- Email: [email protected]
Systems framework for regional-scale integrated modelling and assessment
Mathematics and Computers in Simulation 64 (2004) 41–51
Systems framework for regional-scale integrated modelling and assessment R. Greiner∗ CSIRO Sustainable Ecosystems, Davies Laboratory, PMB Aitkenvale, Townsville 4814, Qld, Australia
Abstract Computer-based methods of integrated modelling and assessment provide an important means for reviewing policy choices in natural resource management (NRM). Research in support of NRM needs to address a wide range of issues involved, from point-scale biophysical, to business-scale human, to regional-scale planning issues. Research covering the full scope of such issues is by default multi-disciplinary and integrative and therefore analytically, methodologically and operationally challenging. The recently initiated Ord–Bonaparte Program is an example of a research and development program attempting to achieve both levels of integration in an applied NRM context. One key requirement for the success of the program lies in developing a systems framework that: (i) enables the integration of the various disciplinary research activities; and (ii) facilitates the implementation of research outputs by making integrated science relevant to decision-makers and translating new knowledge into outcomes for sustainable regional development. This paper proposes an approaches for such a systems framework. © 2003 IMACS. Published by Elsevier B.V. All rights reserved. Keywords: Integrated modelling and assessment; Systems framework; Natural resource management; Participatory research
1. Introduction Computer models can provide powerful tools for informing decision-making in policy and planning processes. Their usefulness is multiple: (1) models facilitate the systematic analysis of a topic by defining and implementing a conceptual framework; (2) computer models can integrate data and knowledge efficiently, specifically for complex problems, and provide reliable and repeatable projections or predictions of the future; and (3) models are powerful communication tools if used to explain and delineate system interactions. Models that analyse dynamic behaviour of multiple issues and include multiple disciplines are commonly referred to as integrated models [1]. Decisions affecting the style and direction of natural resource management (NRM) are embedded in a multi-dimensional context, requiring understanding of point-scale biophysical, business-scale human ∗
Tel.: +61-7-4753-8630; fax: +61-7-4753-8650. E-mail address: [email protected] (R. Greiner). 0378-4754/$30.00 © 2003 IMACS. Published by Elsevier B.V. All rights reserved. doi:10.1016/S0378-4754(03)00119-8
42
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
and regional-scale planning issues. Integrated models that comprise biophysical, ecological, economic and social system components are ideally suited to investigating the expected performance of alternative policies or development pathways against economic, social and ecological objectives, and providing risk and uncertainty analysis. It is increasingly recognised that integrated modelling has to be accompanied by a process of researchers and modellers working with decision-makers in a participatory manner to lead to research implementation [2]. This paper adds to the debate of integrated modelling and assessment (IMA) concepts and methodology in three ways. It adds interpretation to the ‘integration’ terminology. It demonstrates how integrated modelling and assessment can be organisationally entrenched in a research and development (R&D) program. And it presents a theoretical framework for an IMA approach to support decision-making for ecologically sustainable regional development, outlining key requirements for conceptualising and implementing effective IMA. It elaborates on how integration in general, and modelling in particular, can inform decision-making for ecologically sustainable development (ESD).
2. Integrated modelling and assessment There is increasing demand for integration in areas of environmental impact assessment of resource use and the assessment of trade-offs among stakeholders for the effects of water and land use decision in catchments and regions. This is reflected in an increasing number of integrated studies and models, with the bulk of applications related to water management [3,4] and assessment of climate change [5,6]. The use of the term ‘integration’ in the modelling literature can relate to one or several aspects including: the integration of issues, disciplines, methods, models, scales of consideration, and stakeholder concerns. There seem to be different connotations and terms such as ‘integration’, ‘integrated assessment’, ‘integrated modelling’ or ‘systems analysis’, which are used to various degrees to be complementary or synonymous. Neither term represents standard methods or models for integration, rather a wide variety of approaches have been used for different applications. The concept of integrated modelling is well developed. For example, the handbooks by Miser and Quade [7,8] contain a spectrum of approaches to and applications of integrated or systems analysis. In the modelling literature, integrated modelling usually refers to an approach whereby models from different scientific disciplines are coupled, taking into account the major internal dynamics of a system, their interactions and the relevant external forces [9]. Essentially, ‘systems analysis’ and ‘integrated modelling’ may be seen as relating to the understanding of how complex systems function. ‘Integrated assessment’ tends to apply and interpret the knowledge provided by integrated models within a specific context, presenting a bridge from science to the policy domain [10]. The concept of integrated assessment is fuzzy and multiple definitions exist [11]. The editors of Integrated Assessment, a journal launched by Kluwer Academic Publishers in 2000,1 define integrated assessment as ‘the practice of combining different strands of knowledge to accurately represent and analyse real world problems of interest to decision-makers’ [12]. They go on to explain that integrated assessment aims at conveying innovative and sometimes counterintuitive insights into issues. This paper uses the term ‘integrated modelling and assessment’ to provide equal emphasis of modelling and participatory process aspects. IMA is about clarifying multi-dimensional problems and supporting 1
The journal was discontinued in 2001.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
43
rational decision-making in the face of complexity by enabling the comparison of alternative options in the light of their possible outcomes. 3. R&D to support sustainable development 3.1. Geographical context and issues for ecologically sustainable development The ecosystems of tropical northern Australia have seen little modification over the 200 years since European settlement. Tropical northern Australia covers an area of approximately 3 million km2 of land—or 38% of Australia’s surface area. It is rich in natural resources, including minerals, land and water as well as terrestrial and marine biodiversity. The region is sparsely populated with only 650,000 people, equivalent to 3.7% of Australia’s population, and a typical inland population density of 0.08 persons/km2 . However, there is increasing awareness of these resources, improved technology to utilise them and subsequently competition for access to these resources [13]. Development of natural resource use in the southern parts of Australia has happened largely unregulated and in the absence of scientific data and understanding. Various resource degradation problems, including land and water salinity and biodiversity decline, have emerged at a large scale as a direct consequence. Northern Australia has the opportunity to learn from the lessons of the south and develop its resources in a long-term sustainable manner. Development options need to be subjected to considerations of ecologically sustainable development. ESD in essence means that economic objective of development must not trade-off social objectives, such as non-market values and intergenerational equity, and ecological objectives, such as safeguarding ecosystem function. IMA plays an important role in achieving ESD in northern Australia. Commercial development opportunities and proposals in northern Australia include expansion and intensification of grazing, new irrigation, aquaculture, fishing, mining and tourism projects. Indigenous cultural uses and conservation are important land uses. Individual development proposals are subject to project-specific social and environmental impact assessment but there is no scrutiny of indirect, compounded and long-term implications specifically in the ecological and social domain and broader economic context. 