David J. Pannell

Abstract

The roles of plant-based systems and plant-based research and development (R&D) for management of dryland salinity in southern Australia are examined. Logical policy responses to dryland salinity are proposed for protection of water resources, biodiversity, infrastructure, dispersed assets such as agricultural land, and salt-affected land. Recommended responses are sensitive to a range of bio-physical and socio-economic conditions. Extension and incentives to promote currently available perennials or salt-tolerant plants are only appropriate as the main policy response in a minority of cases. Penalties or permits to limit planting of perennials can be justified in certain areas of high-water-yielding catchments. For the majority of agricultural land that is at risk or is contributing to dryland salinity, the most logical policy response is investment in development of improved salinity management technologies, including R&D into new plant-based systems. Situations where plant-based R&D for profitable farming systems is the best option include (a) to reduce salinity impacts on water resources where groundwater systems are responsive and the dependence on fresh runoff for consumptive use is low; (b) to protect infrastructure and biodiversity where there is relatively high responsiveness of groundwater and the urgency of response is low; (c) to protect dispersed assets (e.g., agricultural land, most remnant vegetation on farms, flood risk mitigation) where profitable perennial plant options are lacking; and (d) for land that is already salt affected.

Keywords: plants, policy, research and development, dyland salinity

Several factors have contributed to changing attitudes about the roles of perennial-plant-based systems for management of dryland salinity, including:

• Changing evidence about the effectiveness of plant-based systems in preventing the impacts of salinity. In most regions, prevention of discharge requires much larger areas of perennials to be established than was believed in the 1980s and 1990s.
• Increased knowledge about the diversity of groundwater flow systems with their varying responses to plant-based salinity management systems;
• Recognition of the potential for plant-based options to have adverse impacts on fresh water flows into rivers in some situations;
• Evidence about the disappointing economic performance of most existing plant-based salinity management systems and, related to that, their insufficient current levels of adoption; and
• Increasing prominence of engineering-responses to manage dryland salinity.

These new understandings have led to calls for a greater investment in research to develop improved plant-based systems for management of dryland salinity (or "plant-based R&D") (e.g., Pannell 2001a) on the grounds that adequate adoption will require new perennial plants that are more economically attractive then the existing options. However, policy makers responsible for dryland salinity appear to be reluctant to fund R&D (plant-based or otherwise). Some policy makers consider that there has been considerable funding for salinity R&D in the past, without appreciating that it has largely been R&D of a particular type. While researchers have produced detailed knowledge of the salinity problem and its causes, there has until recently been very limited effort devoted to development of new systems and technologies to better manage it. Salinity policy programs take for granted the availability of appropriate salinity management tools, but this optimistic position is often not justified (Pannell 2001a).

Salinity policy makers are subjected to influences from a range of stakeholders with different interests and areas of expertise. Advocates for research are considered just one interest group among many, also including:

• Private land managers, particularly farmers and their representatives;
• Managers of public lands;
• The broader community;
• Catchment management organisations;
• Environmental advocates; and
• Other governments.

The groups have differing priorities, which are not always understood and/or respected by other groups. Some have multiple roles and all, to some extent, have vested interests in particular aspects of, or responses to, the salinity problem.

Given the above background, it is not surprising that policy makers have failed to develop a clear and well-founded position on the roles and limits of plant-based R&D as a response to dryland salinity. This paper is an attempt to develop such a position. Our aims are:

1. To clarify the role of plant-based systems for salinity management, and
2. To determine the appropriate roles for R&D into plant-based systems for salinity management in the context of various other possible responses by government.

We recognise that salinity is one of a number of degradation issues and that, where possible, governments and catchment management organisations seek multiple benefits in natural resource management. However, we have kept the paper tightly focused on salinity to maintain clarity.

In this section we focus on the variety of groundwater flow systems (GFSs) present across Australia and consider their implications for plant-based R&D. The responses of groundwaters to establishment of perennial vegetation depend on the gradient and the permeability or "transmissivity" of the material through which water flows (gravel, sand, clay). These vary systematically among different GFSs.

