The second dimension of variation is in the values of the assets that are under threat. We have come to appreciate that it is not just agricultural land or water resources but also infrastructure and environmental impacts. Within any of those categories, as well as between them, there is a wide variation in the value of the assets. In general, in the case of say, infrastructure, you can find some very large values concentrated into very small areas so potentially justifying large investments. In the case of agriculture land, which is probably at the other end of that spectrum, quite a lot of agricultural land, although it is valuable for agriculture production, does not really compete very well with some of the concentrated values that are under threat in the other categories.

The third dimension on which things vary is their responsiveness to treatment. New science that has become available in the past few years has emphasised that in different parts of the landscape, in different parts of New South Wales and other States, in some places you can go in and implement land-use changes and get a reasonably rapid response and a reasonable significant response in terms of prevention of salinity. In others, you can put the same intensity of effort in and get almost no measurable response in our lifetimes. The National Land and Water Resources Audit has categorised the ground water flow systems across Australia into about 20 different categories and then grouped them into three broad categories. I show you a little section of the map for south-west Western Australia. The pink areas show the relatively responsive bits, the blue bits are intermediate and the grey bits are very slow to respond. There is a similar mottled pattern effect across all of the States.

I have put Western Australia on the board to remind me to say that sometimes people adopt a position that I am just a Western Australian and what would I know about situations in the east? However, my arguments are not just a Western Australian position or a Western Australian curiosity. I believe that they are relevant throughout southern and eastern Australia, and they are now quite widely accepted amongst many people in all relevant states.

In the talk I will mention a few myths that have snuck into the salinity story during the past couple of decades and that we seem to be getting over now. The first myth is that increased water use of traditional agricultural practices will be sufficient to deal with salinity. There was a view that this could do the job but the science that has come out in the past few years has shown quite clearly that that will not be nearly intensive enough in its effect. That is part of the story of the new science over the past few years that the degree of intervention that needs to happen is generally high, even in the more responsive bits. Even in those pink areas the degree of land-use change that will be required to prevent salinity impacts is much higher than we have previously appreciated.

The fourth and final dimension that I want to cover is the cost of making the land-use change needed to prevent salinity. In a relatively small number of areas the cost of making the change is virtually zero. There are profitable perennial species available that farmers could implement on areas where they would have a beneficial effect on salinity. But it has become much clearer in the past few years that, especially in the low to medium rainfall zones, those options are much fewer and further between than we would need to really handle the salinity problem. This interacts with our relatively new knowledge about the intensity of treatment that we would need to put in place. If you just put in a few trees around the edges, that sort of cost could be handled. But if we are talking about revegetating 50 per cent or more of the landscape, that is a cost impost that farmers will not wear unless we are talking about commercial enterprises being put in place where salinity prevention is a by-product of that commercial activity. That ties in very close with the commercial theme that the Chair mentioned at the start.

I believe there are water efficient irrigation technologies that are available which, based on some work that the Australian Bureau of Agricultural Resource Economics published last year, look like they are not excessively costly. On the other hand, a high cost example is pumping to try to draw down ground water. That is pretty expensive no matter where it is done. It has variable impacts, depending on where you do it, but basically it is expensive and probably something that would only be reserved for high-priority areas. The point is that the cost of salinity management varies.

The next myth I want to mention is that farmers can and will change land use sufficiently with the existing land-use options and existing technologies that are available to them. They have not yet and I do not believe they are ever likely to without some radical changes in the available technologies.

If we put all that together what have we got? We have a very small number of areas where the situation is favourable for salinity management on all four of those dimensions. It may be possible to identify a few cases where you have a high level of threat from salinity, high values at risk, relatively high responsiveness to management and relatively low cost. Those situations are obviously the top priorities for investment. Then there is a group of areas which have positive outcomes for maybe two or three of those dimensions which would be the moderate priority areas. But for most of the agricultural land, most of the rural landscape that needs to be treated if we were to completely eliminate salinity as a threat, they do not rate at all high in priority compared to those top priority areas that are favourable on several of those dimensions.