3.2. Concept of an integrated regional-scale R&D program During 1999–2000, a R&D program was set up to support long-term sustainable natural resource management for the Ord–Bonaparte region, which is part of the Kimberley region in north-western Australia. This R&D initiative is entitled the Ord–Bonaparte Program (OBP) [14]. Land and Water Australia (LWA) and the Commonwealth Scientific and Industrial Research Organisation (CSIRO) are the major funding agencies. The aim of the OBP is to underpin NRM, planning and policy with the knowledge of how natural and human systems function and inter-relate at the catchment and regional-scales. The Ord–Bonaparte region was chosen as ‘case study region’ because of its diversity and significance of natural resource-based industries, the renewed natural resource development interests in the region, the diverse social fabric of the regional community, and the capacity of the region to implement research outcomes through existing planning and management processes. The R&D program is philosophically, operationally and methodologically grounded in the concepts of integrated modelling and assessment. The science provided within the R&D approach is embedded in a
44
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
partnership approach with stakeholders. It aligns with a paradigm shift in NRM away from ‘autocratic’ or ‘interactive natural science based’ NRM towards ‘collaborative natural- and social-science-based management” [15]. If one presumes that a lack of information is a key factor in bad development decisions, then sound data and knowledge are the necessary condition for sustainable development. A series of sufficient conditions have to be fulfilled so that information actually translates into sustainable development. They include [13]: • Effective databases containing baseline and monitoring data in biophysical, ecological, social and economic domains. • Effective tools for data interrogation such as decision-support systems and geographic information systems. • Synthesis and integration of the information. • Analysis of resource management and planning institutions and processes. Individual projects within the OBP investigate a wide range of NRM, planning and policy issues that reflect the various natural resources and their uses and related issues. This work is undertaken in partnership with a diverse range of stakeholders and the research is negotiated with stakeholders. Unlike many other R&D initiatives, the OBP set out to develop an integrated systems framework a priori, to help define individual projects and enable their seamless conceptual and methodological integration. 3.3. Structure of the R&D program The systems analysis and integration philosophy is reflected in the structure of the OBP. The program is made up of five ‘sub-programs’. Three of these contain a suite of research activities centred around key natural resources and their management. One deals specifically with issues related to indigenous contributions to resource management. The fifth is entirely dedicated to various aspects of integration. A representation of the program structure is provided in Fig. 1. To further entrench integration within the program structure and provide methodological focus and consistency across individual research projects, nine research ‘themes’ cut across the five sub-programs. Subprogram 1 Regional Resource Futures
Subprogram 3
Subprogram 2
Water resource planning & management
Rangeland systems Subprogram 4
Subprogram 5 Aboriginal management & planning for ‘country’
Coastal, estuarine and marine resources
Fig. 1. Organisation of the Ord–Bonaparte R&D Program.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
45
“Regional Resource Futures” subprogram project 1
project 2
project 3
Community Resource Information Centre
Regional Futures
Program Evaluation
↔ ↔ ↔ ↔
Assemble, integrate and manage data set;
← ← ← ←
Extract, visualise and disseminate data
← ← ← ←
Set up RIC for ongoing capacity for inquiry and equitable access
↔ ↔ Links from/to OBP ↔ subprograms 2-5 ↔
↔ ↔ ↔ ↔
Common understanding of regional system function / interaction
Scope research needs / stakeholder aspirations ↔ ↔ ↔ ↔
Build interactive tools for regional analysis
↔ ↔ ↔ ↔
↔ ↔ ↔ ↔
→ → → →
Framework for evaluation and feedback Ongoing monitoring / evaluation Synthesis of experience
Explore options for future management / planning and policy
Fig. 2. Research projects within the “regional resource futures” sub-program.
The themes are called ‘biophysical resource inventory’, ‘process understanding’, ‘socio-economic data and understanding’, ‘synthesis and integration’, ‘participation’, ‘institutions’, ‘capacity’, ‘indigenous social and cultural issues’ and ‘monitoring and evaluation’. The ‘regional resource futures’ sub-program provides the hub for integration. Its role is to develop methods and tools for integrated regional analysis and program evaluation by bringing together issues-based and disciplinary research undertaken in the other sub-programs in an interpretative and participatory manner. To fulfil its role, the sub-program is structured into three projects (Fig. 2). Each project has specific integrative components: 1. The ‘community resource information system’ project seeks to establish an easily accessible and user-friendly repository of data relevant to NRM, i.e. data already existing and new data gathered by or generated across the OBP. 2. The ‘regional futures’ project proposes to support the conceptual and quantitative integration of data and process and impact understanding. This is to be done through the design, implementation and application of a series of interconnected qualitative and numerical models. The technical aspects of the work are embedded in a consultative and capacity building process with stakeholders. The following sections of this paper reveal details of this approach. 3. The ‘evaluation’ project assesses the achievements of all projects within the OBP and the program itself against milestones and objectives and distils ‘learnings’ from the case study that can inform future integrated R&D efforts. 3.4. Catchment-based systems definition The OBP aims to address integrated natural resource planning and management issues and as such its natural spatial reference is of bioregional definition, comprising catchments and their coastal and marine impact areas. The Ord catchment is about 80,000 km2 . The hydrological catchment boundaries
46
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
are not congruent with what can be called the limits of the ‘social catchment’ relevant to the study, characterised by administrative boundaries and the social, cultural and economic activities and linkages of the approximately 12,000 people living in this region. The area of the social catchment is about three times the size of the actual Ord River catchment, extending approximately 300 km east–west and 750 km north–south, and covering some 250,000 km2 of land. The catchment focus of the OBP is intended to provide a systems context for natural resource management. In a biophysical sense, this means considering flows from the top of the catchment right through to the coastal and marine environment. Questions of relevance here include, e.g. the effects of changes to grazing regimes in the upper parts of the catchment (around Halls Creek) to surface water run-off and sediment loads, with potential down-stream effects on irrigation-water supply, the lower part of Ord River and adjoining coastal and marine environments of the Cambridge and Josef Bonaparte Gulfs. In a socio-economic sense, the regional focus enables an involvement of all relevant stakeholders— including the large aboriginal population, an assessment of management impacts beyond the immediate user groups or industries implementing changes. Including flow-on and indirect effects provides the opportunity for a comprehensive assessment of alternative management and development options. It also allows for institutional arrangements such as government boundaries and agency responsibilities to be analysed in the NRM context. Natural resource management at the regional-scale provides a complex set of challenges for decisionmakers as it encompasses several dimensions or layers of complexity: • There are multiple