Three broad categories of GFSs (local, intermediate and regional) have been classified (Coram et al. 2000) to indicate the spatial scale over which groundwaters flow. The different flow systems indicate the distances over which changes in agricultural land use may have off-site impacts, and have been used to suggest the scales over which such changes would need to occur to achieve long-term equilibrium in groundwater levels. We acknowledge that there are some limitations to the current GFS classifications (e.g., they are not spatially explicit and the conceptualisations are over-simplified). However, as they have been widely promoted and used to help make decisions about prioritisation of salinity-related investments, they are the current ‘benchmark’ on which to base our analysis.

Time lags in the downstream response of discharge to clearing or revegetation are very different between these three broad categories of GFS (100 years or more in regional GFSs compared to 20-50 years in local GFSs, Coram et al. 2000). The response time has a major influence on the economic value of any off-site benefits. Other things being equal, the economic benefits of off-site salinity prevention resulting from establishment of perennials will be higher (because they are more rapid) in local than in regional GFSs.

In the final column we make comments about plant-based R&D needs and about the likely on-site and off-site responses. These "responses" refer to the time taken for groundwater tables to reach reach equilibrium following a land-use change or engineering intervention. For the purposes of this paper, the term "on-site" is taken to mean the land on which perennials or engineering options are established, plus immediately neighbouring land up to a distance of say 100 metres or less. "Off-site" refers to all more distant land or water.

For local GFSs the estimated on-site responses are less than 10 years for cases where transmissivities are low (e.g., GFS types 1, 7) and 10-30 years for others. Off-site response times are approximately 20 years where transmissivity is high (GFS 4, 5, 10) and 50 years where transmissivity is low (GFS types 1, 7). There is considerable potential for runoff reduction with planting of perennials on duplex soils and in steep areas within GFS types 3, 4 and to a lesser extent 5, so careful targeting of plant-based solutions is required in these regions. These situations are likely to occur more in high rainfall regions of eastern Australia than in WA or SA. In locations with high rainfall and a local GFS, only very deep rooted perennials can use sufficient of the winter rainfall excess, meaning that R&D should emphasise woody rather than herbaceous perennials.

Table 1: Likely on-site and off-site responsiveness of plant-based options in local groundwater flow systems (adapted from Coram et al. 2000)

In considering whether intervention to avert salinity impacts is justified, a key factor is the value of the assets at risk. Putting values on some salinity impacts is difficult because they include both instrumental values (e.g., financial value of water resources to be used for irrigation and town water) and non-instrumental values (e.g., non-financial value of irreversible loss of biodiversity).

For impacts on terrestrial assets, the values per hectare of land affected vary widely among the asset classes. Financial losses per hectare are greatest for impacts on infrastructure, due to the high cost of their repair or replacement. Non-financial losses per hectare are greatest for some environmental assets of outstanding significance. By comparison the potential losses (either financial or non-financial) per hectare of non-irrigated agricultural land are much smaller. The areas of land are very large, so the total value is high, but in considering investment in protection of particular pieces of land, the analysis needs to consider the values threatened on that land, not more generally. For protection of water resources, each case has a particular mixture of financial and non-financial values at stake.

The factors involved in adoption behaviour in response to natural resource management problems have been discussed by Guerin and Guerin (1994), Cary et al. (2002) and Ridley (2004). Pannell (2001b) reviewed the issues in the context of dryland salinity. At the farm level there are two main considerations:

• The perception by farmers of likely net-benefits in adopting perennial plant-based systems relative to annual systems, including any on-site salinity-related benefits.

• The capacity of farmers to incorporate new practices into the farming system. Factors affecting capacity include financial resources, labour resources and managerial resources.

The key factor affecting adoption by commercial farmers is their perceptions about on-site benefits (profitability of harvested products and the value of any local salinity benefits), which should outweigh the on-site costs (direct input costs as well as opportunity costs of activities displaced by perennials). Policy makers who make decisions about salinity intervention must additionally consider the off-site benefits (e.g., reduction in salinity) and costs (e.g., reductions in water volumes in rivers). The crucial importance of farm-level economics in adoption behaviour is underscored by studies showing that existing perennial plant-based options in southern Australia are either unprofitable or lack profitability on a scale that would generate more than localised benefits (Bathgate and Pannell 2002; Lefroy 2002; O’Connell and Young 2002; Abadi et al. 2003; Kingwell et al. 2003).