This brings us to the issue of prioritisation and to a recognition, given the little bit of background that I have already given you, that we cannot afford to buy a comprehensive solution to salinity—even if we wanted to it probably would not be a good investment, but we will put that to the side. The simple truth of the matter is that with current budget availability we cannot even go close to buying a comprehensive solution to salinity so we really are forced into choosing some winners and losers. From the brief evidence I have already put to you it is my firm belief that to get the best bang for the buck in the salinity budget it would require us to focus the public dollars fairly tightly into some very high priority areas or else in ways that get high leverage, and I will talk more about that later on but for now I will talk about prioritising these high-priority areas. I recognise the reality that some catchments and some sub-catchments warrant relatively few dollars in terms of direct financial support. This points to the need for some sort of investment framework to try to make these decisions.

The next myth is that sharing the salinity budget around evenly across the agricultural landscape is in some way fair. This is one comment that comes back when it is said that not everyone will get an equal slice of the pie: "but what about fairness?" Fairness is a complicated concept and what is fair in some criteria is unfair by others. Sharing it around evenly does not sound particularly fair to taxpayers in that it would result in their taxes being spent inefficiently and relatively ineffectively. In many cases it is also not fair to the land-holders who are being tricked into putting their own time and resources into activities which are certainly not in their own best interests, and in some cases not even in the interests of the broader community. So fairness is not just a simple matter of spreading the money thinly.

The next issue on which I wish to touch is integrated catchment management which is a bit like motherhood: you have got to be in favour of it. In principle, there are plenty of good things behind the idea of integrated catchment management but in practice it seems to cause people to do strange things and to think in strange ways. What has crept into the integrated catchment management concept is a set of additional ideas and assumptions that do not bear up. I want to break it down into its three parts: integrated, catchment and management or planning. The "integrated" idea is fine. The "catchment" word causes problems. I want to focus on the idea of the catchment scale intervention. Many people seem to interpret the reasonable need for management of salinity under an integrated catchment management approach as implying that interventions need to be at the catchment scale, whereas the evidence is quite strong and clear that in many situations the most important and effective interventions are at a local scale, not at a catchment scale. I will give some examples in a minute but, for example, it applies to a fair number of farming situations, to a number of country towns and to some environmental situations. It does not always have to be done at a catchment scale. By focusing only on catchment-scale interventions you may be diverting funds into relatively low priority uses.

Now "management". Usually when people talk about integrated catchment management they are really talking about integrated catchment planning. Sometimes it is mapping rather than planning, but let us say it is planning. The problem with the planning part is that it does not get you adoption. Often there seems to be an inadequate appreciation that drawing up a set of plans does not get you to the final outcome of achieving change on the ground. There is a lack of recognition of the need for an incentive of some sort to drive that change on the ground.

I will show you some examples from Western Australia. This slide is an example of where integrated catchment management will, if you do it properly, lead you to undertake quite localised intervention. Lake Toolibin on the bottom left-hand side is the last freshwater lake left in the Western Australian wheatbelt. You will notice that in Western Australia we are not so fussy as to require water in our lakes! But, it does have water in it at times and, when it does, it is fresh. There were a number of other freshwater lakes in the agricultural region of Western Australia but they have all gone saline, and the Western Australian Government has spent over $1 million trying to keep this lake fresh.

On the advice of hydrologists, a whole range of interventions has been undertaken, but the really critical ones are the two that are labelled there. There is a network of bores sprinkled through the floor of the lake. They are pumping out groundwater which is rising in a palaeochannel underneath the lake and at risk of intruding into the lake. And there is a diversion channel built around one edge of the lake to capture surface flows coming off the catchment and send them down the line to Lake Tarbalin which you can see in the background, in order to prevent salt water run-off from entering Lake Toolibin. You will see also in that picture a bit of revegetation in the catchment above Lake Toolibin. I think that is probably oil mallees. They are making a contribution, but they are the second string in the bow of this strategy. There is an environmental example where the treatment needs to the local.

The same story applies to some of the country towns. In Western Australia we have about 50 towns that are under threat from salinity. About 30 of them have had detailed hydrological analyses done and for six of them detailed reports have been published. Strategies have been recommended by the engineers and hydrologists who have studied the situation in the six towns. If you look at the central column of the table on the slide, the strategies for the six towns are: pump, reduced recharge within the town, pump, pump, reduced recharge within the town and pump. If you asked the question,  "Do trees in the surrounding catchments have an important contribution to make to the protection of the infrastructure within these towns?", the answer is: no, no, unlikely, no, no and no.