natural resources to consider including land, fresh water, landscape, terrestrial, coastal and marine biodiversity. • There are multiple uses of these resources through grazing, tourism, recreation, aboriginal land uses, irrigation, aquaculture and fishing. Uses may be complementary or competing for resources. • There are a series of natural resource management and resource development agencies at federal, state and regional level with statutory obligations for the management of natural resources in the region. Their interests and responsibilities may be complementary, overlapping or conflicting. • There is a wide range of interests and aspirations within the regional community to be considered, specifically those of a large aboriginal population, as well as interests of the people of Australian at large. • Issues of risk management and uncertainty arise in the face of various external risk factors, such as international markets and climate change. These dimensions define natural resource management as a complex system. A ‘systems approach’ is needed to track the web of causal relationships within the system and offer solutions to systemic problems. Systems analysis, based on general systems theory [16], offers a way of conceptualising and analysing the complexity of natural resource management by looking at the entirety of a problem in its specific context with all its ‘hard’ scientific and ‘soft’ social dimensions and doing this with the aid of models [17,18]. The benefits of applying a systems approach to environmental and natural resource management applications have been outlined comprehensively and compellingly [19,20].
4. Towards a conceptual model of regional system function This section outlines the participative–conceptual aspect of the integrated analysis proposed as part of the “regional futures” project in the “regional resource futures” sub-program. The objective is to build
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
47
a common understanding, or systems model, of regional system function. This includes a description of the character of the region—its people, industries, natural environments and institutions—and how the East Kimberley works—explaining inter-relationships between the region’s natural resources, their use, the regional economy and community. This qualitative work is complemented by analytical–numerical research, which is described in Section 5. 4.1. Requirement for a qualitative, conceptual regional model Understanding and knowledge, supported by appropriate tools, are essential requirements for managers, planners and decisions makers to make sustainable development happen. Engaging stakeholders and the community right from the beginning in conceptualising integration is part of the strategy to effect learning and common understanding and foster the application and use of tools such as models and expert systems in planning and decision-making. Developing a qualitative model of regional system function provides an important first step. It involves drawing a causal picture of: (i) how the community benefits from the various natural resource uses and how they inter-relate; and (ii) what the management and planning issues are and how decisions impact on various aspects of the system. The resulting agreed conceptual regional framework will: • guide the various disciplinary research activities across all five sub-programs of the OBP; • enable the disciplinary and issues-based research to be brought together in a consistent format for integrated analysis; • provide flexibility to accommodate improved scientific understanding as it becomes available. 4.2. Methodology The task of implementing an IMA approach requires an applied systems thinking [21]. We propose as core method focussed workshopping based on the workshop techniques central to ‘participatory action research’ [22] and ‘organisational learning’ [23]. A series of workshops is envisaged to bring together local experts, key OBP researchers and selected (inter)national experts in regional development. During the initial workshop the issues are being defined and supported by existing data. We hope to define a preliminary (qualitative) systems model upon which we base and prioritise information acquisition activities that provide an overarching integrative systems framework and will help guide the disciplinary research undertaken in other sub-programs. We regard this as pre-condition for successful cross-program model integration later in the program lifetime by ensuring that the models built across various OBP sub-programs and projects are consistent in terms of measurement units, scale and model type, and cover the social, economic and biophysical areas of the OBP. Importantly, that first workshop is also intended to act as a team-building exercise, which brings stakeholders and scientists together to a ‘problem-solving’ team [18]. The subsequent workshops will develop the regional systems model further based on data and knowledge forthcoming. The model is basically a communication interface between stakeholders and researchers all of whom have joint ownership of the model. As such the modelling process is more important than the model itself. The intention is to hold the workshops in Kununurra, the major urban and administrative
48
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
centre in the region and have at least the first workshop professionally facilitated to ensure equality of all participants.
5. Developing tools to enhance the integration, interpretation and delivery of research The participative–conceptual work outlined in Section 4 is complemented by the quantitative–analytical work described in this section. The purpose of this IMA approach is to develop a suite of models to facilitate stakeholder access to data and understanding of the Ord–Bonaparte region and enhance interpretation, integration and delivery of OBP research outcomes. We use the term ‘tools’, which has emerged as the stakeholder-preferred term, synonymous for ‘models’. 5.1. Tools for regional analysis and integration Models are essential in bringing the outputs of the disciplinary sub-programs together to deal with the complexity and provide the holistic understanding of the region that the OBP seeks. It is envisaged to develop a suite of tools, the specifications of which will be developed in consultation with stakeholders to maximise relevance and consequently research implementation. We seek to develop a suite of tools rather than one all-encompassing model as we expect that stakeholders want to investigate a wide range of issues and conduct analysis in various ways, and bring different levels of expertise to the investigation. While the exact specification of the tools are yet unknown, we expect that the tools will, broadly: • integrate various ecological and biophysical models from the disciplinary sub-programs within the overarching conceptual framework; • offer insights into the regional economy, its dependencies, trade-offs between different nature-based industries and natural resource uses, externalities and economic incentive instruments; • explore social response to change in both quantitative and qualitative terms. All models will be built into a ‘toolkit’ to facilitate their delivery and application [24]. 5.2. A preliminary structure for the toolkit The toolkit proposed here comprises four main components: a user interface, a model management system, a multi-objective modelling engine, and an expert system for interpretation of modelling outputs (Fig. 3). The model management system contains biophysical and socio-economic models developed in the OBP projects. It also provides functions for selecting, invoking, running, and integrating models. The multi-objective modelling engine is used to explore and evaluate different regional resource use scenarios through multi-objective optimisation or multi-criteria analysis techniques. After a scenario is defined, this modelling engine develops a mathematical programming model using the biophysical and socio-economic indicators generated from the relevant models in the toolkit to determine optimal resource allocation within biophysical constraints and socio-cultural preferences. It can also be used to develop a multi-criteria decision model that identifies trade-offs among regional resource use options based on stakeholder value judgements. This approach asserts that the integration of multiple scales, issues and objectives can be achieved through the integration of different modelling approaches into higher-level simulation models [25].
