RESEARCH AND APPLICATION  Integrating Catchment Ecosystems and Community Health: The Value of Participatory Action Research Margot Parkes* and Ruth Panelli‡ *Ecology and Health Research Centre, Department of Public Health, Wellington School of Medicine, University of Otago, New Zealand; ‡Department of Geography, University of Otago, PO Box 56, Dunedin, New Zealand ABSTRACT Understanding links between catchment management and community health demands consideration of complex bio-physical, socio-economic, and public health relationships. These relationships cut across a spectrum of health, environment and development considerations and highlight the need for appropriate and integrative modes of inquiry and decision making. What can Participatory Action Research (PAR) contribute towards achieving an integrated approach to catchment management and community health issues? In addition to a methodological overview of Participatory Action Research, this paper reviews other participatory, community, action, and ecosystems-based methods. Commonalities in principles and methods are highlighted across a number of fields of research and practice including rural and com- munity development, public health and health promotion, natural resource management, environmental health, and integrated ecosystem-based approaches. Lessons learned from application of Participatory Action Research are described in relation to a catchment and community health project, based in the Taieri River catchment, New Zealand. The case study emphasizes the importance of both horizontal and vertical connections between diverse coalitions of catchment stakeholders and the contribution of PAR cycles of inquiry, reflection, and action, toward this type of integration. Both generic and location-specific examples highlight the value of participatory methods that respond to the challenge of how to integrate the complex social and bio-physical processes that characterize human and ecosystem health. INTRODUCTION tem change and their impacts on humans (Ewert & Kessler 1996; Vitousek et al. 1997; McMichael et al. 1999). Trends toward degradation of our life-supporting ecosystems highlight the challenges facing both modes of inquiry and decisionmaking in order to address these cross-cutting issues in an integrated way. Some of these challenges are addressed in our consideration of freshwater ecosystems and community health. The problems associated with degradation of freshwater ecosystems at global, regional, and local scales indicate the fundamental nature of the relationship between human health and the sustainability of natural ecosystems (Postel 1997; Gleick 2000). Freshwater resource management— especially at the scale of river catchment management—also highlights the challenge of adopting approaches to research and management with the The vision of communities living in safe, sustainable, productive environments is an intuitively appealing social goal. Fulfilling this vision however is requiring increased attention to the links between social and ecological systems (Berkes & Folke 1998) and socio-economic and environmental determinants of health (Baum 1998). While our understanding of the connections between healthy communities and healthy (natural) ecosystems has increased, so too have growing insights into the rate and scale of global, regional, and local ecosysAddress correspondence to: Dr. Margot Parkes, Ecology and Health Research Centre, Department of Public Health, Wellington School of Medicine, University of Otago, PO Box 7343, Wellington South, New Zealand; E-mail mparkes@wnmeds. ac.nz,Fax: 64-4-3895319 ©2001 Blackwell Science, Inc. 86 capacity to integrate complex bio-physical, socioeconomic, and public health issues (Ewing et al. 2000). This type of integration demands approaches to problem identification, inquiry, and problem solving that can cross traditional knowledge boundaries, such as those between the bio-physical and social sciences (Rapport et al. 1998; Bradshaw & Bekoff 2000; Ludwig 2001). We engage with this disciplinary “boundary crossing” in the current paper as we reflect on appropriate methodologies for approaching catchment ecosystems and community health. The management of freshwater resources within catchments (or watersheds) requires consideration of land and water issues at the scale of entire river basins (Hooper 1997; McGinnis et al. 1999; Hildén 2000). At this scale, water resource management and public health are linked by development, environment, and health issues that cut across traditional sectoral, disciplinary, and community boundaries. The HEAD (health, environment, and development) triangle illustrated in Figure 1 is adapted from Witten et al. (2000) and is presented here as a heuristic tool for recognizing academic and scientific boundaries and territories, including their tendency to emphasize different social and bio-physical systems. While inquiries that aim to integrate health, environment, and development considerations are located within the center of the triangle, the surrounding sections of the HEAD triangle serve as a reminder of the obstacles to achieving an integrated approach to complex issues1. Ad hoc approaches to linking across sectors and disciplines are a limited first step to overcoming traditional barriers. An additional challenge is to prioritize systematic and deliberate approaches to integration (World Health Organization 1997; Neller 2000), including collaborative, inclusive alternatives to research and decision making that explicitly acknowledge the complexity and uncertainty of socio-ecological systems (Funtowicz & Ravetz 1994; Kay et al. 1999; Ewing et al. 2000). Heron (1996, p. 9) argues that beyond the task of “democratizing research content”—across different types of knowledge, inquiry, and decision making, is the further challenge of “democratizing research method”—whereby participants are actively involved in determining methods used. In response to these challenges, our paper moves beyond the question of why it is important to integrate across traditional boundaries and between different stakeholders to focus on how this integration may be achieved, based on the specific context of river catchment management. The paper describes the value of Participatory Action Research as a methodology with principles that overlap and complement many other styles of participatory and ecosystem-based research approaches that have been applied in the context of complex human-ecosystem relationships. Overall, the paper seeks to address the question: “What can Participatory Action Research contribute toward achieving an integrated approach to catchment management and community health issues?” Commonalities are identified between Participatory Action Research (discussed in the second section) and other participatory, community-action, and ecosystem approaches as employed in health, environment, and rural development arenas (reviewed in the third section). Disciplinary fields referred to include natural resource management, public health, rural studies, community development, organizational research, and environmental health2. Following this review, a New Zealand case study is provided in the fourth section to illustrate how such an approach may be implemented. PARTICIPATORY ACTION RESEARCH: SOME OBSERVATIONS In this section we provide a brief methodological overview of Participatory Action Research (PAR) as a methodology that is particularly appropriate for addressing complex human and ecosystem relationships. As will be outlined in later sections, a range of non-PAR approaches have utilized inclusive, cross-disciplinary methods in contexts such as agriculture, natural resource management, water resource management, and environmental health (Chambers 1994b; Waltner-Toews & Wall 1997; McGinnis et al. 1999; Yassi et al. 1999). 