These are examples. I am not saying this is a universal outcome. All I am doing is painting you an example of a quite common situation where catchment-scale intervention—if you interpreted it as meaning you needed to do things throughout the catchment—may well cause you to do things which are not in the best interests of the asset you are trying to protect.

If not integrated catchment management, then what? The problem as I see it is that ICM encourages people to put the problem backwards. It makes people assume that you should go for catchment scale change, whereas what we ought to be doing is starting with the assets that you want to protect and working from there; analyse the best methods of protecting those assets. It may be local, catchment scale or some combination.

We also need to consider the "living with" option because in some situations that is the best option. And we need to compare the priority of the works in that catchment with those in other catchments. That implies prioritising at a level above the catchment, not just at the catchment scale. The final step is that we need to concentrate the funds so that we can create the incentive to get the changes that we need to happen, or provide the resources to actually implement the changes.

I mentioned the "living with" option. The living with salinity option certainly applies at that the agricultural level, as well, but here are a couple of examples from the non-agricultural area.

Water resources is one. That is a picture of a reverse osmosis desalination plant. I believe that desalination is going to become a serious option for us in Australia as a source of freshwater in years to come. Desalination costs have been falling over the last 20 years as the technology has improved. Within the next 10 or 20 years it is going to be competitive with traditional water sources. It is an option that may encourage us to look at living with salinity and then desalinating the saline or brackish water. Desalination of brackish water is already close to being competitive with traditional water sources.

The other one I will mention is built infrastructure. In some situations it is far more efficient and effective to allow the salinity to occur and then repair the damage. That sounds pretty unattractive but the numbers show that the difference in outcomes and the cost of outcomes is quite dramatic in some situations. Not always. I am painting examples here, I am not drawing generalisations. The bottom line there says that the cost of repair to salinity in the Merredin town site in the wheat belt of Western Australia over the next 60 years has a net present value of about $400,000. The cost of preventing that damage is between five and 10 times that much.

I have talked about the need to prioritise and the need to target resources, public funding, into relatively high priority areas. It raises the question: What do you do for the rest if they are not going to get a slice of the main pie? My recommendation is that we need to put a significant chunk of the resources into ways that will get leverage across very large areas. The way I see that happening is through the development and promotion of technologies for salinity prevention. That is one element of it which allows you to get that leverage. That is a picture of some oil mallees that you will hear about later in the day from John Bartle.

The other element needed by the majority is developing and promoting options for living with salinity, for the cases where prevention is impossible—in some cases it is just physically impossible, but it also may be economically unviable in some cases. In Western Australia we have many farmers who are already actively living with salinity, they often do not have a choice, and they are doing things like growing salt-tolerant plants, as you will see on the left; or there is a new trend to establishment of deep drains in salt-affected areas, which you can see in the overhead on the right.

Which technology should we be developing? I cannot give you a simple answer to that because the reality is that we need a wide diversity of technologies. We need every tool in the toolbox to be available to throw at the problem because the diversity of situations is so wide. The key thing is that they need to be profitable if they are going to be adopted on a big enough scale. The existing suite of options is clearly inadequate. As I mentioned earlier, there are only a small number of options which are already profitable. We need different methods to suit different situations and different types of problems. To paint that picture of how we need that diversity of methods, this is a fairly crude lumping together of plant and engineering options to show how they vary in their relevance to different types of problems.

The left-hand column shows the different types of assets that we want to protect: Agricultural land, water resources, biodiversity, physical infrastructure or flood prevention and sub-categories of those. The other columns show: two dots, that particular approach is highly relevant; one dot, it is somewhat relevant; a hollow dot, it is a little bit relevant—that is, a supporting role; and a question mark, I think we need to do more work. I would not want to have an argument about these dots, there is plenty of scope for considering what they should be, but the real point of the exercise is to show that different approaches have different roles for different types of problems. Some approaches do not apply at all well to some of the problems.

That brings me to the last area, which is policy. How do we get a policy system in place that will encourage the sort of prioritisation and attention to the commercial drivers that I believe we need if we are really to be successful in this area? I am going to try to clarify this issue by breaking down the types of policy into three broad groups, which are quite distinct from each other.