Social issues also affect adoption of perennial plant-based management options. Over 16 million hectares of Australian land is managed by farmers with an estimated value of agricultural operations of less than $22,500 (Hooper et al. 2002). These small and "lifestyle" landowners manage a significant quantity of relatively high value, potentially highly productive land, usually in areas that are close to (perhaps within one hour’s drive of) cities or larger towns that provide opportunities for employment. The states with the greatest numbers of small farms are Victoria and NSW (Barr et al. 2000), with more than a quarter of the agricultural land in Victoria being on small farms (Neil Barr, unpublished data). Many are in high rainfall areas (and therefore have the potential to influence significant amounts of freshwater runoff) in the Goulburn-Broken region and parts of the Upper Murray region in Victoria. There are also some significant areas close to regional centres within the NSW and ACT catchment areas of Murray and Murrumbidgee that have many non-commercial land managers (Albury, Wagga Wagga, Canberra).

There are likely to be a number of differences in policy approach needed to effectively influence land managers whose main income is not from agriculture. As well as having different priorities and objectives, they have limited time and resources available for making major land-use changes. One implication is that plant-based options would need to be low cost, low risk and low effort. There may need to be attention to providing incentives for land retirement, rather than only having emphasis on profitability. Beyond this, it seems possible that a different mix of incentives and penalties may be appropriate and in some cases no action may be most appropriate. Extension activities should also be targeted differently for this group of farmers (Hollier et al. 2003). Further investigation is needed to assess the components of this mix, particularly research on the likely responses of these land managers to the various policy instruments available to governments. Barr and Wilkinson (2004) discuss these issues further.

Extension. This is usually the appropriate response where perennials are already economically competitive, although in some cases where they are competitive, penalties may be warranted to discourage adoption. Extension may also be appropriate in cases where farmers generally lack information or have mis-perceptions about the salinity problem or its management. This second type of extension is not specific to particular hydrological or economic circumstances.

Incentives and penalties. The choice between incentives and penalties is somewhat arbitrary. We have assumed that, where a policy instrument is used to change current land use or common farming practice, the instrument used would provide a positive economic incentive (e.g., a form of subsidy or compensation). This is the main mechanism by which policy would be used to promote adoption of existing plant-based options that have non-agricultural benefits but which are not sufficiently attractive to farmers to be adopted spontaneously. In contrast, a policy instrument used to prevent a change away from existing land use or common farming practice (e.g., to prevent the planting of trees where their off-site costs exceed their benefits, or to prevent the installation of deep open drains where this would generate excessive downstream environmental costs), would be based on a penalty (e.g., a regulation backed with fines, or a requirement to purchase a tradable water right).

The use of incentives or penalties, as opposed to one of the other policy approaches, is considered to be appropriate where there is "market failure" due to externalities (e.g., off-site impacts on other types of assets). The criterion for market failure requires not just that there are externalities, but also that the overall benefits from changing the off-site impact would outweigh the costs (the net-benefit test). This is important for dryland salinity because analyses have found that the net benefit test fails in many locations, often because the off-site (public) benefits are outweighed by the on-site (private) costs (e.g., Pannell et al. 2001; Dawes et al. 2002; Heaney et al. 2000) or because preservation of fresh-water flows, where they occur at high levels, tends to be more important than prevention of additions to groundwater (Heaney et al. 2000; Bathgate et al. 2004). The chances of failing the net benefit test are greater where groundwater flow systems are regional or intermediate-scale (since hydrological response times are slow and required changes in land use would be large) and where the cost to land managers of changing their land use in the desired way would be high. A particular use of incentives that may be appropriate in some cases is to promote land retirement, where the costs of doing so are less than the alternatives, including do nothing.  Where incentives are deemed appropriate they are commonly used in combination with extension, where the extension is targeted towards increasing the chances of establishment and persistence of the new plant-based system. Extension in such cases would be directed towards known management issues, such as fertiliser application and grazing management.

