Landscape planning for biodiversity conservation in agricultural regions: A case study from the Wheatbelt of Western Australia
Biodiversity Technical Paper, No. 2
Robert J. Lambeck, CSIRO Division of Wildlife and Ecology
Commonwealth of Australia, 1999
ISBN 0 6422 1423 9
There is increasing recognition that many agricultural enterprises in Australia are not sustainable in their current form. Dry-land agriculture, based primarily on wheat and sheep, represents a prime example. While this industry has generated significant wealth for Australia, it has also created major problems in the form of land degradation and loss of biological diversity. If such industries are to be ecologically sustainable it will be necessary to halt processes of land degradation and loss of biodiversity. Because these processes cannot be managed by individual land-holders acting in isolation, it is essential that there be a shift from traditional farm-based, single-objective land management, to multiple-goal, sustainable management practices implemented at catchment or regional scales.
The success of multi-objective planning will depend on clear goal definition. If goals and objectives are not articulated at the outset, there will be no criteria for assessing the appropriateness of specified actions. Having set objectives, it is then necessary to identify, for each objective, the elements that are required in a landscape, the quantities or area required, their distribution in the landscape, and the appropriate management regime. If these requirements can be determined, they must then be combined in a way that addresses all objectives.
This report focuses primarily on the requirements for retaining biological diversity in situ in agricultural landscapes but, in recognition that the conservation of biological diversity will not be achieved if it is not part of a multi-objective planning process, procedures for integrating nature conservation with land-conservation and production are also considered.
The process of biotic impoverishment in agricultural landscapes has been well documented. Our capacity to reduce, halt or reverse these trajectories of decline will depend, in part, on the quality of information available and the time available to address the problems. Given that both the quality of information and the urgency of the problem vary from one location to another, it is necessary to develop strategies which are appropriate for the situation at hand.
Two broad types of response can be identified. The first of these, termed a 'general enhancement' approach employs ecological principles, derived from the scientific literature or from past experience in other locations, to provide general guidelines for action. This type of approach is appropriate where there is little information available for the landscape being managed and there is an urgent need to take some remedial action. The objective of a general enhancement approach will be to maximise the number of species retained in a landscape within the constraints of other land uses.
The second approach, termed 'strategic enhancement', aims to specify conservation targets and determine the actions required to meet those targets. Strategic enhancement requires a knowledge of the requirements of the biota in the landscape to be managed. This type of approach is relevant where such knowledge is available, or where the problem does not require immediate action and there is likely to be time and resources available to gather the necessary information. Strategic approaches can be further subdivided into two types of objectives. The first of these is to retain the biota, or specified components of the biota which presently occur in the landscape. This 'species retention' objective requires specification of the components of the biota which we wish to retain and an assessment of what is required to ensure their persistence. The other type of strategic objective is to reintroduce species that previously occurred in a landscape but no longer do so.
In reality, a management strategy for a particular region may be based on a combination of these approaches. However, it is important to distinguish between them as the information required, and the outcomes that can be expected, differ considerably.
Where rapid responses to biodiversity losses are required and the information available is poor, there will be little option but to rely on general ecological principles. However, managers must recognise the limitations of conservation planning based on these principles. Such approaches can only hope to increase the probability of species being retained, or conversely, to minimise the probability of species loss. For fragmented agricultural landscapes ecological principles will tell us that bigger remnants are generally better than smaller ones; that wider corridors are better than narrow ones; and that more connected landscapes are better than more fragmented ones. However, they will not tell us how much bigger, how much wider, or how much more connected these different landform attributes should be. An approach based on general principles can identify directions in which management should proceed, but cannot specify the magnitude of the response required, or what species will be retained or lost as a result of that response. It is likely that approaches based entirely on general principles would in most cases offer sub-optimal answers from both conservation and agricultural perspectives.
General principles are of limited value if we wish to retain individual species, nominated groups of species, or all species in a landscape. Such objectives will require the specification of what is required in a landscape to meet the needs of the species to be retained. This will require some level of understanding of the ecology of those species. While this type of objective has been routinely used in attempts to protect single endangered species, the use of species-based approaches to address broader conservation issues have had limited success. Such attempts have usually tried to identify 'umbrella' species or groups of such species whose needs, if met, will also meet the needs of an array of additional species. While this is undoubtedly valid in some cases, such strategies have not provided a procedure for determining which species are most likely to have an umbrella function, or what additional species will be protected or lost. A new approach to addressing this problem is presented in this report.