The first is policy instruments to encourage change on private land. I am applying that to a pretty broad range of policy activities. Within that category I would include: market-based instruments, which has been emphasised in the National Action Plan; small-scale subsidies, which is the heart of the Natural Heritage Trust; extension; information; communication; groups — that is, the Landcare approach; or regulation, which is promoted by some people. They are all methods of trying to push land-holders to change their existing land use.

The second category is fully-funded direct works or direct action of some sort taken by governments. That would include the sort of activity you saw in the picture of Lake Toolibin. It is actually public land where those pumps and that drain are being put. There is no need to encourage the surrounding farmers to do it; the government just goes in and funds it.

The third category is the technology development and industry development that I alluded to in the last couple of slides, where you are not directly funding action, you are funding activities that will create leverage for action down the track.

In a similar spirit to the last table this one shows that those different types of policy approaches have different relevance to different types of problems, again including agricultural land, water resources, biodiversity, physical infrastructure or flood prevention. The diversity of relevance is even greater than I showed you in the last table. You will see, for example, that by my judgment technology development is the primary, almost the only, policy tool that is relevant to the protection of the broad scale agricultural land that is not closely associated with some other high priority assets. Whereas, for something like water resources, the full range of policy instruments is relevant. For something like biodiversity hotspots, technology development probably does not have a major role to play. Direct works are, for environmental hot-spots like Lake Toolibin, very important, and there is probably a role for policy instruments as well.

Again, I would not go to the wall defending these ratings. I think that they are pretty close to being right, but they are my subjective judgment. The real point is that we need to get the thinking about this right because, at the moment, the great majority of policy effort and the policy approaches we are using in Australia fall in the first column, the policy instruments column. Of the three, the policy instruments column has the least broad spread across the range of issues that we wish to tackle. That column indicates that policy instruments are only important for four of the 10 categories of salinity impact that I have put up there. I think there is a real, pressing need to diversify the funding into the other two types of policy approach. We are putting too many eggs into that one basket.

I will finish with some recommendations and hope there will be time for a few questions. You can see that this story, this set of information, poses some pretty serious challenges to the existing policy paradigms that we have in place in Australia. I have four slides of recommendations that result.

First, we need to adopt a framework to assess and target salinity investments. We need to reverse the planning approach so that it is asset-based and not catchment-based.

We need to prioritise at State and national scale, not only at catchment scale because some of the decisions about direct investment, industry development and technology development, need to be prioritised at a higher scale than the catchment scale.

The corollary of that is quite a strong one. It means that we need to modify the current role and the intended role of catchment planning groups, such as catchment management authorities.

We need to allow time and resources for the analysis of the options. This is a very complicated area and I am quite concerned that, because of the delays that have occurred in getting the National Action Plan up and running, we are going to rush into accreditation of plans. It would be very easy for us to accredit plans under pressure to get things done which are not properly developed.

For them to be properly developed they are going to need a very high level of technical support. I do not think it is realistic to expect these catchment authorities to be able to come up with meaningfully sound plans without a quite high level of technical support.

We need to adopt targets that come from analysis, not just from our desires about the outcomes that we would like to see.

We need to embrace and take seriously options for living with salinity, not because they are always going to the best options but because they sometimes are.

We need to scale down the expectations surrounding policy instruments, including market-based instruments, which carry the burden of being flavour of the month at the moment. Of course they have a role but they are not the only game in town.

We need to recognise direct government action, such as fully-funded works or possibly purchase of water from irrigators as a legitimate part of the story.

I think we need to change the nature of the extension and communication activities, particularly in regard to salinity, so that they focus on helping farmers develop and promote properly evaluated technologies, not just technologies that they happen to have available.

We need to get serious about supporting technology development and industry development. I have suggested that something like 10 or 15 per cent of the salinity budget ought to be allocated in that area. That needs to address both the plant options and the engineering options.

The final point I would like to make is that we need to keep an open mind, and expect it to need to be changed. It certainly has needed to be changed in the last few years, as our scientific understanding of the problems has advanced very rapidly. We have reached the point where the scientific knowledge about salinity is way ahead of the policy design that we have in place.