2 In crossing this spectrum of disciplinary territories, our paper seeks to highlight the variety of precedents that can encourage “understanding of what works and what doesn’t work, and why in various community settings” (Witten et al. 2000, p. 381), and to acknowledge the wealth of experience available that can foster willingness “not only to learn from our own mistakes and successes, but also to avoid the mistakes and to emulate the successes of others.” (McTaggart 1991, p. 185) 1 The considerable academic, political, and personal challenges of integrating across traditional knowledge boundaries are recognized as important themes by Reason (1991), Moles (1995), Heron (1996), and Booth (2000) and warrant explicit, ongoing attention in order to move beyond the potential rhetoric of “integration”. Ecosystem Health Vol. 7 No. 2 June 2001 87 FIGURE 1. The “HEAD” (Health, Environment, And Development) Triangle: Links Between Different Disciplinary Territories. While incorporating many of these characteristics, participatory action methodologies are also notable for their capacity to acknowledge the multiple contexts, stakeholders, and processes involved in both the human and environmental systems. They can also provide relevant research-community relations when innovation and change are potential objectives in such studies as well as potentially addressing the separation of research from policy. Participatory Action Research refers to a spectrum of approaches that share interests in combining research and action through a series of processes by which issues are defined, addressed, and reconsidered. These approaches are widely employed in fields as diverse as community development, education geography, health, sociology, gender, and rural studies (Whyte 1991; Garvin 1995; Bowes 1996; Hiebert & Swan 1999; Motteux et al. 1999). Extensively developed in education studies3, PAR involves forms of inquiry where researchers and the researched population form collaborative relations in order to identify and address mutually conceived issues or problems through cycles of action and research. However, the degrees to which they establish participation processes and develop ongoing “action” components will vary greatly. The two core components of PAR–action and participation–characterize both the goals and strategies associated with this type of research. First, PAR involves an interest in action—in doing, or achieving, some wider social or community goal through the process of research. In this form, action is reflective and focused. It involves both ex3 The educational goals of learning, critical reflection, and generation of new knowledge are highlighted by Kemmis & McTaggart’s description of action research as “the way groups of people can organize the conditions under which they can learn from their own experiences and make the experience accessible to others” (Kemmis & McTaggart 1988, p. 5). plicit purpose and reflection (McTaggart 1991; Goff 2000). This form of research may actively get involved with the problems, risks, or politics of the research context. See for instance: (Garvin 1995; Martin 1996; Karabanow 1999). Second, PAR is implemented through goals of “participation.” Participatory processes enable the researched population to be involved (to varying degrees) in the research program. This focus on participation differentiates PAR approaches as often more inclusive (even sometimes emancipatory) when compared to other conventional methodologies (Bowes 1996; Panelli, 2001), however, it can also set up clearly differentiated positions and power relations between researchers (as experts) and the research population (as those to be “helped”)4. The types of relationships that develop between researchers and researched in PAR are varied. Table 1 illustrates one way of differentiating these relations according to the degree of partnership and participation that is afforded the researched population (i.e., ranging from co-option through to co-learning)5. 4 For instance, Wadsworth’s (Wadsworth 1998) definition of PAR as “active co-research, by and for those to be helped . . . [where] those to be helped, determine the purposes and outcomes of their own inquiry” illustrates the emancipatory role of PAR, enabling previously unheard voices or issues to be valued, recognized, or gain profile (for example, through time, resources, or funding opportunities). However, the language of “those to be helped” also runs the risk that researchers and research subjects can be construed in potentially superior-inferior ways. 5 Participatory relationships can also be differentiated according to the emphasis on “reflection” in both research and action - an important determinant of whether outcomes will be based on thoughtful action, as compared to action per se. (see Heron 1996, p. 7-9 and Goff 2000). Highlighting both personal and collective aspects of reflective participation, McTaggart notes that “the collective reflects on observations made about action taken so far and uses this reflective activity to inform decisions about future action steps of the group and of individuals. In this way the public and the personal spheres of thought and action are kept in step.” (McTaggart 1991, p. 175) Parkes and Panelli: Participatory Action Research in a River Catchment 88 TABLE 1 Types of Participatory Research Mode of Participation CO-OPTION COMPLIANCE CONSULTATION COOPERATION CO-LEARNING COLLECTIVE ACTION Involvement of Local/Researched People Token representatives are chosen but there is no real input or power sharing Tasks are assigned with incentives but outsiders decide the agenda and direct the actions Local opinions are sought but outsiders analyze and decide on the best course of action Local people work together with outsiders to determine priorities but responsibility remains with outsiders for directing the process Local people and outsiders share their knowledge to create new understandings and they work together to form action plans with outside facilitation Local people set their own agenda and mobilize to carry it out in the absence of outside initiators and with or without outside facilitators Relationship of Research to People ON FOR FOR/WITH WITH WITH/BY BY (After Cornwall 1995; Pretty et al. 1995; Martin 1996) The central column of Table 1 differentiates according to involvement of local/research people and draws attention to differences in identifying and conceptualizing the problem or issue to be addressed. Problem formulation and recognition varies from being imposed by external sources (in modes of participation at the top of the table), to local people setting their own agenda in terms of problems, issues, and processes of inquiry and action (in modes at the bottom of Table 1). Distinguishing between these different types of participation poses an interesting challenge for catchment studies (Mullen & Allison 1999; Hildén 2000). Choosing an appropriate mode of participation can enable communities, agencies, and researchers to form participatory relationships and work together through shared actions. While the basic rhetoric of participation is now a familiar concept for many involved with both research and policy, PAR explicitly highlights the importance of being aware of the choices that underlie which type of participation is adopted and the different outcomes that may be generated as a result. Researchers need to be aware not only of the types of participation but also the types of parEcosystem Health ticipant involved (Wadsworth 1998; Panelli 2001). This is especially relevant in the cross-cutting catchment context where there is much to be gained from unraveling the rhetoric of “multiple stakeholders.” As indicated by the heading in the central column of Table 1, those participating may include more than local people from within a catchment. PAR invites explicit consideration of the potential to include a spectrum of participants including catchment agencies, industries, community groups, and other researchers. The act of considering researchers and agencies as participants highlights important, often political issues relating to equity and access to information, power and resources that underpin many research processes. This is especially so in relation to the different knowledge generation and construction processes of different types of participant and how these are valued. A critical success factor for mediating between complex bio-physical and social dynamics evident in a river catchment is the exercise of explicit attention to the dynamics of more or less powerful agencies. These dynamics are the interplay between researchers and their information, Vol. 7 No. 2 June 2001 89 community groups and their information, and the interests of political and private sector pressure