Engineering. Expensive works such as groundwater pumping are only justified where a high value asset is under threat and it is judged that plant-based management options will not be cost-effective or will not act quickly enough to protect the asset. Two different versions of this response are included in our analysis: salt interception schemes to protect waterways (i.e. pumping of saline groundwaters that would have discharged into rivers, with disposal of effluent into evaporation basins) and localised engineering works (usually pumping) to protect terrestrial assets. We have narrowed down the circumstances where these approaches are most likely to be relevant (as explained in detail below), but have further qualified our recommendation with the phrase "if economic", since their economic performance is variable and case specific.

At the farm level, engineering is relevant to management of salinised land in some regions (deep open drains in the Upper South-East of SA and parts of the wheatbelt of WA). In cases where engineering has primarily on-site costs and benefits, it is arguably not a subject for government intervention, although some farmer groups in WA are increasingly pushing for government to provide and fund infrastructure (large arterial drainage systems) to facilitate disposal of effluent from farm drains. Complicating this issue are concerns about possible downstream impacts. There is still considerable uncertainty about the extent and significance of these. Given the impetus behind some proposals for large drainage schemes, research to quantify likely downstream impacts appears a high priority.

Plant-based R&D into profitable farming systems. We suggest plant-based R&D for profitable farming systems to be the best option where there is not likely to be market failure from externalities, and where perennials are likely to generate worthwhile salinity benefits (without compromising water yields) if the economics of their production could be turned around. Implicit in this recommendation is that there are worthwhile and as yet untapped opportunities for technology development (Ewing 2004 this conference; Pannell and Ewing 2004), or needs for investment in infrastructure or market development to support new industries based on perennial plants. Plant-based R&D can be a valuable strategy to supplement incentives in some cases. Even if it is not successful in producing economically competitive perennial land uses, such R&D has the potential to reduce the public cost of providing incentives in the medium term, by reducing the farm-level cost of converting to perennials.

Other R&D. This category includes all research other than plant-based R&D for profitable farming systems.

No action. In situations where none of the other responses can be justified, due for example to expense, low effectiveness or a low salinity threat, "no action" becomes the best response.

The recommendations that follow are based on a mixture of research results, theory, rules of thumb, assumptions, judgements and logic. The recommendations are not ‘hard-and-fast’, but provide transparent arguments as a broad guide, a basis for further debate and for reconsidering salinity policy on a more sophisticated and realistic basis.

Recommended policy responses are shown for four sets of cases: recharge areas with salinity impacts on waterways (Table 4), recharge areas with salinity impacts on relatively small scale terrestrial assets (Table 5), recharge areas with salinity impacts on dispersed assets such as agricultural land (Table 6), and salt-affected agricultural land (Table 7). The suggested responses are different for different categories of assets at risk. Research to assist with planning of policy interventions (within "Other R&D") has an overarching role and is relevant to all scenarios in all four tables.

There are four main factors driving the choice of policy approach for protection of water resources (Table 4):

• The potential input of salt from groundwaters into the waterway. This depends on recharge rates (dependent on soil texture and slope), salt stores in soil and salt concentration in groundwaters, all of which can be highly variable, even within a sub-catchment.
• The responsiveness of groundwaters in potential discharge areas to establishment of new perennial vegetation in recharge areas. "Low" responsiveness equates to intermediate and regional GFSs and "High" responsiveness represents the more rapidly responding local flow systems or perhaps on-site effects in the cases of intermediate and regional GFSs.
• Issues surrounding the importance of fresh runoff water. This includes both the level of dependence on the waterway for consumptive use and the volume of surface or near-surface flows of fresh water entering the waterway (dependent on soil texture, slope, vegetation type and rainfall).
• Farm-level economics, particularly whether existing perennial plant-based options for reducing recharge on farm-land in the sub-catchment are more or less profitable than agriculture based on annual plants.

Engineering on-site responses are suggested when the value of the threatened asset is high and the urgency of action is high. The more expensive engineering responses will only be economically justified in such cases, and even then only in a sub-set of them. Extension is recommended in all the scenarios where perennials are more profitable than annuals.

