Piloting a systematic framework for public investment in regional natural resource management: dryland salinity in Australia

Anna M RobertsA,C and David J PannellB,C*

ADepartment of Primary Industries, RMB 1145 Rutherglen, Victoria, Australia 3658

BSchool and Agricultural and Resource Economics, University of Western Australia, 35 Stirling Highway, Crawley, WA, Australia 6009

CFuture Farm Industries Cooperative Research Centre


The Salinity Investment Framework (SIF3) is designed to help environmental managers make better policy decisions about protecting key assets under threat of dryland salinity, a major degradation problem in Australia. This study reports results from applying SIF3 in partnership with two regional organisations (in the states of Victoria and Western Australia) with contrasting institutional powers and structures. We also engaged relevant policy makers with a view to influencing national policy. Available knowledge, science, and data were used to develop region-specific recommendations for public investment in salinity management. SIF3 proved easy to apply, requiring minimal changes from its original version. Establishing trust and credibility with stakeholders was extremely important. More time was required to be spent on communicating what SIF3 was and what it would mean for stakeholders than in applying it in the catchments. In both regions, recommendations included making a clear distinction between localised assets (such as a particular wetland) and dispersed assets (such as agricultural land as a whole); more targeted investment in spatially identified localised assets; using a broader range of policy tools; basing recommendations on the levels of public and private net benefits; anticipating the adoptability of sustainable land-management practices; and formally integrating relevant scientific, economic and social information with community values. The differences between current action and recommendations were significant in both regions, particularly in Victoria. As a result of our work both regions have decided to implementing a similar asset-based approach covering various environmental threats in addition to dryland salinity. The research indicates that it is feasible to use decision frameworks such as SIF3 to improve the rigour of decision making by community-based environmental management organisations. A broader range of policy tools are needed for dryland salinity, with less reliance on extension and small grants, and more emphasis on intervention around key assets and investment in technology development. We have influenced policy makers to some extent at state level and nationally through discussion of SIF3 in a national inquiry, preparation of a policy paper for peak multi-government policy committees, briefings and provision of advice on policy design. Despite this, achieving significant policy change at the state and national levels remains difficult for a number of reasons. Governments give limited signals to environmental managers to achieve environmental outcomes, being more concerned with community participation objectives and political considerations. We hope for significant policy change with increased scrutiny about sound decision making and public accountability.

Keywords: catchment management; integration; targeting; economics; behaviour change; evaluation; institutions; prioritisation; planning.


In countries including Australia, Canada, New Zealand, the United Kingdom and the United States of America, the responsibility for natural resource management (NRM) is being partly devolved to regional bodies defined on watersheds (catchments) (Ewing, 2003; Gutrich et al., 2005; Seymour et al., 2007). In Australia there are 56 regional organisations responsible for catchment management across the six States and two Territories. The institutional arrangements involve publicly supported community partnerships, building on the base of ‘Landcare’, a voluntary community-based model that was established in 1989 (Curtis et al., 1995).

Despite the short history of European-style agriculture (at most 200 years, and less than 50 years in some areas) and despite its low population base, Australia has large problems of environmental degradation in rural areas (Henzell 2007). These include biodiversity decline, salinity, soil acidification, poor water quality in water bodies, soil erosion, other types of soil degradation (physical, chemical and biological), and pest, weed and disease problems (Hobbs and Saunders, 1993; Scott et al., 2000; Ridley et al., 2004; Anon., 2007; Henzell, 2007).

Secondary salinisation (dryland salinity) is one of the major environmental degradation problems in Australia. The causes are well known, largely resulting from the replacement of native vegetation with more shallow-rooted annual species which has altered the water balance and increases recharge to groundwater. The additional recharge causes watertables to rise, dissolving and mobilising stored salts (primarily NaCl) and mobilising them such that they affect rivers, native vegetation, infrastructure and agricultural production (George et al. 1997; Anon., 2001). The largest and most recent program to address salinity in Australia, the National Action Plan for Salinity and Water Quality (referred to hereafter as NAP), invested A$1.4 billion of public funds over 2000-2008 (Pannell, 2001). NAP largely focussed on providing extension and small, temporary grants to landholders based on the premise that the land-use changes that resulted would be sufficient to have environmental benefits. Pannell (2001) argued that the NAP approach was fundamentally flawed. His main reasons were: (a) that there were few technical options that were attractive enough for farmers to adopt at a scale needed to address the problem, even allowing for the available financial support; (b) that applying persuasion, peer pressure and distributing information to encourage a conservation ethic was insufficient to prompt farmers to change management over the scale needed; (c) that the community-based regional bodies to which decision making was devolved would be unable to resist community pressures that would lead to poor prioritisation of investments; (d) that there were major advances in the scientific knowledge of salinity, indicating that it is much more difficult to manage than previously hoped; and that (e) the design of the NAP did not account for these advances in knowledge.

Following advocacy by Pannell to state government bodies in Western Australia, work was undertaken to develop and apply more systematic decision frameworks for public investment in salinity in that state. The Salinity Investment Framework (SIF) versions 1 (McAlinden et al., 2003) and 2 (Sparks et al., 2006) identified high-value assets at high risk from salinity, with high feasibility of protection. Piloting of SIF1 with a regional environmental body in Western Australia, and feedback from other bodies, indicated that an easier-to-apply framework was required. It also reinforced that the magnitude of the salinity problem is very large relative to the available budget, meaning that only highly cost-effective public investments are well justified.
From this experience, the conceptual framework now called SIF3 was developed (Ridley and Pannell, 2005; www.sif3.org). SIF3 was an advance on SIF1 and SIF2 in that it embedded key findings from salinity science and modelling, including transparent choice of policy tools, highlighted the importance of adoptability of land-use changes and highlighted that there is a big range in salinity benefits per hectare of land-use change.