groups and individuals (Wadsworth 1998; Goff 2000). Wadsworth and Goff’s attention to these power dynamics6 highlights the importance of ongoing reflection and/or evaluation at the commencement of PAR activities, at strategic turning points, and at an initiative’s conclusion. Attention to reflection and evaluation enables consideration of the domains of new knowledge generated by the initiative, the PAR capabilities of participants and the facilitators, and learning methods that link research practice with knowledge generation. The interplay between the various parties becomes evident through this process and can play a significant role in validating claims and informing strategic interventions. One form of PAR that has supported the complexities of the current inquiry into catchment and community health involves Community Oriented Participatory Action Research (COPAR), see (Liepins 2000a; Panelli 2001)7. This approach has involved an adaptation of PAR principles to research that explicitly positions (and problematizes) the “community” as the central social arena through which participatory action research might effectively occur8. Research partnerships are established through the use of Community Reference Groups, which shape their own relationship with researchers while electing how they wish to become involved in community research. The development of community partnerships is often a complex and not always successful procedure (Panelli 2001). Nevertheless, by combining relations with formal representatives (through local government structures) and volunteers from different sectors of the “community,” a broadly constituted reference group can result in rich research 6 In relation to the power dynamics between participants, Goff (2000, p. 71-74) highlights that the funding body has the vital role of counter-balancing the specific research endeavor and its integrity. Local participant interests may in some cases contravene systemic, long-term development and public values of interest to the funding body. However, these interests should be worked with in a PAR process and understood by the funding body as an indication of emerging political realities—part of the larger “social-ecology” of the issues at hand. 7 Ruth Panelli previously published under the name of Ruth Liepins. 8 The “community” is acknowledged as both a socio-cultural construct and a rhetorical notion of great power in lay, academic, and policy discourses (Liepins 2000b). While the notion is criticized for the exclusionary and homogenizing tendencies that often result from its use, continued use of the term, at least within quotation marks, is advocated since “community” action and research retains the possibility of recognizing and supporting social diversity (and even disagreement) (see Liepins 2000b; Liepins 2000a; Panelli 2001). relationships. These arrangements can also highlight the collective issues and the disagreements and axes of tension inherent in any given case or “community” (Liepins 2000a). The dynamics between multiply positioned (academic, community) researchers will vary in different forms of COPAR. The processes of participation and action are ongoing endeavors in COPAR and respect for each community’s contexts and decisions means that each case will result in different actions and results. However, some standardization of COPAR design is helpful. The basic phases and processes involved in this type of work are summarized in Table 2 and reflect the key features of PAR inquiry, namely the emphasis on establishing appropriate relations and the (potentially endless) cycle of research, reflection, action. One result will be variation in the degree to which academics and/or community members respectively identify and engage with research problems and the chosen actions that might address these problems9. Finally, COPAR processes also illustrate the need to consider participatory relationships with those beyond the immediate community or catchment. For example, distinctions can be made between co-researchers, reference groups, and researcher/facilitator. Goff (2000) builds on the work of Wadsworth (1998) and Heron (1996) to describe: “co-researchers”—those with publicly recognized responsibility to act on the issue; “critical reference groups” (community reference group in the case of COPAR)—those who often have little power to act but bear the consequences of the actions of those with this power; and “facilitators”— those whose focus of inquiry is developing a research method that is true to the participants’ context10. This categorization is especially helpful 9A recent example illustrates this variety even within a single research program. Liepins (2000a) and Panelli (2001) report on COPAR activities that were originally based on an academic’s generic interest in “communities managing change”. However, the COPAR activities that developed in individual case study communities illustrate the diversity of community decisions and actions that can be encouraged as different local groups choose to focus on contrasting issues. In one case, a problem-solving culture developed within the research/reference group and the community addressed intra-community communication problems and conducted a small education needs assessment exercise. In a different case, a reflective and celebratory culture developed where the research/reference group elected to collate a community profile and hold a weekend event to “Celebrate Community”. 10 Goff details the “rights and responsibilities” of the “facilitator” (p. 66-70) and the participants (p. 75-80) (including “critical reference groups” and “co-researchers”) in PAR practice. Attention is given not only to the responsibilities that guide PAR practice and the rights afforded by each role, but also actions that are not appropriate within the terms the participatory model described. Parkes and Panelli: Participatory Action Research in a River Catchment 90 TABLE 2 Community-Oriented Participatory Action Research (COPAR) Phases and Processes Initiations • meeting a community • recognizing a problem or facing a crisis • seeking understanding through partnership Developing a partnership • recognizing the “players” • establishing forms of representation • forming a core research/reference group Reflection • identification of the community problems and/or issues Research Design • explicitly deliberating over and designing a process of inquiry Conducting Investigations • seeking answers or completing fieldwork/research Reflection and Further Planning • analyzing data and reflecting on findings with a view to: deciding upon and implementing new actions: . . . review, reflection . . . re-participating, re-planning . . . researching, and still more action (After Panelli 2001) RURAL AND COMMUNITY DEVELOPMENT in explicitly acknowledging the power relations that are intrinsically involved in a PAR process involving multiple different types of participants, including community groups, and therefore needing multiple perspectives. We reflect on these acknowledgements shortly, when reviewing the application of PAR to the case of the Taieri catchment. A rich research heritage of participatory approaches has developed in the fields of rural and community development, with a focus on community networking and technology transfer. Many of these approaches owe their origins to the various modes of research and practice used in community-based development projects, initiated by international agencies and non-government organizations in developing countries (World Bank 1996; Botes & van Rensburg 2000). Related methods are currently being applied in rural and urban contexts throughout the world, in order to meet objectives relating to both community and environmental sustainability (Pretty 1998; Wates 2000). Application of participatory methods to agricultural and rural contexts in both development and extension initiatives initially focused on technology transfer from experts to communities. Rapid Rural Appraisal (RRA) developed in the 1970s with an emphasis on communities giving information to experts in order that expert knowledge could be applied. Participatory Rural Appraisal (PRA), with its origins from activist participatory research, agroecosystem analysis, applied anthropology, field research METHODOLOGICAL OVERLAPS