Hajkowicz – One of my main points was thinking through the purpose of mapping salinity, which is really about investments and where we target them. Purchasing environmental services or biodiversity is one of those investments that it is so hard to put dollars on but still it is something that we need the maps to inform. I don't know if that comment helps very much. Working out these maps in such a way that they can inform investment decisions is a good thing to do.

Glen Walker – I was probably one of the people that used the word, and I think in a lot of it when we talk about salinity management we include biodiversity in terms of the asset and then start thinking in terms of public and private good in that sense. So in terms of thinking through salinity management as to whether you are aiming in terms of asset protection and whether that may be more localised actions versus diffuse actions, it really comes down to what is viable and not. So certainly in the way that we have used it, and I think in terms of the text, you would include it, I would have thought.

Phil Dyson – I guess if you put biodiversity and the potential for geophysics together and ask that question, I come back to the same thing: I am looking for how geophysics might be able to inform my water balance and understanding of what is happening beneath the ground, in terms of the impact on biodiversity. So again it comes back to thinking in that same framework. Have I got an aquifer that is sitting underneath that thing that is trying to discharge back to the surface and I don't know about it? Can I drill that thing and find out if it is there? If I do find it with a drill rod, basically is there a geophysical technique that will allow me to locate that spatially within the broader landscape? And can I use that information then to inform my understanding of the water balance and the hydraulics that drive it? That is generally the way most of us would think through those issues.

Ian Rae – Are you suggesting that there is a fourth class of user, then, in addition to the three?

Colin Pain – I would have thought basically the geophysical data was providing background information – geoscience, regoliths, soils information – which are critical layers when we come to consider biodiversity. So I would see these techniques as having direct application, through an interpretation process that may be very similar to the one we would use for mapping salinity.

Phil McFadden – I wonder if we are actually missing talking about the issue that you really raised, Ian. To some extent the question is: if we perceive that protection of biodiversity is an important issue for the nation, and salinity can affect biodiversity, then who is the user out there who is going to be asking the appropriate set of questions about how he manages the salinity from a biodiversity point of view? I think this is where really where you are coming from, isn't it?

Ian Rae – I think it is, yes. I was slowly refining it as Colin got to it!

Phil McFadden – So, really, from the point of view of the document here I guess the question is: how do you feed that concept in to the document so that all the people out there who are asking the questions about salinity have the biodiversity question in their mind at the time when they are asking the questions, and take that into account and recognise that salinity impacts upon that quite significantly and that they need to consider that as well when they are making their choices? Is that really where you are coming from?

Ian Rae – Yes, exactly, Phil. Sue [Serjeantson] has just pointed out that even a case study or two along those lines would help to indicate that we are being catholic about this.

Brian Spies – There were two submissions from the National Parks Service which talked about biodiversity – very detailed surface studies. We will have to go back and see what mapping techniques, if any, they used besides looking at thousands of different types of plants. As far as the users are concerned, we have not identified the different classes of stakeholders. We talked about the scale of the user, and perhaps people at national park level would be on the larger scale, the catchment or community size scale. I can see that maps into there somewhere.

Ian Rae – When I started to think about this, I certainly saw that hierarchy, that as you went up you would tend to get people who were not so closely connected with their own – quite proper – vested interest but took a broader view of things.

Glen Walker – Within the South Australian survey, biodiversity issues in the Riverland were specifically targeted as part of that, and there were specific activities within the overall program within that.

Ian Rae – That's right. Did you see the latest press about it, Glen, that one of the towns has got a terribly smell from the swamp, and among the fixes they are thinking of is to get rid of the swamp! Ho hum.

Glen Walker – I won't go into that.

Andy Green – Not knowing very much about biodiversity I can only observe from examples where the issues have come up. It seems to me that most of the time the biodiversity issue is raised when there is a particular area that is threatened – in other words, it is a fairly restricted geographic area or it is something like a river. Maybe I am wrong. If a plant species is going to disappear from northern New South Wales as a whole, as a part of salinity, maybe so. But it seems to me most of the time it is a Ramsar wetland or it is some fish in a river, or something like that. And there the management problems are very similar to what they are for roads and bridges and towns: you have just got to stop that happening in that spot. And that I think is part of the reason why they are classed jointly and thought of jointly as assets.

Baden Williams – In the Holbrook Landcare Group, which is a pretty big group and has attracted probably getting on for $10 million of funding over the years, 80% of that group listed trees and birds as the number one priority for their landscape. I couldn't believe it, but they did. Now, either they are real smart and know that funding agencies have a strong bias towards planting trees and counting birds, or else they genuinely couldn't care less about their agricultural land and are happy to see it revert to trees and birds. But that is just one biodiversity story.

I would like to come back later, when this subject finishes, to salinity mapping being a surrogate mapping of recharge/discharge, because I think that would perhaps be even more attractive to users down the line.