Having developed SIF3, it was important to test its practical application. This paper: (a) reports on implementation of SIF3 with two Australian regional bodies – the North Central Catchment Management Authority (NCCMA) in Victoria and South Coast Natural Resource Management (SCNRM) in Western Australia; (b) reports how recommendations differ from current investments in the regions; (c) discusses how the participating organisations have responded to the recommendations provided through SIF3; and (d) discusses policy implications and policy responses to the research.

Materials and methods

The regional organisations

We chose to pilot SIF3 in two contrasting regions, both of which were among the 21 priority regions that received funding under the NAP due to their salinity problems. The North Central Catchment Management Authority (NCCMA – www.nccma.vic.gov.au) is located in the south-eastern Australian state of Victoria within the Murray-Darling River Basin. Its major city is Bendigo and it includes four main river catchments, three of which drain into the Murray River. It contains both irrigated and dryland agriculture. SIF3 is focussed on dryland areas where land uses range from dominantly dryland cropping in the northern parts of the catchment, to mostly grazing (sheep and beef) in the hilly, higher rainfall southern zones. Victorian CMAs have had statutory powers since 1997, with responsibilities under the Water Act and the Catchment and Land Protection acts (www.nccma.vic.gov.au). About two-thirds of the CMA’s funding comes from the State of Victoria (Department of Sustainability and Environment, DSE) and the remainder from the national government. The Department of Primary Industries (DPI – www.dpi.vic.gov.au), a Victorian state government agency, has traditionally provided much of the underpinning science for salinity investment, and is also the major provider of public extension services in the region. An asset-based investment tool like SIF3 had not trialled previously in Victoria.

South Coast Natural Resource Management (SCNRM – www.southcoastnrm.com.au) is an incorporated non-profit association, independent from government and without statutory powers. The region encompasses the coast and hinterlands of southern Western Australia, including the major regional centres of Albany and Esperance. It includes a major part of one of the world’s 25 most important biodiversity hotspots, defined as areas with an “exceptional concentration of endemic species undergoing exceptional loss of habitat” (Myers et al., 2000). Dryland agriculture ranges from broadacre cropping with few livestock in the wheatbelt to grazing enterprises in the wetter, more coastal areas. The region has serious problems with dryland salinity and funding comes mainly through major national programs like NAP and the Natural Heritage Trust (NHT).

The SIF3 framework

The SIF3 framework is briefly described below; more details can be found on www.sif3.org. In broad terms, the philosophy of the SIF3 framework and this pilot study included that it should be: firmly grounded in current research knowledge, integrated with local values and knowledge; comprehensive in its assessment of investment options; transparent and repeatable; easy to use; able to be applied using readily available information, or improved information as this becomes available; business-like in recognising limited budgets and in seeking the greatest environmental outcomes for the available budgets; and realistic in terms of the likely behavioural responses of land managers. The approach is intended to make it easy to condense a long list of investment possibilities into a short list of good investment prospects, suitable for detailed feasibility assessment.

SIF3 distinguishes between:

• localised assets: discrete, high-value assets that can be spatially localised, e.g. a wetland; and
• dispersed assets: assets that are spread across the region, such as agricultural land, or the many small parcels of remnant native vegetation on farms.

The distinction between localised and dispersed assets is not clear-cut, but it is helpful in classifying the likely benefits per hectare of land-use change, which has important implications for the type of policy mechanism that should be used.

The payoff from successfully investing in well chosen localized assets is likely to be high. This means that it may be feasible to use relatively expensive approaches, such as engineering works, or high levels of incentive payments, to protect those assets. The assets selected for funding would ideally be particularly valuable, facing high environmental threat, with high feasibility of protection, and marginal adoptability of the relevant works needed to protect them.

To compete with investment in localized assets, investment in dispersed assets needs to be relatively low-cost per hectare, and highly effective over large areas. Appropriate responses may include technology development (developing new land-use options that are sustainable and highly adoptable), extension (where such land-use options exist but have not yet been adopted), and conservation tenders (which may reveal highly cost-effective interventions).

For each asset, SIF3 requires specific information to allow the assessment of likely costs and benefits of investing. The data were used to apply the SIF3 decision tables, as described by Ridley and Pannell (2005). These decision tables integrate the required bio-physical, and socio-economic information to make a specific recommendation about management and the most appropriate delivery mechanism. The required information varies depending on the type of asset (see Ridley and Pannell, 2005) but broadly includes the severity of salinity threat to the asset, the timing of that threat, the technical feasibility of averting the threat, potential negative side effects from protecting the asset, and the adoptability of proposed land management practices that would need to be taken up by landholders (considering characteristics of the practices and characteristics of the landholders who we would like to see adopt them – e.g. Pannell et al., 2006).

SIF3 has some characteristics in common with benefit: cost analysis (BCA). It attempts to capture and interpret previous modelling and research to make recommendations that would be consistent with a BCA, without requiring users to actually conduct a BCA. In that way, it makes the concepts and insights of BCA more accessible to users who lack a background in economics. SIF3 also has some commonality with multi-criteria analysis, in that it draws together a range of criteria relevant to decision making and combines them to support decision making. However, the weights applied to criteria in multi-criteria analysis are arbitrary, whereas in SIF3 they are combined in a more structured way (not simply a linear weighting of different criteria) with an explicit focus on cost-effectiveness.

Piloting the SIF3 framework

The piloting process proceeded through several stages, as follows.

(a) Consultation, communication, participation

We consulted intensively with the regional NRM bodies to understand their histories, their current strategies, and the relevant institutional arrangements and service-provision relationships. We held a field tour for several days in each region where the regional body and government officers helped us understand the region, become familiar with relevant knowledge, view current work, and network with local stakeholders including landholders. We reviewed key documents, which were mostly unpublished, and sought knowledge from experienced local people. In each region we came to rely heavily on several experienced local people.