While Participatory Action Research provides a comprehensive and flexible approach to collaborative and iterative reflection and action, a number of other complementary approaches have developed in diverse fields pertinent to catchment and community health. As shown in Figure 1, these different disciplines and emphases include rural and community development, public and community health, natural resource management, environmental health, and integrated ecosystems-based approaches. This section reviews the relevant qualities and processes of inquiries that parallel or complement PAR, illustrating that a range of values and strategies are common across many of the fields. Ecosystem Health Vol. 7 No. 2 June 2001 91 on farming systems, as well as RRA, came to the fore in the 1980s with an increased attention on shared learning between locals and outside experts (Chambers 1994a). Participatory approaches in the development context use a number of research strategies that complement PAR. Participatory Rural Appraisal explicitly values the analytical abilities of local people and emphasizes the importance of processes used to achieve relaxed rapport between local people and researchers (Chambers 1994b). This attention to types of participation (Table 1), and the emphasis on process as well as product/outcomes are common to many participatory methods used in the development context (McAllister 1999) as well as being core themes in Participatory Action Research. The role of evaluation in order to link participatory projects with their larger temporal or institutional context is another common theme (Estrella et al. 2000). PAR cycles of research, action, and reflection are obviously relevant to meeting the challenge faced by both researchers and agencies (including funders, donors, and institutions) to factor in evaluation throughout the entire project design cycle including pre-project (planning), inproject (implementation), and post-project (followup) phases (McAllister 1999). As well as attention to process evaluation, the growing recognition and valuing of the variety of outcomes that may result from community-driven participatory approaches are also complementary to PAR. Methods for evaluating both community and ecosystem sustainability outcomes, including development of community-based indicators (Flora et al. 2000), are easily incorporated into a PAR approach. PUBLIC HEALTH AND COMMUNITY HEALTH PROMOTION Public health is an area that has traditionally crossed many disciplines and interfaced between multiple community, research, agency, and sectoral stakeholders. In the field of health promotion and in research and practice relating to the socio-economic determinants of health, there has been considerable development of participatory, community-based, and action-oriented initiatives focused on health promotion and harm reduction (Hart & Bond 1995; Boutilier et al. 1997; Isreal et al. 1998). Inclusive and community-based approaches to public health research and practice have particularly been utilized to respond to the challenges of “capacity-building” for public health action as a means to invest in human capital (investment in knowledge and skills), social capital (efforts to maintain mutual trust obligations), and financial capital (adequate investment of financial resources) (Bush & Mutch 1999). Reflection on several decades of public health research and practice has highlighted action research priorities that are applicable to a variety of social and health issues 11. Approaches such as Community-Based (Israel et al. 1998) and Community Reflective Action Research (Boutilier et al. 1997) reflect PAR principles both in their description of the priorities for achieving community action and in their methods for achieving this. In particular, the needs for reflection, analysis, evaluation, and documentation of processes used are identified as priorities for improving both collaborative and intersectoral activity, particularly in terms of the dynamics between the different stakeholders (Sindall 1997; Casswell 1999). The prevention of alcohol and substance abuse is an area of public health research where research strategies particularly reflect and complement many PAR principles. Projects focused on reducing alcohol and related harm through community action research point out the potential to improve partnerships—in particular the relationship between research and public health practice —as well as to improve health and well-being through working with communities to increase their capacity (Conway et al. 2000). Casswell highlights the generic theme of utilizing existing human and information resources when she points out that in ways there is “nothing new about community action, if it is defined as the attempt to coordinate existing resources within the community to prevent (alcohol related) harm” (Casswell 1999, p. 31) [emphasis added]. Reflecting similar community action used at the district level to reduce alcohol and drug-related harm, Bush and Mutch (1999) 11 Hart and Bond (1995) point out that important characteristics of Action Research in the health and social care context are that it: 1. educates; 2. deals with individuals as members of social groups; 3. is problem focused, context specific, future oriented; 4. involves a change intervention; 5. aims at improvement and involvement; 6. is based on cyclic processes in which research, action, and evaluation are interlinked; 7. is founded on a research partnership in which those involved are participants in the change process. In addition, these authors describe multiple potential outcomes of action research in the public health context acknowledging the complementary value of “conceptual”, “structural”, “practical”, and “emancipatory” outcomes. Parkes and Panelli: Participatory Action Research in a River Catchment 92 agement (Ewing et al. 1997; Hooper 1997)13. Principles complementary to COPAR have been applied when using participatory, community-based approaches to catchment management that encourage the integration of perspectives from multiple stakeholders and emphasize capacity building and mobilization of social capital14 as critical to the success of natural resource management initiatives (McGinnis et al. 1999; Mullen & Allison 1999; Pretty & Frank 2000). Integrated Systems for Knowledge Management (ISKM) also focus on involving a range of stakeholders as a means to share existing knowledge, as well as to improve the efficacy and long-term success of Decision-Support Systems (DSS) for specific natural resource management tasks (Bosch et al. 1996; Allen et al. 2001). ISKM particularly aligns with PAR in its attention to the entry and contracting, collaborative planning, information capture, and dissemination and implementation of systems for ongoing monitoring and feedback as part of an iterative process (see Allen et al. 2001, Figure 1). Recent emphasis has been given to the relationships between groups as important for their ability to exchange and benefit from existing community and science resources in natural resource management (Ewing et al. 2000; Pretty & Ward 2001). Pretty and Ward (p. 219) describe stages of stakeholder group relationships as “reactive-dependence,” “realization-independence,” and “awareness-interdependence.” This description of relationships complements PAR methodology not only by highlighting types of participation (Table 1), but also by focusing on the way partnerships can evolve over time based on criteria such as the adoption of reflection and feedback as part of group processes. Discussions of the importance of process and product evaluation as part of integrated and adaptive resource management initiatives are also closely aligned with PAR emphasis on cycles of reflection and action (Allen 1997; Bellamy et al. 1999; McAllister 1999). Generic phases common to many of these ap- outline several valuable insights regarding “capacity building” that draw on similar experiences in community development, organizational development and other public health fields12. It is notable that these proposals could equally be applied to prevent harm in both social or bio-physical systems, by using a PAR approach that pays explicit attention to phases of developing partnerships and reflecting on negotiated methods for integrating both community