The third type of strategy is that of species reintroduction. The aim of such a strategy is to return to a landscape species which have previously occurred but have subsequently disappeared. This approach also requires a knowledge of the habitat and resource requirements of the species to be reintroduced. This knowledge will rarely be available for the region to which the species is to be reintroduced and hence must be acquired from where the species currently exists. However, patterns of local habitat use in a location where a species occurs will not necessarily be appropriate for the new location. Habitat suitability is influenced not only by local habitat attributes, but also by the regional context in which that habitat is located. Attempts to translocate species must therefore take into account both local and regional factors that influence the distribution and abundance of the target species.
In no situation will it be possible to have a total understanding of the requirements of all of the biota. Consequently, management will always represent a compromise between decisions based on good information for the area being managed and general guidelines derived from elsewhere. In any management situation, it is appropriate to strive for the best possible understanding of the biota, but to incorporate general principles when the appropriate information is not attainable. It must be recognised, however, that the greater the reliance on general principles the less certainty there will be that the desired outcomes will be achieved and the less cost-effective the solution is likely to be.
The objective of the current study was to develop a procedure for retaining the biota that currently occurs in the study area. A general enhancement approach aimed at simply maximising the number of species retained within the constraints imposed by other land uses was considered to be inadequate because such an approach is unable to specify the magnitude of the response required or which species were likely to be retained or lost. The third type of objective (species reintroduction) was beyond the scope of the current study and therefore was not considered an option. Consequently, a different approach was required in order to address our objective of keeping all of the native species that currently occur in the region. This report presents a new procedure for identifying the compositional and management requirements for meeting the needs of all of the biota that currently occurs in a landscape. The approach attempts, primarily, to identify the minimum requirements of the flora and fauna in the landscape but also draws on general principles in those situations where our knowledge of these needs is inadequate.
The procedure used to address this objective is termed a 'focal species' approach (Lambeck 1997). It builds on the concept of umbrella species, but differs from previous approaches by focussing on threatening processes and identifying for each threat the species that are most sensitive. Threatening processes may include habitat fragmentation, predation by feral animals, weed invasion and inappropriate fire regimes. For each threatening process, the most sensitive or demanding species is used to define the parameters within which the process must be managed. For example, area-limited species are used to define the minimum area required for different patch types; dispersal-limited species define the configuration of patches and the characteristics of connecting vegetation; and 'process-limited' species define the appropriate rates or limits of ecological processes such as fire, predation and weed invasion. The procedure therefore identifies a limited suite of 'focal species' whose requirements define different attributes that must be present in a functional landscape. If all threatening process are considered and the most sensitive or demanding species are identified, a landscape designed and managed according to their needs should also meet the needs of all other species.
While based on the needs of single species, the procedure reduces the number of species to be considered to a manageable set and provides clear criteria for selecting those species. The importance of this approach is that it provides a bridge between the sometimes divergent fields of species ecology and landscape ecology and uses the combination of these to generate implementable landscape designs and management guidelines.
A limitation of this approach is its focus on the most vulnerable species in the landscape. Such species tend to be relatively uncommon and hence are the most difficult species for which to obtain statistically robust data. Consequently, it is often necessary to rely on 'expert opinion', the reliability of which is not easily verified. In recognition of the limitations of the available data, it is necessary to implement monitoring strategies which are capable of testing the assumptions that must be made when using subjective information.
The focal species approach has the potential to generate explicit design and management guidelines for nature conservation. In this case study, sufficient information was available to specify the patch types required, their size and their configuration. The approach was less able to determine management regimes because data had not previously been gathered with a view to meeting the needs of this type of approach. However, some recommendations regarding predator control and fire frequencies can be made. Recommendations for other management regimes were derived from general principles where these existed.
When applied to a case study area in the wheatbelt of Western Australia, this procedure identified 61 out of a total of 113 habitat patches as currently being too small to support the most area-limited species. The enhancement of all 61 of these patches to a size which exceeds the minimum requirements of the most area-demanding species would require a 4.3% increase in the area allocated to nature conservation. When added to the existing remnant vegetation the total area designated to nature conservation would increase from the current 7.4% to a total of 11.7% of the catchment. Given that the area requirements of the focal species encompass those of all other species that use the same patch types, all other species which are currently limited by habitat availability would also benefit from these actions.
Revegetation of this type and magnitude would not be inconsistent with the expectations of the land-holders in the catchment, given firstly, their perception of a need for significant revegetation to manage hydrology and secondly, their expectation that revegetation need not be commercially viable in its own right. The implementation of such results is therefore highly feasible, particularly in the context of the catchment being identified as a focus catchment under the State Government Salinity Action Strategy.
It is highly unlikely that the use of general ecological principles could generate as efficient a conservation outcome with an equivalent increase in the area under native vegetation. General principles would not be able to specify how that 4% revegetation should be preferentially distributed and would most likely result in effort either being spread too thinly across the landscape, or being over-applied in too few locations. It appears from this study that the use of general principles for landscape planning is likely to generate inefficient solutions that will require greater effort and expense without any clear expectation of what will be achieved. Hence, strategies based on such principles will be less likely to be implemented by private land holders at a scale that is sufficient to meet the needs of the species that are currently at risk.