Understanding the importance of communications in such a venture, we:

• Maintained close relationships with key staff and committees;
• Shared findings with other parties only with agreement of our regional partner;
• Provided briefings, seminars and phone communication on an ‘as needs’ basis to all stakeholders;
• Responded quickly to any query about the work;
• Maintained close contact with state agencies in local and senior management and policy positions;
• Provided seminars and regular briefings to senior staff of the NRM bodies.

(b) Identifying high-value natural resource assets in the region

As stated earlier, SIF3 distinguishes between localised and dispersed assets. We identified high-value localised assets as follows:
(i). Compiling a list of assets based on their official government recognitions in international (e.g. Ramsar listing www.ramsar.org accessed 22 February 2008), national, state and bioregional classifications;
(ii). Conducting workshops with key stakeholders and community members to identify assets of highest value to the local community;
(iii). Consulting local experts;
(iv). Discussing draft asset lists with staff and other stakeholders of the regional bodies.

In the NCCMA region, for example, this process was conducted for 101 river reaches (based on the region’s River Health Strategy; Anonymous, 2005a), and over 200 wetlands of National and State significance, as well as infrastructure and terrestrial biodiversity assets. A list of 80 assets of very high value was developed, of which 40 were judged to be facing high salinity threat. In Western Australia, the short list included 56 localised assets, 22 of which faced notable salinity threats. All remaining land, dispersed native vegetation and wetlands not classified as localised assets were considered as dispersed assets.

For dispersed assets, we did not undertake an asset-by-asset analysis. Rather, we prepared a discussion paper designed to assist each regional body to make an explicit decision about the balance of investment between localised and dispersed assets, and we identified locations that should be priorities for investment in technology change, one of the key strategies for dispersed assets. The intention is that technologies developed for these locations would diffuse to other parcels of the dispersed asset.

(c) Assembling data and conducting the SIF3 analysis

Data were obtained from a wide range of sources including: official government data sets; local expert opinion; unpublished reports commissioned by the regional bodies; observations of landholder behaviour and attitudes; interviews with landholders; and published research. In Victoria the main technical data sets were available across the whole region allowing the analysis to be conducted using geographic information systems (GIS). In Western Australia, this was not possible, and there was a stronger reliance on local expert knowledge.

In both regions there was a lack of formal knowledge about the adoptability of sustainable farming practices. We relied on the knowledge of local extension officers, discussions with landholders, field observations and knowledge of the general literature on adoption of farming practices (Pannell et al., 2006) to assess the extent to which landholders were likely to adopt new practices for salinity management, particularly perennial pastures such as lucerne or phalaris. In general, adoptability at large scales was judged to be poor (or marginal at best) in most parts of both regions, as has also been concluded more broadly in Australia (Roberts et al. 2009). Our assumptions on adoptability were based on discussions with local extension officers, community members and observations in each region.

The data were used to apply the SIF3 decision tables, outlined previously by Ridley and Pannell (2005). In general, this was found to be straightforward, although in a few cases we were prompted to modify the tables, as reported later. A short list of recommended interventions was developed. The items of this short list require further detailed assessment of feasibility prior to investment.

(d) Reality testing and refining the recommendations

Results were initially presented to key decision makers and committees in each regional body. Detailed reports were developed and circulated to interested stakeholders including staff members from the regional body, state government agencies, local landholders, and local government. Feedback was obtained, and recommendations modified where appropriate.
Working with governments

As well as working with the regional bodies, we developed and maintained active engagement with funding investors, policy makers and policy advisers at the national and state level. In Victoria this was principally the Department of Sustainability and Environment (DSE) and the Department of Primary Industries (DPI), and in Western Australia it was the Department of Agriculture and Food Western Australia (DAFWA), the Department of Water (DoW) and the Department of Environment and Conservation (DEC). We also briefed policy makers in the states of New South Wales and South Australia who were keen to find out about our work.

DSE and DAFWA have different, but powerful roles in influencing the success of catchment management. We believed that a combined bottom-up and top-down strategy would be most effecting in advising policy change. Our approach had three main elements:

(i) Briefings in person and through email at strategic times in the project’s life so policy makers and funders knew what we were doing, there were no unexpected surprises for them, and that we understood what was influencing their thinking;
(ii) Inviting their input into the design and process of implementing our work;
(iii) Responding promptly to any queries and concerns that policy makers or funders had.

For the Commonwealth Government, who are more distant from implementation than are their state colleagues, we mainly used periodic briefings on progress and invited feedback at any time. The NCCMA and SCNRM were kept aware of what we were communicating with state and national governments.


Applicability of the SIF3 decision tree

Once the relevant data had been collated, we found that it was straightforward to apply the SIF3 framework to derive recommendations.
However, in applying the framework we identified the need for some changes to the original decision tables published by Ridley and Pannell (2005). The amended decision tables can be found at www.sif3.org. Changes were as follows:

• Several scenarios previously labelled as ‘not applicable’ (e.g. waterways with high salt input, low groundwater responsiveness to intervention, and high fresh water runoff) were found to exist. Recommendations were thus developed and included in the framework.
• The original criteria referred to the farm-level economics of recommended practices, as an indicator of the adoptability of those practices. It was recognised that it should refer more generally to their adoptability, which is affected by many non-financial factors (Pannell et al., 2006).
• The framework was adjusted to include landholders with predominantly lifestyle objectives, rather than business-related objectives, particularly for those with smallholdings. Additional social research was initiated to address this (Wilkinson 2007).

We intended SIF3 to be usable with existing, readily available data. It proved to be so, but in applying it we identified a number of important data deficiencies. For example, there appears to be a need for improved data sets on groundwater salinity and groundwater depth in Victoria. In general the scale of data layers (commonly 1:250,000 scale for groundwater flow systems in Victoria) was too coarse for assessing conditions at the scale of the asset. Expert and local knowledge was at least as important as available data, especially in Western Australia. These issues were identified and reported to regional bodies as priorities for research.