Robin Greenwood-Smith – I think the biodiversity or any of the other environmental-like topics are going to be addressed by the stakeholders, who will set a requirement for a survey or an analysis that requires a particular level of performance, like how many Murray cod per cubic mile or whatever in the Murray, or what sort of trees are going to have to survive despite salinity. That will set the scene for: where do we do surveys, what sorts of surveys are required? This may be outside the farming areas and so forth. It is really not a matter for us, as the people who are working on surveys and at the front end, I suppose, of the process, to make any decisions about it. It is simply that if the community has a requirement, then we will have to find a way of measuring or satisfying that requirement.

Therefore, on the economics side – and this is what actually drove me to this, this morning – as soon as you introduce an intangible such as biodiversity, which it tends to be, then you can't do benefit:cost analyses, because the benefit is undefined. You have got to find some other way of ordering the way in which you put money into projects.

As far as the report goes, it is probably worth while just pointing that biodiversity is one of the variables that do have to be considered in planning a survey.

Phil McFadden – I guess, though, I would view it from a slightly different perspective. It is important for us to start learning how to figure out how to value that biodiversity, because if we simply say we will leave the biodiversity out because we don't know how to value it, then it gives the economists the opportunity of saying that in fact it is a poor cost benefit ratio and so we shouldn't do that. If we have left out something that we know is critical to us, then it is a real problem.

I think you [Ian Rae] have stirred up a point here that is going to add some strength to the document. After Mike makes his point I think we should bring this portion of it to a close and back off to what Baden was picking on.

Mike Lee – Under this regional delivery model that all the governments across Australia have signed on to, the regional bodies have to produce a holistic natural resource management plan which addresses biodiversity and other natural resource related issues. Biodiversity is fair and square in the target for National Action Plan funding, to the extent that salinity and water quality issues are threatening processes for biodiversity. That is why this discussion about assets comes up, because salinity is a threatening process for civic assets, for productive assets and for biodiversity assets.

It works the other way from what you have suggested, Phil. Because it is difficult to cost and value biodiversity assets, and because their values are conceptually so high, I think we will find increasingly that the protection of biodiversity assets is going to drive a lot of the direction of salinity funding, and already is doing so. One of the reasons again is that it does come down to specific and totemic assets that really need protecting, like Toolibin Lake in Western Australia. And that is not a bad thing.

So I wouldn't like people to go away thinking that biodiversity is not part of the story here. But the discussion is not really about mapping biodiversity, it is mapping to reduce the threat of salinity to biodiversity assets.

Andy Green – My comment is almost entirely cynical about valuing biodiversity assets. In reality, the only practical measure of the value of biodiversity assets is what a government will pay to have them protected, in the face of public opinion. It doesn't seem to me it is going to be any different from that. How you get that into your model I don't know, but that's the practical result.

Brian Spies – I am not sure how many people are aware that the Australian Bureau of Statistics does publish shadow environment reports along the UN guidelines. This is a publication they started testing last year, and rather than the GNP of the country it looks at the total environmental accounting of the country in terms of assets being added and degraded over time, and looking at that figure. So they are publishing that to see what the public reaction is, and we might reference that in the report just to give a pointer to it.

Baden Williams – I think the report made it pretty clear that most of the salinity mapping techniques are surrogate measures of salinity, but I don't see anywhere a statement of why to map salinity – unless perhaps the biodiversity; I could see that would be one thing. But I don't see that posed up the front: 'Why are we out mapping salinity?' And again I think it is because we have got caught up with the media hype of something that can be very spectacular in its expression, but really that is just a discharge point for groundwater.

What all of these techniques are doing is using salinity as an indicator of another property – the electrical methods are measuring electrical conductivity. Now, you can interpret that in a number of ways, salinity being one of them. Or you can interpret it in terms of relative recharge/discharge, which I personally think is a topic which is far more likely to attract interest from catchment management authorities and the landholder community than just talking about salinity per se. So perhaps there should be a sentence or two in the report, up the front, saying, 'Why measure salinity?' I think someone in the presentations had that question up.

I don't know what other people think, but we have been using these electromag methods now for nearly 25 years and it is really only in recent years that we have come to realise – we've had the data sitting in front of us for donkey's years – that what we are measuring is the holeyness of the land surface, for water moving up and down the profile. I would love to see this report, which will get a wide distribution, make some point like that so that people realise that they are talking about water, largely, and that salt is there as an indicator.

Paul Lamble – This is just a general comment. We have talked a lot about acquiring data, and different systems for doing so. Surely we also need to consider, when we are looking at trying to get people to make such a big investment, that we need to facilitate arrangements for data sharing and ensuring data quality so that that investment gets the maximum return over multiple time frames and through multiple user groups.