Incentives to grow existing plant-based options are only appropriate in two of the scenarios (cases 2 and 5) where the asset value is high, groundwater responsiveness is high and perennials are only slightly less profitable than annuals. In case 2, the problem is urgent, meaning that plant-based systems alone would not protect the asset in time, requiring on-site engineering works (if economically justified), potentially supplemented by incentives. If modest incentives are sufficient to achieve substantial land-use change in these sub-catchments, they may be justifiable, provided that the problem is caused by water from agricultural land rather than sources local to the threatened assets. In case 2, incentives to grow plant-based systems may allow cost-savings due to reductions in running costs of on-site engineering works in the long run. In case 5 where the situation is not urgent, analysis would be required to assess whether development of plant-based options, incentives or a mixture provided the most benefit.

R&D for development of new plant-based options is suggested for all remaining cases, apart from where the asset value is low and groundwaters are unresponsive when no specific action is suggested. The low value, unresponsive categories are relevant to much of the agricultural areas, and thus for most land, no specific action will be justified to protect terrestrial assets.

Where the value of the asset is high, the response to groundwater under the asset is low and urgency is low (cases 11 and 12), the rationale for recommending development of plant-based solutions is similar to cases 11 and 12 in Table 4. We recognise that the economic value of off-site benefits from revegetation will be low in many cases, due to the scale of perennial vegetation needed to be effective and the time lags involved. However, the cost of successful plant-based R&D per hectare of perennials established is also low. It can be adopted and have impacts over very large areas, eventually resulting in worthwhile public benefits in cases where farms are contributing to off-site salinity. Although the salinity-related benefits from technology development for assets in this category are likely to be modest, they are benefits that are not efficiently attainable by any other means. Further, ‘successful’ R&D in this context means that the development results in solutions that are more profitable than current farming practices, resulting in economic benefits to farmers. There would also be a range of other economic, environmental and social benefits (e.g., diversification of income, provision of habitat for native fauna, carbon sequestration, provision of jobs in harvesting, processing and transport of products from woody perennials). So if salinity-related benefits are not sufficient to justify plant-based R&D, the combination of benefits is likely to be. Of course, in cases where perennials on farms would have no impact on the assets, or where opportunities for development of plant-based solutions do not exist or are too expensive, the recommendations would revert to "no action", at least with regard to salinity.

It is possible that land may not remain in one category, as groundwaters rise or technological progress occurs. For example, in the longer term, the situation for an asset may become urgent, potentially justifying on-site engineering responses that were not previously appropriate.

In contrast to the salinity impacts in relatively small, concentrated areas referred to in Table 5, Table 6 refers to more dispersed effects of salinity and shallow water tables, including impacts on agricultural land, on remnant areas of native vegetation on farms, and on flood risk. The common feature of these impacts is that, relative to some of the scenarios in Tables 4 and 5, there are low benefits per hectare from establishing perennials to prevent salinity. For agricultural land, this is partly because even highly productive agricultural land is not comparable in value per hectare to expensive public infrastructure; all agricultural land would be rated "low" using the value scale in Table 5. For flood risk, the reason is that the establishment of perennials on any given hectare of agricultural land makes only a tiny contribution to flood prevention. For protection of remnant native vegetation the reason is that most small remnants would be of lower conservation value than larger reserves. (When that is not the case, the remnant would be considered within Table 5 rather than Table 6.)

Table 6. Suggested policy responses for recharge areas on agricultural land that is contributing significantly to salinity on dispersed assets, including agricultural land, flood risk, and remnant native vegetation on farms.

Table 7 refers to land that is currently salt-affected. The two factors driving the recommended policies are the downstream impacts and the on-site economics. It is believed that revegetation of saltland can generate downstream benefits from improved water quality (salinity, sediment and nutrients), reduced flood risk and potentially provision of habitat for biodiversity but quantitative evidence is currently lacking. Deep open drains that are popular among some farmers in WA and SA may have negative downstream impacts due to the disposal of saline effluent in waterways, potentially contaminated with acidity and heavy metals in some cases. They may also worsen downstream flood risk by concentrating flows into a shorter period of time, although there are also suggestions that they may reduce flood risk. Again, quantitative evidence is lacking, either way.