Although the SIF3 decision tables (Tables 4-7 detailed in Ridley and Pannell 2005) proved simple to apply, through various communication activities we recognised the need to more clearly explain and test the underlying logic of the framework. From this, we developed the public benefits: private benefits framework, which presented graphically part of the core logic of SIF3 as it applies to changes on private land (Pannell 2008). It proved to be a particularly powerful communication tool, giving confidence to people that the logic of SIF3 was sound, clarified some concepts and removed some of the “black box” feel of the original framework. The public benefits: private benefits framework has become a useful tool in its own right and is being used as an underpinning, generic component as we broaden the framework to encompass threats other than salinity.

Participation and communication

Stakeholders with whom we engaged were as follows:

• The Chief Executive Officer (CEO) and chair of the regional NRM bodies;
• Senior staff of the regional NRM bodies;
• Boards and relevant committees of the regional NRM bodies;
• Community members with close links to the regional NRM bodies;
• The community generally through the media;
• Researchers;
• State government officers who are close collaborators with or service providers to the regional NRM bodies;
• State government officers who provide funds and detailed policy directions to the regional NRM bodies (in Victoria);
• Senior state government policy officials and committees;
• Senior policy officers in the national government;
• Peak national committees that bring state and national governments together to set overall policy directions, including a committee of government agencies, and a committee of government ministers;
• Government inquiries into the performance of NRM policy.

Maintaining communications with so many disparate groups was very demanding of time and resources, but was essential in order to increase the probability of acceptance of the framework. Purposes and outcomes of the communication included:

• Initially gaining formal endorsement and support from senior players in the NRM bodies;
• Promoting interest and participation of CMA staff and stakeholders, to build our credibility and empower the Board to change the current approach;
• Raising awareness and understanding of the framework;
• Increasing detailed understanding by CMA staff and relevant service providers;
• Learning details of the existing institutional arrangements and key personalities;
• Countering negative perceptions or negative messages being promulgated by some stakeholders;
• Identifying such issues at a very early stage;
• Fostering champions for SIF3 in the different organizations;
• Increasing broader awareness and acceptance of the work and its implications; and
• Advocating for change at the policy level.

Comparison of SIF3 recommendations with current approach

Compared to existing strategies in the two regions, the SIF3 analysis differed by: targeting investment more explicitly on spatially identified localised assets; considering a full range of policy tools; making a clear distinction between localised and dispersed assets; basing recommendations on the levels of public and private net benefits; anticipating the adoptability of sustainable land-management practices; and using the SIF3 decision tree to formally integrate the relevant scientific, economic and social information with community values for assets.
In the NCCMA investment was made in 10 priority sub-regions (called target areas). The predominant tool used was extension, supported by small temporary grants (commonly $60/ha) to encourage trialling of new practices. The target areas were mainly locations with large areas of salinity-threatened agricultural land, with some consideration of salt loads in waterways (Anon., 2002). The focus on agricultural land was largely due to the membership of committees and participants in key processes, which were dominated by farmers and by current and past staff from the state agricultural agency, as has been found previously by Curtis et al. (1995).

Example of SIF3 recommendations for some of the localised assets in the North Central region are shown in Table 1. By our assessment, extension and small temporary grants are limited as the main response for any of these assets for salinity protection, due to the lack of land-use options that will avert salinity and be adoptable on a sufficient scale. In general, the commonly recommended practices of planting lucerne or phalaris perennial pastures are only adoptable on scale insufficient to reduce the extent of dryland salinity. This is evidenced by their limited uptake despite extension and incentive payments in the region for many years, although we acknowledge that dry years were also a contributing factor. For effective salinity mitigation, adoption would be needed over a large proportion of target sub-catchments. From the SIF3 analysis we recommend engineering works or larger, long-term incentive payments (subject to further analysis on feasibility and cost-effectiveness), and/or development of improved technologies that will be more widely adoptable. The recommendations for localised assets are highly context specific.


Table 1. Selection of localised assets in the Avon-Richardson and Loddon catchments and summarised SIF3 recommendation.

Asset or  cluster

Asset values

Summarised SIF3 recommendation

Avon-Richardson Catchment    

Lake Buloke (wetland)

National significance, high community value

TDA for region, no direct action – intractable problem and degraded asset

Donald township (town, river)

High economic and community value

Engineering is economic, otherwise TD

Avon Plains Lakes (wetlands)

Bioregional significance, high community value

Perennials around lakes with long-term incentives and land use caveats, subject to community interest, as engineering is unlikely to be economic

Cope Cope lakes

Bioregional significance, community value

No action – intractable problem

Mt Jeffcott biodiversity reserve

Bioregional significance, high community value

No action – not threatened by salinity

York plains wetlands and river reach

Bioregional significance, very high community value

Area too large for engineering. As for Avon Plains lakes

Chirrup wetlands

Bioregional significance

No direct action due to asset value not being high enough. TD for region

Richardson River reach

High priority river reach

No direct action – intractable problem. TD

Loddon Catchment    

Tang Tang wetland

Wetland of national significance

Engineering if economic, as for  Avon Plains lakes if not.

City of Bendigo

High economic and community value

Engineering if economic.

Laanecoorie reservoir

Community and economic value

No direct action – TD

Bet Bet, Timor West area (river reach, land, native vegetation)

Dispersed land and vegetation, river reach, not as high in value as many other areas

Land retirement and/or TD

Moolort region -large cluster of wetlands, waterways and reservoirs

One wetland with national significance, reservoirs have economic and community value

Need more information on salinity risk.  Probably long-term incentives for perennials. Given the size of the area engineering not cost-effective

Tanfards Swamp

Low value

No action – low asset value

Upper Loddon river reach

Very high priority river reach

Establish salinity risk. Recommendation could be penalties to prevent perennials and maintain runoff

TD = technology development.