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
49
Fig. 3. Preliminary toolkit structure.
The expert system component provides explanation and interpretation of the models and model results. It shows assumptions, explains the modelling process, provides necessary guidelines for use of models, and interprets and translates information derived by the models to a certain format, which is desired by a certain group of stakeholders. The user interface provides users with controls for the implementation of the system. This may include managing scenario investigation and supporting selection of models or tools that seem most useful for an intended application. The toolkit evaluates the effects of the various scenarios on multiple indicators and the trade-offs among them according to multiple criteria and multiple objectives. This toolkit will be linked to the resource information system (Fig. 2), which does not only manage information and data required for modelling, but also provides tools for presenting modelling results in particular forms such as maps. 6. Conclusions This paper highlights the key role that integrated modelling and assessment plays in a new R&D program designed to support regional-scale NRM. Modelling provides the integrating framework across research in
50
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
various disciplines and across NRM issues related to various natural resources and associated industries. It also provides the link from research to real-life application of data and understanding by natural resource managers and planners. A toolkit approach is proposed to support this task and accommodate the variety of models and tools to be built during the lifetime of the program. The road to integration and integrated assessment is paved with many conceptual, methodological, technological and process challenges [26,27]. Thorough organisational and theoretical preparation and research implementation is required to make this a process that truly supports ecologically sustainable development. Acknowledgements The author is indebted to a large number of researchers and stakeholders who have, in the course of numerous meetings and conversations, contributed to the content of the OBP and therefore ultimately to this paper. References [1] J. Rotmans, M. Van Asselt, Integrated assessment: growing child on its way to maturity. An editorial essay, Climate Change 34 (1996) 327–336. [2] J.L.A. Geurts, C. Joldersman, Methodology for participatory policy analysis, Eur. J. Operational Res. 128 (2001) 300–310. [3] M.S. Krol, A. Jaeger, A. Bronstert, J. Krywkow, The semi-arid integrated model (SIM), a regional integrated model assessing water availability, vulnerability of ecosystems and society in NE-Brazil, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (7–8) (2001) 529–533. [4] H. Staudenrausch, W.A. Flügel, Development of an integrated water resources management system in southern African catchments, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (78) (2001) 561–564. [5] C. Pahl-Wostl, D. Schlumpf, M. Büssenschütt, A. Schönborn, J. Burse, Models at the interface between science and society: impacts and options, Integr. Assess. 1 (2000) 267–280. [6] J. Alcamo (Ed.), IMAGE 2.0: Integrated Modeling of Global Climate Change, Kluwer Academic Publishers, Dodrecht, 1994. [7] H.J. Miser, E.S. Quade (Eds.), Handbook of Systems Analysis—Overview of Use, Procedure, Applications, and Practices, Wiley, Chichester, 1985. [8] H.J. Miser, E.W. Quade (Eds.), Handbook of Systems Analysis—Craft Issues and Procedural Choices, Wiley, Chichester, 1988. [9] A. Bronstert, River basin research and management: integrated modelling and investigation of land-use impacts on the hydrological cycle, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (7–8) (2001) 485. [10] M. Hisschemoeller, R.S.J. Tol, P. Vellinga, The relevance of participatory approaches in integrated environmental assessment, Integr. Assess. 2 (2001) 57–72. [11] R. In’t Veld, Afterword, Integr. Assess. 2 (2001) 93–94. [12] H. Dowlatabadi, J. Rotmans, P. Martens, Integr. Assess. 1 (2000). [13] A.K.L. Johnson, S. Cowell, N. Loneragan, G. Dews, I. Poiner, Sustainable development of tropical Australia: R&D for management of land, water and marine resources, Occasional Paper 05/99, LWRRDC, Canberra, 1999. [14] CSIRO, Ord–Bonaparte Program: R&D Plan 2000–2005, CSIRO Tropical Agriculture, Brisbane, 1999. [15] S.A. Mullner, W.A. Hubert, T.A. Wesche, Evolving paradigms of landscape-scale renewable resource management in the United States, Environ. Sci. Policy 4 (2001) 39–49. [16] W.R. Ashby, An Introduction to Cybernetics, Wiley, New York, 1956. [17] P.B. Checkland, Systems Thinking, Systems Practice, Wiley, Chichester, 1981. [18] W.E. Grant, Ecology and natural resource management: reflections from a systems perspective, Ecol. Modell. 108 (1998) 67–76. [19] C.S. Holling, Adaptive Environmental Assessment and Management, Wiley, New York, 1978.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
51
[20] C.W. Walters, Adaptive Management of Renewable Resources, Macmillan, New York, 1986. [21] M.C. Jackson, Critical systems thinking and practice, Eur. J. Operational Res. 128 (2001) 233–244. [22] A. Martin, J. Sherington, Participatory research methods—implementation, effectiveness and institutional context, Agric. Syst. 55 (2) (1997) 195–216. [23] P. Senge, The Fifth Discipline, Doubleday, London, 1990. [24] R. Grenier, X. Zhu, Systems Framework for Intergrated Regional-Scale R & D Projects. Proceedings MODSIM (2001) 1201–1206. [25] F. Villa, R. Costanza, Design of multi-paradigm integrating modelling tools for ecological research, Environ. Modell. Software 15 (2000) 169–177. [26] R.M. Argent, R.B. Grayson, S.A. Ewing, Integrated models for environmental management: issues of process and design, Environ. Int. 25 (6–7) (1999) 693–699. [27] U. Dahinden, C. Querol, J. Jäger, M. Nilsson, Exploring the use of computer models in participatory integrated assessment— experiences and recommendations for further steps, Integr. Assess. 1 (2000) 253–266.