With the development of the focal species approach, it will be possible to focus further scientific investigation towards identifying the species which are most sensitive to other threatening process and determining acceptable levels for those threats. By specifying clear objectives and providing a framework for addressing those objectives, it is possible to develop much more effective research strategies which focus on obtaining the critical data that are required to address the problem at hand.
While the primary results of this study refer to spatial parameters defined by area-limited species, it is possible that an examination of the spatial requirements for managing other threats may exceed the values presented here. For example, if the use of fire to produce patchy burns and provide opportunities for recolonisation requires areas greater than those required by the most area-limited species, then the minimum sizes may turn out to be greater than those specified in this report. If this turns out to be the case, fire will become the focal 'species' for specifying the minimum area requirements of the different patch types. This information is not currently available. The acquisition of additional information about the requirements for addressing the needs of focal species for other threats, such as predation, would be unlikely to alter the spatial recommendation that this study has produced.
It is important to recognise that planning for nature conservation at a sub-catchment scale, as was done in this project, may not be sufficient to guarantee the long-term persistence of all species in the sub-catchment unless it is part of a broader, regional strategy. The solution produced in this study could be considered 'adequate' in that it meets the immediate needs of the species present. However, it may not be a 'viable' solution because it has not been applied at a large enough scale to support viable populations of all species. Failure of species to persist in the catchment will not be due to inadequacies in the study catchment but will be due to the poor quality of surrounding catchments. It is therefore necessary to determine the spatial scale at which a solution will not only meet the immediate needs of the species present, but will also ensure their long term persistence. Such an outcome is much more likely to be achieved if planning is conducted at a bioregional scale.
The application of these procedures at a larger scale will enable an assessment of the extent to which planning at larger scales reduces local costs and effort. Regional planning will identify a greater array of options for implementing a strategy and hence has the potential to reduce the contribution required by any individual land-holder and will therefore increase the likelihood of their participation.
The focal-species approach does not provide a template that can be applied to all agricultural landscapes. Instead, it provides an efficient procedure for identifying what is required to address the conservation priorities in any particular landscape. When the solution derived from the case study area was assessed for its suitability in another area with a similar local landscape configuration, it was immediately apparent that patterns of patch occupancy by focal species differed markedly between locations. This difference appeared to be attributable to the fact that the two sites were placed in very different regional contexts. In one location, the pattern of fragmentation was uniform throughout the region, whereas in the other, a nearby expanse of extensive bushland appeared to influence patterns of patch occupancy in the surrounding region. Regional factors must therefore be taken into account when considering local conservation plans.
The fact that a conservation solution derived from one location may not be appropriate for other areas is due to a number of factors including differences in vegetation patterns, landscape configurations and species composition between regions. For this approach to have broader relevance, it will be necessary to conduct the analysis at a larger regional scale, or to determine the circumstances under which a solution developed in one location can be applied legitimately to another. The latter approach will require the identification of catchments or sub-catchments that are not only broadly similar in their biotic, climatic and topographic attributes but which are also similar in terms of their clearing patterns and clearing history. A framework for developing such an approach is presented in Appendix 1.
Attempts to retain the biota in agricultural regions will invariably require the re-allocation of land from production to conservation uses. Land-holders will be reluctant to allocate more land than is absolutely necessary to meet a conservation outcome. Given the ever present competition for a limited land resource it will be necessary to specify the minimum requirements needed to meet a conservation goal. It will therefore be essential to develop methods which identify the minimum area, composition, configuration and management regimes needed to meet that goal. Approaches such as the focal species approach presented here begin to provide this precision. General principles, on the other hand, are unable to specify the magnitude of the response required. If nature conservation in production landscapes is to achieve more than simply improving on current conditions in some unspecified manner, then it will be necessary to further develop strategic planning procedures and to identify the appropriate scales for their application.
Attempts to incorporate multiple objectives into farm planning, combined with shifts to catchment-based management, have created an urgent need to be able to assess the implications of adopting alternative management strategies.
If farmers are to seriously consider addressing nature conservation issues, they will need to know what actions must be taken, where they need to be implemented, and how much of the landscape is required. Most importantly, they need to know the implications of the specified actions for agricultural production.
Decision support tools can play a useful role in facilitating a planning process that attempts to address this complexity. The primary value of using a formal decision support process is that it brings together the various stakeholders and encourages the articulation of important issues and objectives. In the current case-study stakeholders were primarily land-holders and State Government conservation agencies. In larger scale planning exercises the array of stakeholders may increase to include various levels of government and other parties such as aboriginal groups and mining companies. It is therefore essential that the planning process is transparent, comprehensible by the users, enables all participants to assess the performance of their interests, and allows them to explore alternative options and assess the implications of adopting alternative strategies. Where multiple stakeholders with potentially conflicting interests are involved, the procedure may require facilitation by an independent agent.