Figure 1 shows, for North Central region, the old target sub-catchments (on the left) compared with locations of localised assets identified using SIF3 as good prospects for investment in engineering or grants to landholders (on the right). Only one of the new priorities lies within one of the original target areas, indicating that the assets that should have been of the highest-priority were not being protected from salinity.


Figure 1. Comparison of old salinity target areas (left hand side) with localised high-value assets identified using SIF3 as priorities for protection (right hand side).


The differences between current action and SIF3 recommendations were greater for the NCCMA than for SCNRM. Several major assets that were identified in the SIF3 analysis were already targets for substantial investment by SCNRM (Lake Warden, the Lower Kalgan River, the Wellstead Estuary), and several others had been been subjects of major investment by government agencies (Upper Kent River, Lake Magenta, Lake Muir). With the exception of Lake Warden, the actual investment strategies are somewhat different to those recommended by SIF3 (for salinity). The existing plans and interventions for Lake Warden are well aligned with the SIF3 recommendations. Overall in the South Coast, the current investment places less emphasis on extension than in the NCCMA and there were also several research projects aimed at finding profitable perennial options, in recognition that there were insufficient options for landholders currently. These projects were in good agreement with SIF3 recommendations. A concern for SCNRM that was similar to NCCMA was whether the resources allocated to particular projects are sufficient to achieve the desired NRM outcomes. We suspect that generally they are not. As is common in other regions, some projects are pursuing ambitious targets with modest budgets.

Influence on planned salinity investments

The SIF3 pilots have strongly influenced both regional NRM bodies. Results have been presented during a sequence of meetings with the CMA’s CEO, Chair and Board. Its responses have included the following:

• The NCCMA has rewritten its investment plan under which investments in dryland areas are made, based firmly on the SIF3 recommendations;
• The Board and NCCMA staff conducted workshops in regions which had been target areas under the previous strategy, communicating results to communities and stakeholders and letting them know that they would not continue to receive grants as previously;
• The Board has committed to use an asset-based approach addressing multiple environmental threats to redevelop its regional catchment strategy;
• Training sessions have been conducted for NCCMA staff, its main committee responsible for implementation of programs, and extension staff who provide services to the NCCMA;
• The NCCMA has formed a research partnership to address the need of further technology development;

SCNRM has also viewed our results positively. A number of steps have been taken in response to the pilot.

• Early in the pilot, the Land Reference Group of SCNRM began to discuss a greater emphasis on technology development and participatory research than currently. They have fostered and supported field research of this type in the region;
• The research team has regularly been invited to brief all relevant committees and stakeholders, and they have indicated a very strong interest in the results and a willingness to change direction;
• The organization has been an enthusiastic and active collaborator in providing data and organising events related to the project;
• The CEO of SCNRM organised for us to brief the CEOs of other regional NRM bodies in the state about the framework and results;
• They have partnered with us in research to broaden the SIF3 framework to cover other environmental threats and have actively promoted this work to other Western Australian regions.

Influence on policy

The influence on policy makers from this project has included:

• Extensive discussion of SIF3 in the report of an inquiry into salinity by a committee of the Senate (the upper house of the national government), including "Recommendation 22: The Committee recommends that the Australian Government in cooperation with the states and territories keep a watching brief on the development of the Salinity Investment Framework 3 (SIF3), with a view to potentially implementing it (or a modified version of it) across the country. It is recommended that the framework be applied within the context of the new (post-2008) program(s)." (Anonymous, 2006, p229-230).
• Subsequent questioning of government officials by members of the above Senate Committee at Senate Estimates hearings regarding progress with implementation of SIF3.
• Preparation by the Victorian State Government of a paper based on the lessons and implications of SIF3 for presentation to peak multi-government policy committees, including principles for future public investment in salinity. The paper was endorsed in 2006 by the NRM Standing Committee (consisting of senior officers from each state and national government) and in 2007 by NRM Ministerial Council (consisting of relevant ministers from each state and national governments).
• Active support by the Australian government for briefings of senior policy officials in states other than Western Australia and Victoria by members of the project team.
• Invitations from state government bodies to provide briefings, and subsequently to provide specific advice on policy design issues (e.g. on the ‘NRM Standard’ developed by the Natural Resources Commission in the state of New South Wales (Anon. 2005b), and the NRM Investment Framework being developed by the Natural Resource Council in Western Australia).
• Invitations to address a number of other regional NRM bodies.
• Earlier criticism of our work by some policy advisers has largely ceased.


Participation and communication

The implementation of SIF3 was transdisciplinary (Tress et al. 2007) and adaptive. The project directly involved the public, government, and experts from a number of academic disciplines (Norton and Noonan 2007). This approach was important, given that we were seeking substantial changes in understanding and investment patterns.

The project team gave a very strong emphasis to communication. Inadequate information and prior tensions are powerful motivators for generating mistrust (Gezelius and Refsgaard 2007), and we observed that this had been evident in one of the regions. Communication was more difficult and time consuming than application of the framework itself, and was at least as important. Due to the newness of the asset-based approach concept, especially in Victoria, without such active and broad-ranging communication, we believe this work would not have succeeded in changing investment decision making at the regional level, and creating an environment for broader change at the policy level. The project perhaps provides a salutary lesson to other researchers who aspire to change institutional and policy arrangements in a well-established area like salinity: the research is likely to be the easiest part. Changing mindsets takes time and a lot of communication.

Carolan (2006) suggests that increasing complexity and uncertainty should be accompanied by increasing democratisation of the process, where scientific information is integrated with community values and preferences. Our approach was consistent with this. Community participation and involvement in NRM was stated to be a major reason for establishing the regional NRM system in Australia. For example, the SCNRM’s purpose as stated on its website ‘is to bring together people, organisations and information to ensure that the regional community drives the management of natural resources’ (www.southcoastnrm.com.au). By our observation, the regional NRM bodies tend to do a better job of the community consultation aspect of their role than of the scientific side. Part of our motivation in this project was to assist regions to strike a better balance between the two.