Systems framework for regional-scale integrated modelling and assessment R. Greiner∗ CSIRO Sustainable Ecosystems, Davies Laboratory, PMB Aitkenvale, Townsville 4814, Qld, Australia
Abstract Computer-based methods of integrated modelling and assessment provide an important means for reviewing policy choices in natural resource management (NRM). Research in support of NRM needs to address a wide range of issues involved, from point-scale biophysical, to business-scale human, to regional-scale planning issues. Research covering the full scope of such issues is by default multi-disciplinary and integrative and therefore analytically, methodologically and operationally challenging. The recently initiated Ord–Bonaparte Program is an example of a research and development program attempting to achieve both levels of integration in an applied NRM context. One key requirement for the success of the program lies in developing a systems framework that: (i) enables the integration of the various disciplinary research activities; and (ii) facilitates the implementation of research outputs by making integrated science relevant to decision-makers and translating new knowledge into outcomes for sustainable regional development. This paper proposes an approaches for such a systems framework. © 2003 IMACS. Published by Elsevier B.V. All rights reserved. Keywords: Integrated modelling and assessment; Systems framework; Natural resource management; Participatory research
1. Introduction Computer models can provide powerful tools for informing decision-making in policy and planning processes. Their usefulness is multiple: (1) models facilitate the systematic analysis of a topic by defining and implementing a conceptual framework; (2) computer models can integrate data and knowledge efficiently, specifically for complex problems, and provide reliable and repeatable projections or predictions of the future; and (3) models are powerful communication tools if used to explain and delineate system interactions. Models that analyse dynamic behaviour of multiple issues and include multiple disciplines are commonly referred to as integrated models [1]. Decisions affecting the style and direction of natural resource management (NRM) are embedded in a multi-dimensional context, requiring understanding of point-scale biophysical, business-scale human ∗
Tel.: +61-7-4753-8630; fax: +61-7-4753-8650. E-mail address: [email protected] (R. Greiner). 0378-4754/$30.00 © 2003 IMACS. Published by Elsevier B.V. All rights reserved. doi:10.1016/S0378-4754(03)00119-8
42
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
and regional-scale planning issues. Integrated models that comprise biophysical, ecological, economic and social system components are ideally suited to investigating the expected performance of alternative policies or development pathways against economic, social and ecological objectives, and providing risk and uncertainty analysis. It is increasingly recognised that integrated modelling has to be accompanied by a process of researchers and modellers working with decision-makers in a participatory manner to lead to research implementation [2]. This paper adds to the debate of integrated modelling and assessment (IMA) concepts and methodology in three ways. It adds interpretation to the ‘integration’ terminology. It demonstrates how integrated modelling and assessment can be organisationally entrenched in a research and development (R&D) program. And it presents a theoretical framework for an IMA approach to support decision-making for ecologically sustainable regional development, outlining key requirements for conceptualising and implementing effective IMA. It elaborates on how integration in general, and modelling in particular, can inform decision-making for ecologically sustainable development (ESD).
2. Integrated modelling and assessment There is increasing demand for integration in areas of environmental impact assessment of resource use and the assessment of trade-offs among stakeholders for the effects of water and land use decision in catchments and regions. This is reflected in an increasing number of integrated studies and models, with the bulk of applications related to water management [3,4] and assessment of climate change [5,6]. The use of the term ‘integration’ in the modelling literature can relate to one or several aspects including: the integration of issues, disciplines, methods, models, scales of consideration, and stakeholder concerns. There seem to be different connotations and terms such as ‘integration’, ‘integrated assessment’, ‘integrated modelling’ or ‘systems analysis’, which are used to various degrees to be complementary or synonymous. Neither term represents standard methods or models for integration, rather a wide variety of approaches have been used for different applications. The concept of integrated modelling is well developed. For example, the handbooks by Miser and Quade [7,8] contain a spectrum of approaches to and applications of integrated or systems analysis. In the modelling literature, integrated modelling usually refers to an approach whereby models from different scientific disciplines are coupled, taking into account the major internal dynamics of a system, their interactions and the relevant external forces [9]. Essentially, ‘systems analysis’ and ‘integrated modelling’ may be seen as relating to the understanding of how complex systems function. ‘Integrated assessment’ tends to apply and interpret the knowledge provided by integrated models within a specific context, presenting a bridge from science to the policy domain [10]. The concept of integrated assessment is fuzzy and multiple definitions exist [11]. The editors of Integrated Assessment, a journal launched by Kluwer Academic Publishers in 2000,1 define integrated assessment as ‘the practice of combining different strands of knowledge to accurately represent and analyse real world problems of interest to decision-makers’ [12]. They go on to explain that integrated assessment aims at conveying innovative and sometimes counterintuitive insights into issues. This paper uses the term ‘integrated modelling and assessment’ to provide equal emphasis of modelling and participatory process aspects. IMA is about clarifying multi-dimensional problems and supporting 1
The journal was discontinued in 2001.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
43
rational decision-making in the face of complexity by enabling the comparison of alternative options in the light of their possible outcomes. 3. R&D to support sustainable development 3.1. Geographical context and issues for ecologically sustainable development The ecosystems of tropical northern Australia have seen little modification over the 200 years since European settlement. Tropical northern Australia covers an area of approximately 3 million km2 of land—or 38% of Australia’s surface area. It is rich in natural resources, including minerals, land and water as well as terrestrial and marine biodiversity. The region is sparsely populated with only 650,000 people, equivalent to 3.7% of Australia’s population, and a typical inland population density of 0.08 persons/km2 . However, there is increasing awareness of these resources, improved technology to utilise them and subsequently competition for access to these resources [13]. Development of natural resource use in the southern parts of Australia has happened largely unregulated and in the absence of scientific data and understanding. Various resource degradation problems, including land and water salinity and biodiversity decline, have emerged at a large scale as a direct consequence. Northern Australia has the opportunity to learn from the lessons of the south and develop its resources in a long-term sustainable manner. Development options need to be subjected to considerations of ecologically sustainable development. ESD in essence means that economic objective of development must not trade-off social objectives, such as non-market values and intergenerational equity, and ecological objectives, such as safeguarding ecosystem function. IMA plays an important role in achieving ESD in northern Australia. Commercial development opportunities and proposals in northern Australia include expansion and intensification of grazing, new irrigation, aquaculture, fishing, mining and tourism projects. Indigenous cultural uses and conservation are important land uses. Individual development proposals are subject to project-specific social and environmental impact assessment but there is no scrutiny of indirect, compounded and long-term implications specifically in the ecological and social domain and broader economic context. 3.2. Concept of an integrated regional-scale R&D program During 1999–2000, a R&D program was set up to support long-term sustainable natural resource management for the Ord–Bonaparte region, which is part of the Kimberley region in north-western Australia. This R&D initiative is entitled the Ord–Bonaparte Program (OBP) [14]. Land and Water Australia (LWA) and the Commonwealth Scientific and Industrial Research Organisation (CSIRO) are the major funding agencies. The aim of the OBP is to underpin NRM, planning and policy with the knowledge of how natural and human systems function and inter-relate at the catchment and regional-scales. The Ord–Bonaparte region was chosen as ‘case study region’ because of its diversity and significance of natural resource-based industries, the renewed natural resource development interests in the region, the diverse social fabric of the regional community, and the capacity of the region to implement research outcomes through existing planning and management processes. The R&D program is philosophically, operationally and methodologically grounded in the concepts of integrated modelling and assessment. The science provided within the R&D approach is embedded in a