and science knowledge (Table 2). NATURAL RESOURCE MANAGEMENT There is increasing recognition that “interdisciplinary and participatory research approaches are essential to address complex natural resource management issues, to involve local communities in these processes and to promote sustainable and equitable natural resource management systems” (McAllister 1999, p. 7). A body of participatory research and practice with principles similar to PAR has emerged from the broad fields of adaptive natural resource management, environmental management, sustainable agriculture, and community-based management, with particular focus on partnership building, co-ordination of existing resources, knowledge exchange, and capacity building (including investment in social capital) (Pretty et al. 1995; Imperial 1999; Evans & Birchenough 2001; Allen & Kilvington, 2001). Equity of, access to, and exchange of existing catchment information are core challenges of Integrated Catchment Management (ICM) that prioritize both community and science knowledge in order to improve understanding and implementation of best practices for land and water man- 12 In outlining priorities for capacity building Bush and Mutch (1999) emphasize that: A. local organizations and groups, as the fabric of community life, are untapped resources for harm reduction; B. building capacity for harm reduction appears to depend upon progressing partnerships in which there are reciprocal transfers of knowledge and a growing ability to solve problems; C. adequate measures of capacity are required, because outcome measurements limited to a change in health status [functional or product outcomes] undercut the value of community capacity [process] outcomes; D. assuming that civil society possesses an inexhaustible supply of social capital is a major threat to achieving harm reduction—social capital should therefore be considered a finite resource requiring investment. (A-D are adapted from Bush & Mutch 1999, p. 77-78 “What can we learn about district level capacity building from this experience?”). Ecosystem Health 13 Precedents for collaborative community, web- and/or GISbased catchment information exchange systems are increasing. Australian examples include: http://www.catchment.com; http://hric.tag.csiro.au/information/publications/collgis.html. 14 Pretty and Frank (2000, p. 211) describe social capital as a term that captures the idea that social bonds and social norms are an important part of the basis for sustainable livelihoods, and refer to the contributing roles of “relations of trust”, “reciprocity and exchanges”; “common rules, norms and sanctions”; and “connectedness, network and groups”. Vol. 7 No. 2 June 2001 93 proaches and with considerable overlap with PAR are described in relation to co-management of natural resources (Borrini-Feyerabend et al. 2000). Details of Borrini-Feyerabend’s three phases of I: Preparing for partnership, II: Negotiating plans and agreements, and III: Learning-by-doing, with a reflective feedback loop between Phases I and II, are closely aligned with the phases of COPAR outlined in Table 2. ENVIRONMENTAL HEALTH Addressing the many spatial and temporal scales at which human health is affected by the earth’s ecosystems presents important and complex challenges to the fields of Environmental Health and Environmental Epidemiology. Evidence of methodological and conceptual development is indicated by ecosystem approaches to framing health promotion strategies (Cole et al. 1999b), the emergence of eco-epidemiology (Hales et al. 1997), an emphasis on ecosystem-level analysis (McMichael 1999; Pekkanen & Pearce 2001), and consideration of vulnerability to environmental change (Woodward et al.1998). The complex social and institutional implications of an ecosystem approach to environmental health have highlighted the importance of cross-disciplinary and inter-sectoral approaches, as well as the need to break down barriers at the interface between research and policy (Briggs et al. 1996; World Health Organization 1997; Neller 2000). Implications, in terms of how these interfaces may be crossed have received less attention. Explicit attention to participatory, inclusive methods to address these challenges in environmental health is slowly emerging (Witten et al. 2000), but is limited by the traditional divisions between health protection and heath promotion and related barriers between quantitative and qualitative research. Origins of participatory and community-based environmental health efforts have links with health promotion initiatives such as Healthy Cities and more recently Local Agenda 21, which incorporate concepts of safe, healthy, and sustainable ecosystems (Von Schirnding 1997; Baum 1998; Dooris 1999). Cole et al.’s (1999a) description of community involvement in the development of environmental health indicators exemplifies participatory research based on consultation and touches on some important issues arising regarding whether the indicators are being developed with or for the participants (see Table 1). Yassi et al. (1999) provide a detailed account of a multi-stakeholder “eco- systems approach” to determinants of health, including the development of community-based indicators. Participants in the research could conceivably be categorized as “co-researchers” (municipal agencies and fellow researchers) and “critical reference groups” (community-based focus groups). The several phases of research and action with these participants also indicate the applicability of PAR as a potential framework for this kind of environmental health research, where attention to the processes involved with negotiating partnerships, reflection, and further planning are important matters deserving explicit attention. It is obvious that there are a number of precedents of participatory and community-oriented research applied in the environmental health field. However, in addition to these, and as a contribution to the need for development of principles and best practice for participatory approaches to environmental health research (Witten et al. 2000), PAR is notable as a systemic methodology with explicit attention to reflection in research, action, and evaluation. INTEGRATED ECOSYSTEMSBASED APPROACHES The emerging “ecosystems approach” has emphasized the need for modes of inquiry that can respond to complex human-ecosystem relationships. In addition to institutional and social considerations15, ecosystem-based management must consider the complex interactions between multiple species, unpredictable non-linear changes over time, and the multiple spatial and temporal scales of nested hierarchies16 that are characteristic 15 Imperial’s description of core principles in ecosystem management and collaborative decision making include: improving integration of government policies; enhancing the coordination of various governmental and non-governmental organizations; broad public participation; and the involvement of key stakeholders in government decision making (Imperial 1999, p. 449). 16 VanLeeuwen et al. (1999, p. 204) refer to an example of an ecological hierarchy as : field, farm, land use district, watershed, ecological region, ecological zone, and biosphere. A nested hierarchy can also be described in terms of social units: individuals, families, neighborhood, villages, and communities within larger socio-ecological structures. In relation to socioecological systems, Waltner-Toews describes nested hierarchies where each unit is itself a whole thing, while containing other wholes and being part of something larger (Waltner-Toews 1996, p. 686). Descriptions of nested hierarchies relate closely to Checkland and Scholes’ work describing “holons” (whole units) within “holarchies” (Checkland & Scholes 1990). Parkes and Panelli: Participatory Action Research in a River Catchment 94 of ecosystems (Imperial 1999; VanLeeuwen et al. 1999; Gilbertson 2000). In response to this spectrum of considerations, PAR provides the kind of methodological framework that can move beyond academic and institutional comfort zones to incorporate perspectives of both social and bio-physical sciences and incorporate multi-stakeholder participation and collaborative decision making. The integrated, ecosystems-based approaches referred to in this section include those