We suggest that where downstream impacts are positive or neutral, the policy approach should be broadly similar to that in Table 6; that is, extension where the management options are profitable, development of new options (plant-based or engineering) where they are not. In this case, of course, the plant-based options would be based on productive salt-tolerant species. There may be cases where treatment of saltland would have downstream impacts that are sufficiently positive and on-site costs that are sufficiently low to justify payment of incentives, but this needs further analysis. Where a practice that is being adopted has downstream impacts that are sufficiently negative, penalties (e.g., a regulatory approach) may be justified.

The role for existing plant-based options in managing dryland salinity is represented in Tables 4 to 7 by three policy options: extension in cases where profitable plant-based systems already exist, incentives in cases where off-site benefits from perennials exceed the on-site costs (including opportunity costs from foregoing production of annuals) and penalties in cases where plant-based systems have adverse impacts on non-agricultural values, particularly reduced water yields in waterways.

Overall, it is rarely possible to justify intervening to force or provide incentives for adoption of perennials for salinity benefits alone. The time lags are so long and the required scales of planting are so high that the present value of salinity benefits per hectare of new perennials is generally small. This is true for all regional and intermediate GFSs and some of the local systems. If we want to do something about salinity in these catchments, the most realistic policy option is to invest in plant-based R&D, meaning the development of new and improved types of plants and plant-based systems that are more economically attractive to landholders. This approach applies to the largest area of land, by far. There are a number of elements to the rationale for recommending this option in so many of the cases considered in Tables 4 to 7, as outlined below.

• The existing suite of plant-based options is limited. Plant-based R&D can extend the reach of salinity management beyond the minority of cases where extension and incentives are appropriate.
• Salinity budgets are limited. The available policy budget for salinity is sufficient to influence management of only a minority of the land contributing to salinity problems. Plant-based R&D offers the only prospect for effectively managing salinity on a larger scale.
• Plant-based R&D will generate additional benefits beyond salinity mitigation, including diversification of farm income, provision of habitat for native fauna, carbon sequestration, and social benefits from provision of jobs in harvesting, processing and transport of products from woody perennials.
• Partial success in plant-based R&D will still be of benefit, particularly through reducing the incentives required for adoption of perennial plants.
• There are unrealised R&D opportunities. We believe that the current shortage of profitable options reflects the absence of past investment in this type of plant-based R&D, rather than the absence of opportunities to develop profitable systems. It is early days in the systematic search for new improved systems, but the opportunities are exciting.
• Plant-based R&D reduces the problem of justifying investments to prevent very long-term impacts of dryland salinity. The standard economic technique of discounting is used to convert future benefits into present values so that they can be compared with current costs of mitigating actions. However, for very long-term outcomes, even low discount rates mean that current costs dominate the decision. Plant-based R&D sidesteps this problem by using commercial levers to promote plant-based systems, with salinity mitigation as an ancillary benefit.
• The downside is low. Compared to the cost of existing salinity programs, the cost of appropriate plant-based R&D is very small, with the possible exception of managing the risks associated with the introduction of new plants. The ability of widely adapted plants to become weeds is a clear possibility, as raised by Semple et al. (2004, this conference).

Plant-based R&D is not as direct an instrument as other policy options. It is unrealistic to expect that ideal plant-based options will be created for all of the categories where plant-based R&D has been recommended. The R&D that delivers benefits may not even have been targeted at the particular category it ends up addressing. It may be best to think of plant-based R&D as a process of attempting to create opportunities for the community to benefit in a variety of ways: environmental, economic and social. Salinity-related benefits would come as part of a package of outcomes, and they may not even be the most substantial of the benefits that result. Indeed, they probably will not be. Nevertheless, plant-based R&D remains the approach that is by far the most likely to generate broad-scale benefits from salinity management.

Choices about the most appropriate government response to salinity should be sensitive to the hydrological and socio-economic conditions and to the types of assets under threat. Where the main aim of salinity management is to reduce impacts on water resources, the logical approach in some upper catchment areas is for penalties or permits to prevent loss of fresh water runoff entering waterways. There are few cases where providing incentives to grow existing plant-based options is the most appropriate response. Investment into plant-based R&D is justified in several cases, particularly where groundwater systems are responsive and the potential for runoff generation is low. In a minority of locations, salt interception schemes are technically and economically feasible.