44
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
partnership approach with stakeholders. It aligns with a paradigm shift in NRM away from ‘autocratic’ or ‘interactive natural science based’ NRM towards ‘collaborative natural- and social-science-based management” [15]. If one presumes that a lack of information is a key factor in bad development decisions, then sound data and knowledge are the necessary condition for sustainable development. A series of sufficient conditions have to be fulfilled so that information actually translates into sustainable development. They include [13]: • Effective databases containing baseline and monitoring data in biophysical, ecological, social and economic domains. • Effective tools for data interrogation such as decision-support systems and geographic information systems. • Synthesis and integration of the information. • Analysis of resource management and planning institutions and processes. Individual projects within the OBP investigate a wide range of NRM, planning and policy issues that reflect the various natural resources and their uses and related issues. This work is undertaken in partnership with a diverse range of stakeholders and the research is negotiated with stakeholders. Unlike many other R&D initiatives, the OBP set out to develop an integrated systems framework a priori, to help define individual projects and enable their seamless conceptual and methodological integration. 3.3. Structure of the R&D program The systems analysis and integration philosophy is reflected in the structure of the OBP. The program is made up of five ‘sub-programs’. Three of these contain a suite of research activities centred around key natural resources and their management. One deals specifically with issues related to indigenous contributions to resource management. The fifth is entirely dedicated to various aspects of integration. A representation of the program structure is provided in Fig. 1. To further entrench integration within the program structure and provide methodological focus and consistency across individual research projects, nine research ‘themes’ cut across the five sub-programs. Subprogram 1 Regional Resource Futures
Subprogram 3
Subprogram 2
Water resource planning & management
Rangeland systems Subprogram 4
Subprogram 5 Aboriginal management & planning for ‘country’
Coastal, estuarine and marine resources
Fig. 1. Organisation of the Ord–Bonaparte R&D Program.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
45
“Regional Resource Futures” subprogram project 1
project 2
project 3
Community Resource Information Centre
Regional Futures
Program Evaluation
↔ ↔ ↔ ↔
Assemble, integrate and manage data set;
← ← ← ←
Extract, visualise and disseminate data
← ← ← ←
Set up RIC for ongoing capacity for inquiry and equitable access
↔ ↔ Links from/to OBP ↔ subprograms 2-5 ↔
↔ ↔ ↔ ↔
Common understanding of regional system function / interaction
Scope research needs / stakeholder aspirations ↔ ↔ ↔ ↔
Build interactive tools for regional analysis
↔ ↔ ↔ ↔
↔ ↔ ↔ ↔
→ → → →
Framework for evaluation and feedback Ongoing monitoring / evaluation Synthesis of experience
Explore options for future management / planning and policy
Fig. 2. Research projects within the “regional resource futures” sub-program.
The themes are called ‘biophysical resource inventory’, ‘process understanding’, ‘socio-economic data and understanding’, ‘synthesis and integration’, ‘participation’, ‘institutions’, ‘capacity’, ‘indigenous social and cultural issues’ and ‘monitoring and evaluation’. The ‘regional resource futures’ sub-program provides the hub for integration. Its role is to develop methods and tools for integrated regional analysis and program evaluation by bringing together issues-based and disciplinary research undertaken in the other sub-programs in an interpretative and participatory manner. To fulfil its role, the sub-program is structured into three projects (Fig. 2). Each project has specific integrative components: 1. The ‘community resource information system’ project seeks to establish an easily accessible and user-friendly repository of data relevant to NRM, i.e. data already existing and new data gathered by or generated across the OBP. 2. The ‘regional futures’ project proposes to support the conceptual and quantitative integration of data and process and impact understanding. This is to be done through the design, implementation and application of a series of interconnected qualitative and numerical models. The technical aspects of the work are embedded in a consultative and capacity building process with stakeholders. The following sections of this paper reveal details of this approach. 3. The ‘evaluation’ project assesses the achievements of all projects within the OBP and the program itself against milestones and objectives and distils ‘learnings’ from the case study that can inform future integrated R&D efforts. 3.4. Catchment-based systems definition The OBP aims to address integrated natural resource planning and management issues and as such its natural spatial reference is of bioregional definition, comprising catchments and their coastal and marine impact areas. The Ord catchment is about 80,000 km2 . The hydrological catchment boundaries
46
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
are not congruent with what can be called the limits of the ‘social catchment’ relevant to the study, characterised by administrative boundaries and the social, cultural and economic activities and linkages of the approximately 12,000 people living in this region. The area of the social catchment is about three times the size of the actual Ord River catchment, extending approximately 300 km east–west and 750 km north–south, and covering some 250,000 km2 of land. The catchment focus of the OBP is intended to provide a systems context for natural resource management. In a biophysical sense, this means considering flows from the top of the catchment right through to the coastal and marine environment. Questions of relevance here include, e.g. the effects of changes to grazing regimes in the upper parts of the catchment (around Halls Creek) to surface water run-off and sediment loads, with potential down-stream effects on irrigation-water supply, the lower part of Ord River and adjoining coastal and marine environments of the Cambridge and Josef Bonaparte Gulfs. In a socio-economic sense, the regional focus enables an involvement of all relevant stakeholders— including the large aboriginal population, an assessment of management impacts beyond the immediate user groups or industries implementing changes. Including flow-on and indirect effects provides the opportunity for a comprehensive assessment of alternative management and development options. It also allows for institutional arrangements such as government boundaries and agency responsibilities to be analysed in the NRM context. Natural resource management at the regional-scale provides a complex set of challenges for decisionmakers as it encompasses several dimensions or layers of complexity: • There are multiple natural resources to consider including land, fresh water, landscape, terrestrial, coastal and marine biodiversity. • There are multiple uses of these resources through grazing, tourism, recreation, aboriginal land uses, irrigation, aquaculture and fishing. Uses may be complementary or competing for resources. • There are a series of natural resource management and resource development agencies at federal, state and regional level with statutory obligations for the management of natural resources in the region. Their interests and responsibilities may be complementary, overlapping or conflicting. • There is a wide range of interests and aspirations within the regional community to be considered, specifically those of a large aboriginal population, as well as interests of the people of Australian at large. • Issues of risk management and uncertainty arise in the face of various external risk factors, such as international markets and climate change. These dimensions define natural resource management as a complex system. A ‘systems approach’ is needed to track the web of causal relationships within the system and offer solutions to systemic problems. Systems analysis, based on general systems theory [16], offers a way of conceptualising and analysing the complexity of natural resource management by looking at the entirety of a problem in its specific context with all its ‘hard’ scientific and ‘soft’ social dimensions and doing this with the aid of models [17,18]. The benefits of applying a systems approach to environmental and natural resource management applications have been outlined comprehensively and compellingly [19,20].