with explicit attention to the full health, environment, and development spectrum, and with overt methodological parallels with PAR. This represents a selection of literature that has focused on methods to conceptualize, analyze, and address previously described relationships between human and ecosystem health or social and ecological systems (Ewert & Kessler 1996; Karr 1997; Berkes & Folke 1998). The analysis and assessment of agroecosystem health17 are focused not only on developing indicators of agroecosystem health across the biophysical and human domains, but also providing an integrated, systemic framework within which such indicators can be analyzed (Gallopin 1995; WaltnerToews 1996). The new types of questions about health18 associated with this agroecosystem health approach have lead to development of an adaptive Methodology for Ecosystem Sustainability and Health (aMESH) that explicitly draws on theories of PAR and complex adaptive systems (Murray et al. 1999). The aMESH research process describes a number of iterative cycles of research action in keeping with its guiding principles of multiple stakeholder perspectives, methodological pluralism, and hierarchical cross-scale interactions. The SOHO method (Self Organizing Open System Approach to Ecosystem Sustainability and Health) (Kay et al. 1999, Figure 2) also explicitly addresses complex adaptive systems involved with health, environment, and development issues and outlines methodological proposals with themes familiar to the COPAR phases of Table 2. The SOHO approach includes a “problem definition” phase in terms of a socio-ecological system description, and requires attention to development of partnerships in order to achieve a combined “systems” and “collaborative” approach to promote the desired vision. These establishing phases are followed by the design and planning of an adaptive, iterative management program. OVERLAPS–A SUMMARY This third section of the paper has been far ranging in its identification of relevant literatures but space has not allowed for a detailed discussion of each field of work. Rather, in keeping with the methodological aim of this paper, we can conclude that all five fields of endeavor are generating strategies and methods that are pertinent to our interest in links between catchment and community health. The core contexts, techniques, principles, and processes of these fields are summarized in Table 3. As shown in Table 3, these different fields of research have all engaged with various elements of the PAR methodology introduced earlier in this paper. First, in some fields, great emphasis is placed on appropriate relationships of participation, partnership or collaboration (e.g., rural development and public health work). Second, in several cases attention is devoted to different forms of reflection or identification of current states or future goals (e.g., resource management projects). Third, in most cases, some kind of action strategy is mobilized. Finally, several fields emphasize further reflective techniques in the form of substantive or procedural evaluation; thus providing valuable feedback for continued cycles of research. Each of the disciplinary or multi-disciplinary fields represented in Table 3 provides academic support for studies of catchment and community health. However, we have concluded that in a practical sense most fields are focusing either on why over-arching principles or concepts are important (such as “capacity building” or the enhancement of “social capital” or the need for “ecosystems-based” approaches), or on specific sets of research techniques (such as ‘appraisal’ analysis in rural development studies, or development of Decision Sup- 17 Smit et al. (1998, p. 4) utilize the following definition to conceptualize, assess, and interpret health of agroecosystems: “Agroecosystems are managed primarily for the purposes of producing food and fiber and other agricultural products; they comprise domesticated plants and animals, biotic and abiotic elements of the underlying soils, drainage networks, and adjacent areas that support natural vegetation and wildlife. Agroecosystems explicitly include people, who comprise the most influential mammalian community within them among their essential elements. Hence they have socioeconomic and public health, as well as environmental dimensions. Agroecosystems defined in this way fall into a nested holarchy extending from small gardens to farms to large regional zones”. 18 Assessment of complex links between human and (agro)ecosystem health are highlighted by the question: “Are the quality and quantity of internal and external resources sufficient, and is their organization appropriate for the system to meet its goals?” (Waltner-Toews & Wall 1997, p. 1743). Ecosystem Health Vol. 7 No. 2 June 2001 95 TABLE 3 Participatory Approaches to Address Human-Ecosystem links Commonalities between research areas Core contexts Rural Community Development RRA, PRA, PAR, COPAR Public Health Natural Resource Management Environmental Health Participatory methods and techniques Healthy Cities, AR, PAR, Co-management, Ecosystem Community -based ICM, ISKM Approaches Integrated, systems-based approaches Ecosystem Management, aMESH, SOHO Processes and principles • Complex partnerships and networks: multiple stakeholders, disciplines, sectors perceptions, knowledge types. Are relationships dependent, independent, interdependent? • Integration of resources and knowledge exchange: includes co-ordination of existing resources (to prevent harm), and issues regarding equity and access to information: • Cycles of research and action: includes initiation and preparation, developing a partnership, designing and implementing, with reflection and feedback between each phase. • Iterative problem solving and planning: each iteration may include (re)consideration of visions, issues, objectives, strategies, tasks. • Evaluation: includes both process inquiries (procedural, during cycles, formative) and outcome-focused evaluation (retrospective audit--reviews, summative). • Capacity building for sustainability of social and ecological systems: what emphasis on investment into social, human, financial capital? • Multiple potential outcomes: conceptual, structural, practical, and emancipatory outcomes may contribute to meeting overlapping public health, environmental, social, and community objectives. Acronyms: AR- Action Research, aMESH - adaptive Methodology for Ecological Sustainability and Health, COPAR - Community Oriented Participatory Action Research, ICM – Integrated catchment management, ISKM – Integrated Systems for Knowledge Management, PRA - Participatory Rural Appraisal, PAR- Participatory Action Research, RRA - Rapid Rural Appraisal. (After Hart & Bond 1995; Israel et al. 1998; Bush and Mutch 1999; Casswell 1999; McAllister 1999; Borrini-Feyerabend et al. 2000; Pretty & Ward 2001; Allen et al, 2001). port Systems in resource management works). In contrast, Participatory Action Research provides a comprehensive methodology that focuses on how the complexity of research topics, such as catchment and community health, can be addressed in an integrated way. In PAR, we find that the elements of research relationships, reflection, action, and evaluation, are all equally recognized in a comprehensive approach that encourages researchers (irrespective of their social or bio-physical topics) to recognize the contexts of their work and engage in a relational and adaptive iteration of inquiry. In closing this review, we recognize that such an optimistic attitude to PAR is in great measure a reflection of our attraction to the theoretical possibilities of a form of research that can be integrating, dynamic, and reflexive. The prac- tice of such research provides us with the opportunity to review this position, thus, the following section outlines the structure and methodological lessons gained from a specific catchment study. PARTICIPATORY ACTION RESEARCH IN THE TAIERI RIVER CATCHMENT BACKGROUND: A CATCHMENT AND COMMUNITY HEALTH CASE-STUDY Concerns regarding the safety, sustainability, and management of New Zealand’s freshwater ecosystems are receiving increased attention (Ministry Parkes and Panelli: Participatory Action Research in a River Catchment 96 for the Environment 1997; Ministry