The high level of communication and consultation involved in the project was not specific to SIF3 – it was consistent with what one would need for any new decision framework if one wished it to be understood and used. As more users become familiar with the approach, we are finding that the communication requirements are reducing. We spent much less time communicating to stakeholders in Western Australia, where the environmental management body was familiar with earlier versions of SIF, than in Victoria, where it was novel.

Institutional issues

The decisions and practices of environmental managers and environmental policy makers are strongly shaped by their institutional arrangements. We needed good knowledge of these institutional arrangements in order to understand the constraints that managers and policy makers face, and to know how best to pursue change.

Each of the regional NRM bodies obtains almost all of their resources from governments, state and national. Projects conducted by NRM bodies must address priorities set by governments. Sometimes the priorities are well defined, documented and reasonable. Sometimes there are expectations that are understood but not documented (for example getting money spent on on-ground activities within short time-frames). At other times the regional bodies struggle to work out what the funding agencies really want. Further, the NRM bodies must be responsive to several government agencies, and the requirements of different agencies are not always consistent. Indeed we have observed conflicting messages coming from within a single organization.
Institutional issues pose a threat to long-term success of both the regional model itself and to the use of investment frameworks such as SIF3. Likely sources of institutional challenges include: a) priorities of state and national goverments that are not well aligned, sending mixed or confused messages to regions; b) differences in regional priorities compared with those of the state or national governments; c) the institutional relationship between a regional NRM body and its local NRM government agencies, who may provide information, services or resources; and e) whether statutory powers and financial independence devolved from the state to the region are sufficient.

In this pilot project, despite different institutional arrangements, and different degrees of familiarity with the SIF approach between the regions, our partnership with both organisations produced very encouraging results. It would, however, be unwise to conclude that we could achieve these results with every regional NRM body. Both NCCMA and SCNRM were actively seeking to improve their investment processes, wanted to be seen as leading regional bodies and wanted a rigorous, scientific approach to investment. The next phase of our work will be to expand the pilot to several new regions and continue work in both NCCMA and SCNRM.

Reasons for success

This project prompted rapid change in salinity investment in the two regions. A number of elements of our strategy have contributed to the success in beginning to change investment at the regional level. Our timing was good. We chose regions that were keen to improve their current approach to investment planning. The current national NRM programs were coming to an end (in mid 2008) and the regions were keen to position themselves for the future. We had a strong background in understanding the technical and institutional aspects of the salinity problem and used the science and local knowledge that was available. We invested considerable time working with people from the regional NRM bodies and their stakeholder groups and were supported by key local champions who alerted us to problems and supported us on the ground. They helped maintain the momentum for change and helped us respond to challenges as they arose. We maintained strong communications with institutions that had funding relationships with the regional bodies. The decision framework we used was rigorous and the processes were transparent. We engendered trust through communication and encouraged feedback by all stakeholders. As outlined by Gutrich et al. (2005), trust, integration of science, transparency and inclusion of public feedback are important if science is to be integrated within decision making about co-operative ecosystem management.

Policy implications

The experience of piloting SIF3 in two regions in Australia has revealed a number of implications for policy. Most importantly, we believe that the pilots have demonstrated that it is feasible to apply a systematic decision framework such as SIF3 to select NRM investments that are likely to yield more valuable environmental and natural resource outcomes for the available budget. SIF3’s focus on outcomes is much needed; existing processes imposed by government funding agencies tend to encourage a focus on activity and expenditure of funds.
Our experience reinforces the view that good regional NRM planning and prioritisation requires a strong evidence base and analysis, combined with judgement by decision makers. Many regional plans have large room for improvement in the use of evidence and analysis. For example, for many of the works that have been funded, there is little knowledge of their likely NRM outcomes (Auditor General, 2008). For government, this suggests the need for an improved process of accreditation of regional plans. As part of this accreditation, plans should be assessed to see whether they make appropriate use of science to determine likely NRM outcomes from planned actions. In addition, regions need support from governments through the provision of quality-assured data. There are important gaps in the required data in many (and perhaps all) regions.

Current processes for selecting NRM targets are often weak. Targets should be based on analysis, and should realistically reflect the available resources and the likely behavioural responses on landholders. Very few of the existing targets do so. Existing targets have an aspirational flavour (which has been encouraged by government guidelines). Commonwealth and States should revise their requirements for target setting to ensure that targets are measurable and achievable, which implies the need for them to be spatially explicit and time-bound.
In the early days of the current regional arrangements, regional bodies received little specific guidance on planning and prioritisation activities. Over time, more guidance has been forthcoming, usually in the form of advice about processes and principles that should be adhered to (e.g. the Natural Resources Commission of New South Wales has developed an NRM “Standard”) (Anon. 2005b). Our experience in this project indicates that insufficient support has been given. Integrated decision frameworks like SIF3 are required to help regional decision makers.
The reasons that regions and governments would be interested in pursuing an asset-based framework such as SIF3 can be found in a recent report by the Australian National Audit Office (Auditor General, 2008). The report concludes that, amongst other things, a) there is room for improvement in addressing the transparency and accountability of Australian government funds managed by the States; b) there has been little evidence of substantial landscape repair and replenishment of natural resources from the NAP and NHT programs; c) that reporting of outcomes needs to be substantially improved; d) that stronger targeting towards the highest priorities and critical national assets is necessary to achieve measurable results. An investment framework such as SIF3 would assist in all of the above areas. It serves in part to illustrate both the size and complexity of the task given to regions and also the importance of getting institutional arrangements and accountability well aligned.

Influence on policy

Our project included active engagement with policy makers and policy advisers at the state and national level, our judgement being that a combined bottom-up and top-down strategy would be most effecting in promoting policy change. Where seeking to improve policy for the public good, influence can be enhanced by: a good understanding of the policy process, good networks with policy officials, excellent communication free of jargon, an orientation on solutions rather than problems, pragmatism, practicality, persistence, good timing, demonstrated broader support, and the credibility of advocates or advisers (Pannell, 2004, Pannell and Roberts 2009).