4. Towards a conceptual model of regional system function This section outlines the participative–conceptual aspect of the integrated analysis proposed as part of the “regional futures” project in the “regional resource futures” sub-program. The objective is to build
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
47
a common understanding, or systems model, of regional system function. This includes a description of the character of the region—its people, industries, natural environments and institutions—and how the East Kimberley works—explaining inter-relationships between the region’s natural resources, their use, the regional economy and community. This qualitative work is complemented by analytical–numerical research, which is described in Section 5. 4.1. Requirement for a qualitative, conceptual regional model Understanding and knowledge, supported by appropriate tools, are essential requirements for managers, planners and decisions makers to make sustainable development happen. Engaging stakeholders and the community right from the beginning in conceptualising integration is part of the strategy to effect learning and common understanding and foster the application and use of tools such as models and expert systems in planning and decision-making. Developing a qualitative model of regional system function provides an important first step. It involves drawing a causal picture of: (i) how the community benefits from the various natural resource uses and how they inter-relate; and (ii) what the management and planning issues are and how decisions impact on various aspects of the system. The resulting agreed conceptual regional framework will: • guide the various disciplinary research activities across all five sub-programs of the OBP; • enable the disciplinary and issues-based research to be brought together in a consistent format for integrated analysis; • provide flexibility to accommodate improved scientific understanding as it becomes available. 4.2. Methodology The task of implementing an IMA approach requires an applied systems thinking [21]. We propose as core method focussed workshopping based on the workshop techniques central to ‘participatory action research’ [22] and ‘organisational learning’ [23]. A series of workshops is envisaged to bring together local experts, key OBP researchers and selected (inter)national experts in regional development. During the initial workshop the issues are being defined and supported by existing data. We hope to define a preliminary (qualitative) systems model upon which we base and prioritise information acquisition activities that provide an overarching integrative systems framework and will help guide the disciplinary research undertaken in other sub-programs. We regard this as pre-condition for successful cross-program model integration later in the program lifetime by ensuring that the models built across various OBP sub-programs and projects are consistent in terms of measurement units, scale and model type, and cover the social, economic and biophysical areas of the OBP. Importantly, that first workshop is also intended to act as a team-building exercise, which brings stakeholders and scientists together to a ‘problem-solving’ team [18]. The subsequent workshops will develop the regional systems model further based on data and knowledge forthcoming. The model is basically a communication interface between stakeholders and researchers all of whom have joint ownership of the model. As such the modelling process is more important than the model itself. The intention is to hold the workshops in Kununurra, the major urban and administrative
48
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
centre in the region and have at least the first workshop professionally facilitated to ensure equality of all participants.
5. Developing tools to enhance the integration, interpretation and delivery of research The participative–conceptual work outlined in Section 4 is complemented by the quantitative–analytical work described in this section. The purpose of this IMA approach is to develop a suite of models to facilitate stakeholder access to data and understanding of the Ord–Bonaparte region and enhance interpretation, integration and delivery of OBP research outcomes. We use the term ‘tools’, which has emerged as the stakeholder-preferred term, synonymous for ‘models’. 5.1. Tools for regional analysis and integration Models are essential in bringing the outputs of the disciplinary sub-programs together to deal with the complexity and provide the holistic understanding of the region that the OBP seeks. It is envisaged to develop a suite of tools, the specifications of which will be developed in consultation with stakeholders to maximise relevance and consequently research implementation. We seek to develop a suite of tools rather than one all-encompassing model as we expect that stakeholders want to investigate a wide range of issues and conduct analysis in various ways, and bring different levels of expertise to the investigation. While the exact specification of the tools are yet unknown, we expect that the tools will, broadly: • integrate various ecological and biophysical models from the disciplinary sub-programs within the overarching conceptual framework; • offer insights into the regional economy, its dependencies, trade-offs between different nature-based industries and natural resource uses, externalities and economic incentive instruments; • explore social response to change in both quantitative and qualitative terms. All models will be built into a ‘toolkit’ to facilitate their delivery and application [24]. 5.2. A preliminary structure for the toolkit The toolkit proposed here comprises four main components: a user interface, a model management system, a multi-objective modelling engine, and an expert system for interpretation of modelling outputs (Fig. 3). The model management system contains biophysical and socio-economic models developed in the OBP projects. It also provides functions for selecting, invoking, running, and integrating models. The multi-objective modelling engine is used to explore and evaluate different regional resource use scenarios through multi-objective optimisation or multi-criteria analysis techniques. After a scenario is defined, this modelling engine develops a mathematical programming model using the biophysical and socio-economic indicators generated from the relevant models in the toolkit to determine optimal resource allocation within biophysical constraints and socio-cultural preferences. It can also be used to develop a multi-criteria decision model that identifies trade-offs among regional resource use options based on stakeholder value judgements. This approach asserts that the integration of multiple scales, issues and objectives can be achieved through the integration of different modelling approaches into higher-level simulation models [25].