For protection of high value, non-agricultural terrestrial assets (infrastructure and biodiversity), each of the policy approaches is relevant in some circumstances, although the role for incentives is very limited. Engineering (subject to economic analysis) may be appropriate when the value of the asset is high and the urgency for action is high. Plant-based R&D is relevant in a number of situations, particularly where the asset value is high but the urgency is low. It is justified on the basis of reducing the public cost per hectare of treatment.

Compared with infrastructure and biodiversity, agricultural land is generally of low relative value. Extension is the main tool where profitable perennial options exist. More commonly, where current plant-based options are not sufficiently profitable, R&D to develop improved options should be undertaken.

Where land is already salt-affected, development of plant-based or engineering options is justified where the downstream impacts are positive or neutral and where profitable options are lacking. A choice between penalties and no action applies where the downstream impact of managing salt-land is sufficiently negative.

Investment in profitable plant-based options remains an important management option to maintain water quality in rivers in the long term, and it still has an important contribution to make to the protection of land and other assets. There are three main plant-based management strategies that remain important in particular situations: (1) to reduce or delay saline discharge with existing plant-based options; (2) to make productive use of salt-affected land; and (3) to develop new perennial and/or salt-tolerant species and systems (using plant-based R&D) that are economically competitive with annual plant-based options. The third strategy, in particular, has a crucial role in achieving salinity management over much larger areas than will be possible with current perennial plant options.

Abadi A, Lefroy T, Cooper D, Hean R, Davies C (2003) ‘Profitability of medium to low rainfall agroforestry in the cropping zone’. RIRDC/Land & Water Australia/FWPRDC, Joint Venture Agroforestry Program, Rural Industries Research and Development Corporation, Canberra.

Anon. (2001) Australia’s Dryland Salinity Assessment, 2000. National Land and Water Resources Audit. (Land and Water Resources Research and Development Corporation: Canberra).

Anon. (2002). Salinity on Australian Farms, report 4615.0, Australian Bureau of Statistics: Canberra.

Barr N, Ridges S, Anderson N, Gray I, Crockett J, Watson W, Hall N (2000) ‘Adjusting for catchment management, structural adjustment and its implications for catchment management in the Murray-Darling Basin.’ Dryland Technical Report No. 2 (Murray-Darling Basin Commission: Canberra)

Barr N, Wilkinson R (2004). Social persistence of plant-based management of dryland salinity (this conference).

Bathgate A, Pannell DJ (2002). Economics of deep-rooted perennials in Western Australia, Agricultural Water Management 53, 117-132.

Bathgate A, Woolley J, Evans R, McGowen I (2004) Downstream benefits of salinity management: A case study for the Boorowa Catchment, unpublished paper presented at the 48^th Annual Conference of the Australian Agricultural and Resource Economics Society, Melbourne, 11-13 February 2004.

Bowman S, Ruprecht JK (2000) ‘Blackwood River Catchment Flood Risk Study, Water and Rivers Commission Report No. SWH 29’. Western Australian Government, East Perth, WA.

Cary J, Webb T, Barr N (2002) Understanding landholders’ capacity to change to sustainable practices; insights about practice adoption and social capacity for change (Bureau of Rural Sciences: Canberra)

Coram JE, Dyson PD, Houlder PA, Evans WR (2000) Australian groundwater flow systems contributing to dryland salinity. Bureau of Rural Resources project for the National Land and Water Resources Audit’s Dryland Salinity Theme.

Crabb P (1997) ‘Murray-Darling Basin Resources.’ (Murray-Darling Basin Commission: Canberra)

Dames and Moore - NRM (2001) The economics of predicting groundwater rise and salinity in rural towns. Final report for Rural Towns Steering Committee and Agriculture Western Australia.

Dawes WR, Gilfedder M, Stauffacher M, Coram J, Hajkowicz S, Walker GR, Young M (2002) Assessing the viability of recharge reduction for dryland salinity control: Wanilla, Eyre Peninsula, Australian Journal of Soil Research 40, 1407-1424.

Ewing M (2004) New trees and pastures – when, where and likely role in farming systems (this conference).