A software package (LUPIS – Land Use Planning and Information System) was used in this exercise to develop a range of alternative strategies that placed different levels of emphasis on nature conservation, land conservation and production actions. By doing this it was possible to quantify the amount of land required to meet the specified objectives and to determine the impact on all objectives of changing the relative emphasis on particular objectives. The procedure does not generate a single plan, but rather, presents a range of alternatives. Convergence towards a mutually acceptable plan is achieved through a series of iterations that better reflect the social preferences attached by the participants to different objectives.
Integrated management has the additional benefit of identifying opportunities for positive interactions between biodiversity and land-conservation measures. For example, comparison of the different land uses which contribute to nature conservation, agriculture and hydrological management enabled the identification of new combinations of land uses which address two or more objectives simultaneously. For example, the requirement for dense plantings of perennial species on some recharge areas to manage hydrology can be combined with a nature conservation objective by selecting native plants which provide resources required by the native biota. Alternatively, hydrological and production goals can be merged by planting timber or oil producing tree species in areas where they contribute most to managing recharge. Such strategies can bring further benefits, including shelter for stock and reduced wind and water erosion.
The retention of biological diversity in agricultural landscapes requires that biodiversity protection is explicitly recognised as an objective for ecologically sustainable land management. The retention of species, as opposed to the minimisation of species loss, will require more precise management strategies than can be provided by general ecological principles. The focal species approach developed in this study provides a mechanism for determining the compositional, spatial and management requirements for ensuring that the needs of the biota will be met in any specified landscape.
The project has demonstrated that solutions based on the needs of the resident biota can be extremely efficient when actions are directed towards identifying and ameliorating the threatening processes that limit population viability. The spatial requirements for providing adequate habitat for the most area-demanding species in this case study were surprisingly efficient, with a 4% increase in the area of native vegetation (from 7 to 11% of the catchment) creating 61 additional habitat patches that would be sufficiently large to support these area-limited species.
Such efficiency is unlikely to be attainable through the use of general principles alone. It is therefore necessary to ensure that conservation objectives are clearly stated and that the knowledge required for addressing those objectives is acquired. However, in recognition that in many situations there is an urgent need to act and much of the necessary data for a comprehensive strategic approach will not be readily available, it may be necessary to use general principles to guide conservation actions. It must be recognised however that the use of general principles diminishes the capacity to provide the explicit solutions that are required by landholders who legitimately strive to maximise the area available for production.
A limitation of the current exercise was the spatial scale at which it was applied. While production objectives may be usefully addressed at farm or even sub-catchment scales, many ecological and hydrological processes operate at larger scales. It is therefore necessary to develop a planning process that is based on natural bioregions but which also takes into account human-induced landscape influences such as the pattern and history of land clearing. This regional planning should aim to provide a context for developing local plans and should clearly identify the types of actions that local groups could implement in order to contribute to meeting the regional objectives.
That the retention of biodiversity be explicitly recognised as a goal in Commonwealth, State/Territory, regional and local development programs and that, at regional and local levels, this goal be clearly expressed in terms of general or strategic enhancement objectives.
That all Governments recognise the importance of facilitating and resourcing the retention of biodiversity on privately owned production lands.
That planning for biodiversity conservation in production landscapes be based on natural bioregions that also reflect temporal and spatial patterns of human impact.
That species' requirements be recognised as the primary basis for conservation projects which aim to retain the biota at local and regional scales in agricultural landscapes. General principles should only be employed where urgent decisions need to be made and the appropriate information cannot be acquired in time, or when the objective is to maximise the number of species retained without regard to the precise number or type of species to be kept or lost.
That species-based conservation projects focus on species whose protection can be expected to deliver significant and identifiable benefits to other components of the biota in the region being managed unless a species has high conservation value in its own right.
That bioregional-scale projects be initiated which incorporate biodiversity into production landscapes using the focal-species approach. These projects should be based on regionalisations which identify areas that are homogeneous in terms of natural biophysical patterns as well as anthropogenic landscape patterns. These projects should evaluate the efficiencies that accrue from regional-scale planning and should incorporate comprehensive economic assessments of integrated management solutions which consider long-term, off-site and non-monetary costs and benefits.
That monitoring of biodiversity in at least one bioregion be carried out during and subsequent to implementation of a bioregional scale project to test the assumption that landscapes designed to ensure viability of focal species also ensure viability of non-focal taxa.
That consideration be given to developing standards for spatial data acquisition for regional planning exercises to ensure consistent data quality throughout the planning area.