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
49
Fig. 3. Preliminary toolkit structure.
The expert system component provides explanation and interpretation of the models and model results. It shows assumptions, explains the modelling process, provides necessary guidelines for use of models, and interprets and translates information derived by the models to a certain format, which is desired by a certain group of stakeholders. The user interface provides users with controls for the implementation of the system. This may include managing scenario investigation and supporting selection of models or tools that seem most useful for an intended application. The toolkit evaluates the effects of the various scenarios on multiple indicators and the trade-offs among them according to multiple criteria and multiple objectives. This toolkit will be linked to the resource information system (Fig. 2), which does not only manage information and data required for modelling, but also provides tools for presenting modelling results in particular forms such as maps. 6. Conclusions This paper highlights the key role that integrated modelling and assessment plays in a new R&D program designed to support regional-scale NRM. Modelling provides the integrating framework across research in
50
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
various disciplines and across NRM issues related to various natural resources and associated industries. It also provides the link from research to real-life application of data and understanding by natural resource managers and planners. A toolkit approach is proposed to support this task and accommodate the variety of models and tools to be built during the lifetime of the program. The road to integration and integrated assessment is paved with many conceptual, methodological, technological and process challenges [26,27]. Thorough organisational and theoretical preparation and research implementation is required to make this a process that truly supports ecologically sustainable development. Acknowledgements The author is indebted to a large number of researchers and stakeholders who have, in the course of numerous meetings and conversations, contributed to the content of the OBP and therefore ultimately to this paper. References [1] J. Rotmans, M. Van Asselt, Integrated assessment: growing child on its way to maturity. An editorial essay, Climate Change 34 (1996) 327–336. [2] J.L.A. Geurts, C. Joldersman, Methodology for participatory policy analysis, Eur. J. Operational Res. 128 (2001) 300–310. [3] M.S. Krol, A. Jaeger, A. Bronstert, J. Krywkow, The semi-arid integrated model (SIM), a regional integrated model assessing water availability, vulnerability of ecosystems and society in NE-Brazil, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (7–8) (2001) 529–533. [4] H. Staudenrausch, W.A. Flügel, Development of an integrated water resources management system in southern African catchments, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (78) (2001) 561–564. [5] C. Pahl-Wostl, D. Schlumpf, M. Büssenschütt, A. Schönborn, J. Burse, Models at the interface between science and society: impacts and options, Integr. Assess. 1 (2000) 267–280. [6] J. Alcamo (Ed.), IMAGE 2.0: Integrated Modeling of Global Climate Change, Kluwer Academic Publishers, Dodrecht, 1994. [7] H.J. Miser, E.S. Quade (Eds.), Handbook of Systems Analysis—Overview of Use, Procedure, Applications, and Practices, Wiley, Chichester, 1985. [8] H.J. Miser, E.W. Quade (Eds.), Handbook of Systems Analysis—Craft Issues and Procedural Choices, Wiley, Chichester, 1988. [9] A. Bronstert, River basin research and management: integrated modelling and investigation of land-use impacts on the hydrological cycle, Phys. Chem. Earth B: Hydrol. Oceans Atmos. 26 (7–8) (2001) 485. [10] M. Hisschemoeller, R.S.J. Tol, P. Vellinga, The relevance of participatory approaches in integrated environmental assessment, Integr. Assess. 2 (2001) 57–72. [11] R. In’t Veld, Afterword, Integr. Assess. 2 (2001) 93–94. [12] H. Dowlatabadi, J. Rotmans, P. Martens, Integr. Assess. 1 (2000). [13] A.K.L. Johnson, S. Cowell, N. Loneragan, G. Dews, I. Poiner, Sustainable development of tropical Australia: R&D for management of land, water and marine resources, Occasional Paper 05/99, LWRRDC, Canberra, 1999. [14] CSIRO, Ord–Bonaparte Program: R&D Plan 2000–2005, CSIRO Tropical Agriculture, Brisbane, 1999. [15] S.A. Mullner, W.A. Hubert, T.A. Wesche, Evolving paradigms of landscape-scale renewable resource management in the United States, Environ. Sci. Policy 4 (2001) 39–49. [16] W.R. Ashby, An Introduction to Cybernetics, Wiley, New York, 1956. [17] P.B. Checkland, Systems Thinking, Systems Practice, Wiley, Chichester, 1981. [18] W.E. Grant, Ecology and natural resource management: reflections from a systems perspective, Ecol. Modell. 108 (1998) 67–76. [19] C.S. Holling, Adaptive Environmental Assessment and Management, Wiley, New York, 1978.
R. Greiner / Mathematics and Computers in Simulation 64 (2004) 41–51
51
[20] C.W. Walters, Adaptive Management of Renewable Resources, Macmillan, New York, 1986. [21] M.C. Jackson, Critical systems thinking and practice, Eur. J. Operational Res. 128 (2001) 233–244. [22] A. Martin, J. Sherington, Participatory research methods—implementation, effectiveness and institutional context, Agric. Syst. 55 (2) (1997) 195–216. [23] P. Senge, The Fifth Discipline, Doubleday, London, 1990. [24] R. Grenier, X. Zhu, Systems Framework for Intergrated Regional-Scale R & D Projects. Proceedings MODSIM (2001) 1201–1206. [25] F. Villa, R. Costanza, Design of multi-paradigm integrating modelling tools for ecological research, Environ. Modell. Software 15 (2000) 169–177. [26] R.M. Argent, R.B. Grayson, S.A. Ewing, Integrated models for environmental management: issues of process and design, Environ. Int. 25 (6–7) (1999) 693–699. [27] U. Dahinden, C. Querol, J. Jäger, M. Nilsson, Exploring the use of computer models in participatory integrated assessment— experiences and recommendations for further steps, Integr. Assess. 1 (2000) 253–266.