Guerin LJ, Guerin TF (1994) Constraints to the adoption of innovations in agricultural research and environmental management: a review. Australian Journal of Experimental Agriculture 34, 549-571.

Heaney A, Beare S, Bell R (2000) Targeting reforestation for salinity management, Australian Commodities, vol. 7, pp. 511-518.

Hobbs RJ, McIntyre S (2004). Setting limits to generalisation: a framework for considering landscape characteristics across Australia, Global Ecology and Biogeography (accepted for publication)

Hollier C, Francis J, Reid M (2003). Shrinking extension to fit a growing small farm sector. Proceedings of the 9th Australian Pacific Extension Network Conference, Hobart 26-28 November 2003. www.regional.ag.au

Hooper S, Martin P, Love G, Fisher BS (2002) ‘Get big or get out’ - Is this mantra still appropriate for the new century? Australian Bureau of Agricultural and Resource Economics Conference Paper 02.12, 24th Biennial Conference of the Australian Society of Animal Production, Adelaide, 11 July 2002.

Hutchinson MF, McIntrye S, Hobbs RJ, Stein JL, Garnett S, Kinloch J (2004). Integrating a global agro-climatic classification with bioregional boundaries in Australia. Global Ecology and Biogeography (submitted)

Keighery G (2000) Wheatbelt wonders under threat. Landscope 16, 37-42.

Kingwell R, Hajkowicz S, Young J, Patton D, Trapnell L, Edward A, Krause M, Bathgate A (2003) ‘Economic Evaluation of Salinity Management Options in Cropping Regions of Australia’. Grains Research and Development Corporation, Canberra.

Lefroy EC (2002) Forage trees and shrubs in Australia: Current use and future potential. (Rural Industries Research and Development Corporation Publication 02/039, Canberra).

O'Connell M, Young J (2002) The role of saltland pastures in the farming system - a whole-farm bio-economic analysis, SEA Working Paper 02/08, School of Agricultural and Resource Economics, University of Western Australia, Crawley, Australia.

Pannell DJ (2001a) Dryland Salinity: Economic, Scientific, Social and Policy Dimensions. Australian Journal of Agricultural and Resource Economics 45, 517-546. Final journal version (212K pdf file) also available via the Journal homepage:  www.blackwellpublishing.com/ajare

Pannell DJ (2001b) Explaining non-adoption of practices to prevent dryland salinity in Western Australia: Implications for policy. In: A. Conacher (ed.), Land Degradation, Kluwer, Dordrecht, 335-346.

Pannell, D.J. and Ewing, M.E. (2004). Managing secondary dryland salinity: Options and challenges, New directions for a diverse planet: Proceedings of the 4th International Crop Science Congress, Brisbane, Australia, 26 Sep - 1 Oct 2004.  Full paper (66K)
Later
Pannell, D.J. and Ewing, M.A. (2006). Managing secondary dryland salinity: Options and challenges, Agricultural Water Management 80(1/2/3): 41-56. Full paper (66K)

Pannell, D.J., McFarlane, D.J. and Ferdowsian, R. (2001). Rethinking the externality issue for dryland salinity in Western Australia, Australian Journal of Agricultural and Resource Economics 45(3): 459-475. full paper (114K) brief version (11K). newspaper article based on the paper (6K). Final journal version (164K pdf file) also available via the Journal homepage:  www.blackwellpublishing.com/ajare

Ridley AM (2004). The role of applied science to help farmers make decisions about environmental sustainability. Australian Journal of Experimental Agriculture (accepted)

Ridley AM, Christy BP, White RE, McLean T, Green R (2003) North-East Victoria SGS National Experiment Site: water and nutrient losses from grazing systems on contrasting soil types and levels of inputs Australian Journal of Experimental Agriculture 43, 799-815.

Semple W, Cole I, Koen T (2004). Exotic v. native salt-tolerant species: our choice now but a potential burden for future generations? (this conference)

Vertessy R (2001) Impacts of plantation forestry on catchment runoff. In Nambiar and Brown (eds) Plantations, Farm Forestry and Water. Proceedings of National Workshop 20-21 July 2000 Melbourne RIRDC publication 01/20 9