for the Environment 1999; Tong & Cox 2000). So too is the need for greater integration of research, policy, and relevant local action to address water resource issues and related environment and health considerations (Public Health Commission 1995; Parliamentary Commissioner for the Environment 2000). One area of explicit inquiry involves the study of ecosystem processes underlying New Zealand’s high rates of potentially water-borne disease (Russell et al. 1998; Weinstein et al. 1998; Duncanson et al. 2000). The current investigation complements this type of work by considering the social and physical dimensions of a particular catchment in an integrated way that respects the complex and dynamic nature of the catchment. Positioned within the nexus of health, environment, and development fields, this project is a multi-disciplinary study of the Taieri River catchment, situated in the southeast of Otago in the South Island of New Zealand. The predominantly rural Taieri catchment covers an area of 5650 km2 and is host to a population of approximately 18000 people. Farming, including dairy, cattle, sheep, deer farming, cropping, market gardening, and agri-forestry, dominates land use; and other forestry, residential, wetland, and conservation19 land uses occur in smaller areas of the catchment (Otago Regional Council 1999). The Taieri catchment can be considered in two parts. The sparsely populated upper catchment draining from the headwaters is comprised of native tussock grassland and exotic pasture and characterized by two upland floodplains, rural depopulation, and combined challenges of drought and water-quality issues arising from the recent intensification of agricultural activities. The more densely populated lower catchment also faces rapid intensification of land use, as well as historical sedimentation from the upper catchment, ongoing channelization and floodplain drainage, and more recently, concerns regarding interactions between surface-water, groundwater, and estuarine coastal lakes (Otago Regional Council 1999; Townsend & Riley 1999). The Taieri catchment was chosen as the study area for investigat- ing links between freshwater ecosystems and human health in order to build upon the extensive history of research in the Taieri River and catchment20. The catchment demonstrates the rapid environmental and social change characteristic of many rural New Zealand river catchments (Ministry for the Environment 2000; Tong & Cox 2000) and exemplifies the challenges and opportunities faced when seeking to undertake Integrated Catchment Management of New Zealand river catchments (Bowden, 1999). The Taieri Catchment and Community Health Project (TCCHP) is a participatory, communitybased and collaborative project that was established in 1999 as a means to study the complex links between community health, freshwater ecosystems, and land and water resource management in the Taieri River Catchment. Informed by the literature outlined in the previous section, and in particular the relevance of catchment-scale initiatives to address the complex, multi-stakeholder processes of freshwater management (Ewing et al. 2000; Bowden 1999; McGinnis et al. 1999; Mullen & Allison 1999; Clean Water Action Plan 2000), the researchers were especially interested in the challenge of equitable “co-ordination of existing resources within the catchment community to prevent (catchment-related) harm” (Table 3; Casswell 1999, p. 31). “Harm” in this sense refers to both direct effects on community health from water-related environmental hazards and indirect effects from ecosystem change affecting rural development and socio-economic determinants of health (Figure 1; Witten et al. 2000, p. 374). The decision to use Participatory Action Research as the overarching methodology for the project was a choice made by the authors21 in order to enable both a collaborative and community-based research focus 20 by the University of Otago, Dunedin (Taieri and Southern Rivers Programme 1994-1999; Ecology Research Group 1997; Townsend & Riley 1999). 21 In complement to considerations of Table 1, Heron (1996, Chapter 4) pays particular attention to the “initiation” and “initiator” of a collaborative and co-operative inquiry process. Heron notes that “Some initiating researchers may be academic staff, some may be postgraduate research students, some may work in other professional settings, and others may be part of a special interest group in society that has no institutional form. For those who work in academia there are particular challenges” (1996, p. 72). In the case of the PAR component of the TCCHP, the research was initiated across two academic Departments (Public Health and Geography). The research has contributed toward a doctoral Ph.D. thesis in Public Health (Parkes, Ph.D. Thesis In Progress), and was funded by a three year Training Fellowship in Environmental Health research from the Health Research Council of New Zealand. 19 The catchment’s high ecological and conservation values relate to native tussock grassland habitats, extensive upland scroll-plains, coastal wetlands, and rare native fish and insect species. The catchment also provides freshwater ecosystem services for irrigation, municipal water supplies, hydro-electricity, mining, and many recreational uses. (Otago Regional Council 1999; Townsend & Riley 1999). Ecosystem Health Vol. 7 No. 2 June 2001 97 on learning and action. Collaboration with other researchers optimizes links and integration between a number of bio-physical research initiatives22. The community-based component of the TCCHP involves a variety of collaborative efforts, including communities, researchers, and agencies. The TCCHP thus involves a spectrum of stakeholders in a process of inquiry aimed at:   identifying and mobilizing existing community and science knowledge within the catchment; strengthening capacity for integrated, catchment-based approaches to water resource management and related public health issues23. The conduct of the community-based component of the project is addressed in the following section. APPLYING PARTICIPATORY ACTION RESEARCH TO INTEGRATE CATCHMENT AND COMMUNITY HEALTH ISSUES Participatory Action Research was chosen by the authors as an appropriately comprehensive methodology to engage with the complex bio-physical and socio-economic issues of the Taieri Catchment. Rather than selectively choosing some of the strategies of the previously reviewed literatures, a PAR approach has been adopted as a coherent and sufficiently flexible methodology that can support a cyclic process of research, action, and reflection/evaluation. This process provides the flexibility required to adapt to both context and research findings as they change throughout the period of study. As discussed in the first section, PAR activities also pay explicit attention to the power relations inherent in catchment initiatives with multiple stakeholders and perspectives. 22 These included new and existing research between the departments of Public Health, Zoology, Geography, Information Sciences and Microbiology. In particular, a new multi-disciplinary study investigating the bio-physical relationship between land-use, climate, and the human pathogen Campylobacter was initiated. At over 200 cases per 100 000, New Zealand’s rates of campylobacteriosis are some of the highest in the western world (Withington 1997). 