Our role as scientists has moved well beyond the boundaries of being ‘pure scientist’ (not being involved in policy at all) or ‘science arbiter’ (answering factual questions which policy makers ask). Pielke (2007) describes two additional roles of science in policy decision making, the ‘issue advocate’ (limiting the scope of choices available by promoting some alternative over others), and the ‘honest broker’ (who seeks to expand or at least clarify the scope of choices to allow the decision maker to reduce choice based on their own preferences and values). We have certainly been ‘issue advocates’ in promoting SIF3 as the preferred framework. We have also been ‘honest brokers’, especially with regard to expanding the use of high quality scientific information and local knowledge to assist with decision making, which has not been done before in either region in a systematic way. SIF3 also recommends a broader suite of policy tools for regional bodies to use than the current reliance primarily on extension and incentives.

Demonstrated success at the regional level markedly increased our credibility with policy makers. Arguments based on the science alone would not have been so effective in influencing policy. As noted by Myšiak (2005), involving local stakeholders raises the legitimacy of policy decisions.

Despite the positive developments found in our work, and the recent Australian National Audit Office report (Auditor General, 2008), we suspect that achieving the large policy change needed if major NRM outcomes are to be achieved remains elusive. The State and Australian governments prefer to support a process of voluntary uptake by regional NRM bodies, rather than strongly recommending the use of a rigorous and transparent framework such as SIF3. The SIF frameworks in Western Australia, the asset-based approach promoted in Victoria and the New South Wales Natural Resources Commission “Standard” are encouraging, but in their current forms do not provide sufficient guidance to regional bodies. Funding to achieve the stated NRM targets is also inadequate and the application of frameworks such as SIF3 will serve to highlight this. We are concerned that regional bodies currently have little incentive to pursue NRM outcomes cost effectively, but we remain hopeful that stronger incentives to do so will be introduced eventually. In the meantime, a slow increase in the use of our framework is, in some respects, preferable, being less demanding and giving time to refine the framework and process for applying it.


SIF3 proved easy to apply and highly relevant to regional natural resource managers in Australia. It highlighted the need for substantial changes to existing investments in order to achieve high-value natural resource outcomes. It points to more targeted investment in localised assets, with different assets being prioritised when all relevant criteria are considered. It recognises a separate category of dispersed assets and highlights the need for different policy tools to be used relative to localised assets.

There has been excessive reliance on extension and small temporary grants to landholders, although less so in SCNRM than in NCCMA. In most parts of these regions, these tools are unlikely to result in enduring changes in land management of sufficient scale to achieve worthwhile salinity outcomes, although they may achieve other outcomes in human capacity and business management skills. The SIF3 analysis points to greater reliance on engineering, larger grants provided in a more targeted way, regulation, and technology development to protect assets from salinity.

The success of these pilots owed much to the strategies employed, including strong communication, collaboration, consultation, fostering local champions for the work, and being responsive to the needs and time lines of the two regional bodies. We developed strong credibility and became trusted partners. We formed close links to government bodies providing funding and guidance to the regions. Influence on policy makers and advisers has only partially succeeded to date.

Future work will include: working with other interested regional bodies to improve their salinity investment strategies; broadening SIF3 beyond salinity to encompass water quality problems, processes that threaten remaining biodiversity, and environmental pests; and continuing to influence state and national policy.


We particularly thank the staff, community members and the Board of the NCCMA and SCNRM regions. In particular Geoff Park, Jennifer Alexander, Malory Weston, Gavin Hanlon, Jock Leishman and Ian MacBean were vital supporters in the North Central region. Lindsay Ezard in particular, and other extension staff, provided historical, institutional and local knowledge. Shayne Annett and Adam Hood of DSE were also strongly supportive and actively engaged. In Western Australia, there were also many contributors, notably including Rob Edkins, Gary English, Ruhi Ferdowsian, John Simons, Alex Campbell, Louise Hillman and John Blake. We thank the Cooperative Research Centre (CRC) for Plant-Based Management of Dryland Salinity (now Future Farm Industries CRC), the Cooperative Venture for Capacity Building of Rural Industries Research and Development Corporation, the Australian Research Council, and the Department of Environment, Water, Heritage and the Arts (CERF program) for funding and support.


Anonymous, 2001. National Land andWater Resources Audit. Australian Dryland Salinity Assessment 2000. National Land and Water Resources Audit, Canberra.

Anonymous, 2002. Second generation dryland salinity plan for the North Central CMA region – Draft B. (Sinclair Knight Mertz)

Anonymous, 2005a. North Central River Health Strategy. North Central Catchment Management Authority, available at www.nccma.vic.gov.au , accessed 22 February 2008

Anonymous, 2005b. Standard for quality natural resource management. Natural Resources Commission. Document D05/5274 available at www.nrc.nsw.gov.au accessed 22 February 2008

Anonymous, 2006. Environment, Communications, Information Technology and the Arts References Committee. Living with Salinity - A Report on Progress. The Extent and Economic Impact of Salinity in Australia, Government of Australia, Canberra.

Anonymous, 2007. Catchment Condition Report. The State of Victoria, Victorian Catchment Management Council, October 2007. ISBN 978 1 74208 114 4 4 (available at www.vcmc.vic.gov.au, accessed 22 February 2008)

Auditor General, 2008, Regional delivery model for the Natural Heritage Trust and the National Action Plan for Salinity and Water Quality. The Australian National Audit Office. The Auditor-General Performance Audit Brochure Report No. 21 2007-08.

Carolan, M.S. 2005. Science, expertise and the democratization of the decision-making process. Society and Natural Resources 19, 661-668.