Phil McFadden – Yes, sure – in other words, as the phrase goes, 'Think global, act local.' When you are doing this sort of thing, think about whether you can hook up with other people and get a better bang for the buck through the process.

Greg Street – That was a good introduction, really, because I wanted to talk about availability of data. We seem to have drifted in this discussion towards airborne electromagnetics more than anything else, and we seem to have focused on that. But the availability of the other data sets, which are also critical in this process, varies from state to state, and the cost also varies, as does quality. We need those basic data sets – I am talking about starting from straight maps, geology maps, topographic maps – and we expect to buy those. Air photos, if we are buying stereo pairs, we expect to buy those too.

But then there are a lot of government departments or catchments that seem to be able to get access to this data free through government agencies, and there is quite a double standard there where you have almost got government agencies competing in the contracting area with consultants, and it is a very grey area. Geophysics in Victoria and South Australia, you only have to ring up and they will send you all the CDs of the magnetics and radiometrics. There is very little magnetics and radiometrics that is usable that is owned in the public domain in Western Australia and some of the other states.

When we have got surveys that have flown for natural resource management, they are almost unavailable; I touched on this before. They are quite protected and not allowed out, although there are some now resident in GA.

This availability of data sets is very important if we are going to deal with these issues, and I think it should be in the report.

Phil Dyson – I think there are a few things there that you have said, Greg, that actually I have been trying to sort out in my mind as well. The comment I want to make is that, from my experience, catchment management authorities don't actually do their own analysis. They use a range of people to provide them with very technical advice. In some instances that is people in government departments, in some instances it is consultants. Nearly all of them have technical support groups to actually advise them. And again, in terms of the supply chain or the food chain I talked about this morning, I'm not sure that you really want to focus this document around individuals sitting within a catchment management authority. I think you really want to focus it around the people that are advising them, and those people are fairly informed in terms of their technical knowledge.

The other thing I want to pick up on is that you need to think through also and be very aware of the comments that Greg made about the interface between the broader community and the technical knowledge, and how that is likely to be handled in the future. Who is going to actually deliver this information that we are collecting?

Brian Spies – I would make a quick observation here. A number of people have touched upon the availability of previous data. I know the states and federal government make public domain all their mineral exploration data that is either gathered by the states or the federal government or by private industry, and this is designed to go back for people to use it for free in order to promote further exploration in those states. This sounds very, very analogous to, 'Why not have the same policy in place for environmental management, not just mineral and resource exploration?' It sounds very, very analogous. It has got to do with legislation and reasons behind it, and that is really outside the scope of this study. But it seems to me, listening to what people are saying, that the analogy is there and policy people could perhaps take this up at some stage.

Phil Tickle – I would just like to make a comment in terms of who the real customer is for this document. I have spent a lot of time in the couple of years travelling around the regions, and as Mike correctly pointed out a while ago, every region is meant to have a catchment strategy that they produce to secure large-scale funding. But at the end of the day, every state is implementing a property planning system, which is where this money ultimately gets spent. In some cases those property vegetation management plans are going to be legislated as, effectively, plans, legally binding documents, that have time frames of upwards of 15 years in the case of New South Wales – I think that is what they are talking about – where someone wants to make an investment on the ground and be left alone to get on with their job.

That raises some real issues in terms of the salinity risk mapping that is being used and how the investment gets put on the ground. And, coming back to Mike's comments earlier on, the issue of identifying where salinity is not a risk is a critical one.

The other aspect, then, is how these sorts of technologies can be implemented in the property planning process, because it is the vehicle that is being used to spend the money in the future. I think at the moment there is equal weight being put on farm plans and local/regional scale mapping. Most of the AEM work and other work to date has been focused on regional scale mapping, but at the end of the day the real pressure at the moment is on property planning, and every catchment that I am aware of is putting up to hundreds of property plans together per year, where they are funding on-ground works. And it doesn't really get touched on in this document.

Richard Price (National Dryland Salinity Program) – I guess we spent not much time on communication, and where we have dealt with issues of communication it has been about who the user is and a little bit on the adviser. But I have in the National Dryland Salinity Program an excellent communication team that keep throwing back at me, 'Who is the promoter?' – which is a totally different way of looking at some of these issues.

Before I make a couple of comments on promotion of the document, I would preface the comments by saying that I do realise that science moves forward through disagreement and not consensus, and I don't expect that coming out of the process of debate today or the wider peer review we are going to get total consensus on the document. In fact, I think we are hearing that ourselves.

But in terms of promotion of the document, I just want to ask the question – it is a rhetorical question to some extent – about whose document it is. One definition might be that this is Brian and Peter's document, and it is the result of a project in a point of time. If that is the case, then I've got a communication team of 12 that can go and flog it in a particular way. We know that the client is the Commonwealth, and it was first initiated as a project by Environment Australia and AFFA. If it is their document, if it is their view that they are pushing, that they consider to be a fairly objective view, having gone through the peer process, then I've got a team of NDSP communication people, plus those associated with the NAP, that can promote it. It would be promoted in a very, very different way.

Or, because it has gone through a peer review process and has in fact involved the Academy of Science, is this in fact a document of the scientific fraternity? Given the comments that I made earlier, it could still potentially be viewed as a document of the scientific fraternity, and if it was, it would be promoted yet again in a very different way and it would be promoted by not just the NDSP people, not just those associated with the National Action Plan, but I guess by everyone in this room as well – or maybe not everybody in this room.

If that is the case, are we likely to expect a document that people might stand behind and not take too many pot shots at it? That would kill it, no matter which way we promote it. It is a rhetorical question, but if anyone wants to put their hand up and say, 'Yes, I'll be happy to promote it,' please let me know, because it is going to make my life easier, running the NDSP communications year, in taking this document forward.

Simon Veitch – Richard, I take on board your comment about the opportunity with the National Dryland Salinity Program, and of course that would be an excellent opportunity to communicate this report more widely. But I have to correct a point of detail – that is, who owns this document. It has been sponsored by most states and the Australian government, so it is for the Natural Resource Management Standing Committee, under the Ministerial Council. They are doing that with an appreciation that it is needed to assist regional planning, investment decisions, whether that be at the Australian government level, state government level, or in regions for the catchment management planning, and ultimately, of course, down to the farm and paddock level for individuals to invest as well.

So it has got a wide ownership, but I suppose at the peak level the NRM Standing Committee and the Ministerial Council...[end of tape]

Richard Price – I accept that, Simon, but I am using the word 'ownership' in a very different sense – not ownership of the document per se but ownership of the content of the document.

Peter Woodgate – First let me echo Brian's comments in relation to thanking you all. It has been very professionally run today, and there is a whole range of comments which have come back which will demonstrably improve the quality of the documents and their usability.

My closing thought is really picking up Richard's comment: who is going to own the content? It could be another one of those rhetorical questions, but we could pose it to the panel. Would they be prepared to own the science of the documents, or would they rather leave as neutral observers?

Peter Woodgate – I like the idea of going through each of them.

Phil McFadden – How do you guys feel about that?

Andy Green – I doubt that I want to be part of an ownership process, not being part of an organisation.

Colin Pain – I'd be reasonably happy with being associated with it.

Phil Dyson – With modification.

Peter Woodgate – Was it your own work you wanted to modify, or somebody else's?

Phil Dyson – I think that there are some things in there that I would like to see changed a little bit – altered, modified, whatever. I'm not at the stage yet where I could give you that commitment, sorry.

Peter Woodgate – We'll pick up your comments – you've conveyed them to us, or will – and on that basis then you might be able to?

Phil Dyson – Sure.

Glen Walker – I am pretty much the same.

Richard Lane – I hold my powder dry until I see the next draft.

Tim Munday – Snap!

Bruce Dickson – I'm happy with the section that I have dealt with, yes.

Stefan Hajkowicz – This is tricky. You didn't tell us we were going to do this. Yes, I think I possibly could, but there's a hell of a lot of science I don't understand in the report, too, and I don't know if I would feel able to have ownership over any of that. Do you [Peter and Brian] have our names on it?

Phil McFadden – I think they understood where you guys are coming from. Sue Serjeantson wants to make a comment.

Sue Serjeantson – Phil, I wanted to say with respect to the role of the two Academies that our role has been to oversee a process and so the actual ownership or endorsement is not an issue for us. Our undertaking was to ensure that there was a consultative process. We did this in a two-step manner. We assured the consultants of the independence of the program for today, and that we assured wide dissemination of the information that a public forum was to be held and everybody who had an interest was invited to attend. So I hope that we have been able to do that.

The part that is worrying me a little bit at the moment is that the Academies have been asked to sign off on the User Friendly Guide. My husband is a farmer in an area affected by salinity, Boorowa, and the way we measure salinity is to walk along the Boorowa River and look at how thick the salt is on the rocks. I think most of our neighbours do exactly the same. And so to call what we have got before us at the moment a User Friendly Guide causes me some embarrassment. I think that I would like people to consider whether what we have got is really an advisers' technical guide, a very important resource which can be drawn upon by science writers, pamphlet writers and others who are perhaps more experienced in communicating with the farming environment.

Phil McFadden – A useful comment there, I think. I will draw this to a close now. I would like to thank everybody involved, I would like to thank the speakers very much – that was absolutely fantastic stuff you gave us. Obviously, I would like to thank Peter and Brian. They have put a vast amount of work in over the past little while in producing these documents. They now have a lot of work on their hands. Thanks very much indeed.