Curtis, A., Birckhead, J. and de Lacy, T. 1995. Community participation in Landcare policy in Australia: the Victorian experience with regional landcare plans. Society and Natural Resources 8, 415-430.

Ewing, S. 2003. Catchment Management Arrangements. Chapter 16. In: Dovers, S., Wild River, S. (Eds.) Managing Australia’s Environment. The Federation Press, Leichhardt, pp. 393-412.

George, R.J., McFarlane, D.J, Nulsen, R.A. 1997. Salinity threatens the viability of agriculture and ecosystems in Western Australia. Hydrogeology Journal 5(1) 6-21.

Gezelius, S.S. and Refsgaard, K. 2007. Barriers to rational decision-making in environmental planning. Land Use Policy 24, 338-348.

Gutrich, J., Donovan, D., Finucane, M., Focht, W., Hitzhusen, F., Manopimoke, S., McCauley, D., Norton, B., Sabatier, P., Salzman, J., Sasmitawidjaja, V. 2005. Science in the public process of ecosystem management: lessons from Hawai, Southeast Asia, Africa and the US Mainland. Journal of Environmental Management 76, 197-209.

Hatton, T.J., Nulsen, R.A. 1999. Towards achieving functional ecosystem mimicry with respect to water cycling in southern Australian agriculture. Agroforestry Systems 45, 203–214.

Henzell, T 2007. Australian agriculture, its history and challenges. CSIRO Publishing, Collingwood, 308pp.

Hobbs, R.J., Saunders, D.A. 1993. Reintegrating fragmented landscapes.Springer-Verlag, New York, 332pp.

McAlinden, D., Sparks, T., Burnside, D., Stelfox, L. 2003. Salinity Investment Framework Interim Report – Phase I, Report No. SLUI 32, Department of Environment, Perth.

Myers, N., Mittermier, R.A., Mittermier, C.G., da Fonseca, G.A.B., Kent, J. 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853-58.

Myšiak, J. 2005. Decision support systems for integrated water resource management: value and success factors. International Symposium on Environmental Software Systems. Hotel do Mar, Sesimbra, Portugal.

Norton, B.G., Noonan, D. 2007. Ecology and valuation: Big changes needed. Ecological Economics 63, 664-675.

Pannell, D.J. 2001. Salinity policy: A tale of fallacies, misconceptions and hidden assumptions. Agricultural Science 14(1), 35-37.

Pannell, D.J. 2004. Effectively communicating economics to policy makers. Australian Journal of Agricultural and Resource Economics 48(3), 535-555.

Pannell, D.J. 2008. Public benefits, private benefits, and policy intervention for land-use change for environmental benefits, Land Economics (in press).

Pannell, D.J., Marshall, G.R., Barr, N., Curtis, A., Vanclay, F., Wilkinson, R. 2006. Understanding and promoting adoption of conservation practices by rural landholders. Australian Journal of Experimental Agriculture 46(11), 1407-1424.

Pannell, D.J. and Roberts, A.M. 2009. Conducting and delivering integrated research to influence land-use policy: an Australian case study. Environmental Science and Policy (forthcoming).

Pielke, R.A. 2007. The honest broker, making sense of science in policy and politics. Cambridge University Press, New York.

Ridley, A.M., Mele, P.M., Beverly, C.R. 2004. Legume-based farming in southern Australia: developing sustainable systems to meet environmental challenges. Soil Biology and Biochemistry 36, 1213-1221.

Ridley, A.M., Seymour, E.J., Huhn, K.-J., Park, G. 2007. Priority environmental issues for monitoring – mismatch between farmers and catchment management perspectives. Australian Journal of Experimental Agriculture 47, 356-366.

Ridley, A.M., Pannell, D.J. 2005. The role of plants and plant-based research and development in managing dryland salinity in Australia. Australian Journal of Experimental Agriculture 45, 1-15.

Roberts, A.M., Helmers, M.J. and Fillery, I.R.P. 2009. The adoptability of perennial-based farming systems for hydrologic and salinity control in dryland farming systems in Australia and the United States of America. Crop and Pasture Science 60, 1-17.

Scott, B.J., Ridley, A.M., Conyers, M.K. 2000. Management of soil acidity in long-term pastures of south-eastern Australia: a review. Australian Journal of Experimental Agriculture 40, 1173-1198.

Seymour, E., Pannell, D., Ridley A., Marsh, S., Wilkinson, R. 2008. Understanding the capacity of catchment management organisations to make decisions about natural resource management (NRM) in Australia. Australasian Journal of Environmental Management (submitted)

Seymour, E.J., Ridley A.M. 2005. Toward Environmental Management Systems in Australian agriculture to achieve better environmental outcomes at the catchment scale. Environmental Management 35, 311-329.

Sparks, T., George, R., Wallace, K., Pannell, D., Burnside, D., Stelfox, L. 2006. Salinity Investment Framework Phase II, Report No. SLUI 34, Department of Water, Perth Western Australia, 86 pp.

Tress, G., Tress, B., Fry, G. 2007. Analysis of the barriers to integration in landscape research projects. Land Use Policy 24, 374-385.

Wilkinson, R. 2007. Social issues in asset-based management of dryland salinity: Case studies of commercial and lifestyle landholders in North Central Victoria and the South Coast of Western Australia. SIF3 Working Paper 0704, Future Farm Industries CRC, Perth. Available at http://dpannell.fnas.uwa.edu.au/sif30704.pdf, access date 22 February, 2008

Citation: Roberts, A.M. and Pannell, D.J. (2008). Piloting a systematic framework for public investment in regional natural resource management: dryland salinity in Australia, INFFER Working Paper 0808, University of Western Australia, Perth. (submitted to Land Use Policy) http://dpannell.fnas.uwa.edu.au/dp0806.htm

David Pannell home page

Copyright © Anna M Roberts and David J. Pannell, 2008
Last revised: June 16, 2013.