23 Public health issues of primary concern included ecological determinants of waterborne disease, and the links between rural development, ecological sustainability, and community well-being. In practical terms, the basic research process conducted in the TCCHP is summarized in Table 4. Here the standard PAR elements of initiating an inquiry, relationship building, reflection and information exchange, action, and further reflection/evaluation were proposed for the Taieri catchment context. As well as reflecting the initial planning of TCCHP, the structure and stages of Table 4 indicate how the basic framework and principles of the COPAR methodology (Table 2) have been complemented by contributions from other participatory and action research methods. The participatory methods referred to at the base of the table are especially relevant and provide a working example of the benefits achieved when researchers are open to the wealth of methodological experience and precedent available from the fields indicated in Figure 1. Table 4 extends on themes of COPAR described in Table 2 and also explicitly distinguishes between different types of participants in the process. Based on the partnerships developed in Phase II, it is possible to distinguish the ways that different types of participation in the TCCHP have featured at different times in the project. The project initially focused on participation and involvement of “co-researchers”—those with the recognized responsibility to act on the issue. In the Taieri context these “co-researchers” include agency and research stakeholders with a formalized active role and/or responsibility in researching or decision making in the catchment regarding water resources and related environmental health issues. The involvement of these co-researchers as integral participants in the research process from its early stages has proven very valuable in ensuring the success of ongoing cycles of research and action in the Taieri Catchment. Whilst maintaining communication with co-researchers, the community-based component of the project was initially based around fieldwork involving a series of meetings between four different “community reference groups” (CRGs) based in different locational communities (two upper and two lower catchment) around the catchment, and the “researcher/facilitator.” Whilst useful for differentiating between types of participant, the groupings of “co-researcher,” “community reference group” and “facilitator” are not mutually exclusive. Exploring these distinctions provides a useful heuristic exercise highlighting exceptions and overlaps due to particular individual or group characteristics. An important case in point is the extensive and varied contribution of indigenous Mäori knowledge and experience of Parkes and Panelli: Participatory Action Research in a River Catchment 98 Ecosystem Health Vol. 7 No. 2 June 2001 99 catchment and community health issues where contributions would often be as both co-researcher and community member24. Attention to different types of participant in the TCCHP highlights the differences in the associated modes of inquiry (Phase III) and types of action (Phase IV). Reflection in both research and action is required to acknowledge the power relations and knowledge construction processes at work. In the case of the TCCHP, attention was paid to the important processes of dialogue, documentation, and feedback of the three rounds of Community reference group meetings around the catchment. Appropriate use and emphasis on qualitative research methods can ensure that knowledge and data gained about catchment and community health issues from the CRG participatory processes are valued and available for incorporation into the larger PAR process25. Having established initial participation, planning and action with co-researchers and CRGs, the TCCHP progressed through a number of stages beyond the initial framework of Table 4. Future iterations of the project have been based on connections between stakeholder groups, such as those between co-researchers and CRGs. The different types of stakeholder relationships are described below in relation to horizontal and vertical connections. Figures 2 and 3 provide examples of how the framework of PAR can be applied to contribute toward integrated, collaborative, and comprehensive approaches to complex human–ecosystem issues in order to foster the safe and healthy environments. Describing horizontal and vertical connections highlights the temporal and social dynamics of long-term, multi-stakeholder processes that may go beyond the basic cyclic framework of PAR. This differentiation also underscores the flexi24 Contribution by Mäori to the TCCHP occurs across community, researcher, and agency levels, reflecting extensive iwi (tribal) input into natural resource management, development, and public health research and practice within the region. As the local iwi, Ngai Tahu’s contribution was mostly at the co-researcher level through links with the Executive Komiti of Te Rünanga Otakou (sub-tribe executive committee) and Kai Tahu Ki Otago (the organization responsible for local iwi input into resource management in Otago). In addition, participation included fellow researchers actively involved in related research in the Taieri river catchment as well as local whanau (family) living the area. 25 The Community Reference Group meetings generated a variety of qualitative data to support the research supposition that catchment scale ecosystem change has important implications for community health. This relationship was viewed both in terms of illness and disease (direct effects) but more importantly through perceived sense of well-being (including indirect socio-economic and aesthetic effects) (Parkes PhD Thesis In Progress). bility of PAR methodology to respond to new concepts and incorporate to multiple axes of activity. HORIZONTAL INTEGRATION Horizontal integration refers to connections and capacity for links across similar stakeholder groups. The notion of horizontal integration resonates with notions of social cohesion, connectedness, and social capital that have been applied in the analysis of a variety of natural resource management initiatives, including other catchment contexts (McGinnis et al. 1999; Mullen & Allison 1999; Pretty & Ward 2001). Pretty and Ward (2001) use horizontal connections to describe local-local connections and external connections26. The different layers of Figure 2 indicate horizontal connections enabled by application of PAR in the TCCHP, specifically:    links across community reference groups (CRGs) from different catchment locations; links across different disciplinary researchers forming new research collaborations27; links across different health, environment, and development agencies. VERTICAL INTEGRATION The concept of vertical integration between groups can be used to refer to connections between local groups and external agencies or organizations (Pretty & Ward 2001) especially where they convey enhanced partnerships, exchanges of knowledge, or multi-stakeholder actions. Vertical integration 26 Pretty and Ward describe five elements of connectedness: “Local-Local connections: horizontal connection between groups within communities, or between communities which sometimes become platforms and new higher-level institutional structures. External-External connections: horizontal connections between external agencies leading to integrated approaches for collaborative partnerships” are two types of horizontal connections (Pretty and Ward 2001, p. 212). 27 Work in progress from these research projects is indicated by recent conference presentations: (i) Eyles, R., Parkes, M., Macgillivray, K., Benwell, G. Environmental Health Research at the Catchment Scale: Issues, Problems and Challenges in Data Integration and Analysis. The 12th Annual Colloquium of the Spatial Information Research Centre. 10-13 December 2000. Dunedin, New Zealand. (ii) Parkes, M., Stewart, B., Weinstein, P. Links between Marine Ecosystem Health and Human Health: A Comparative Analysis of Ecosystem Health in Pristine and Disrupted Coastal Marine Ecosystems, and Assessment of Public Health Implications. International Symposium on Ecosystem Health. 12-14 July, 2000. Brisbane, Australia. Parkes and Panelli: Participatory Action Research in a River Catchment 100 FIGURE 2. Integration Fostered by PAR in the Taieri River Catchment: Horizontal Connections Across Stakeholder Groups. implies connections to other groups, and thus can also be used to describe initiatives that cross the interface between researchers and policy agencies, or that create new partnerships between researchers and communities. Consideration should be given to whether these links are undertaken as onesided or, preferably, as two-way initiatives. Traditionally, one-sided initiatives have tended to be ‘top-down’ and imposed. In contrast, “bottom-up” (sometimes referred to as “grassroots”) projects have been described as a critical factor in establishing successful, ongoing community catchment initiatives (see Mullen & Allison, 1999) and are in keeping with the emancipatory role of PAR described in previous sections. In the context of the TCCHP vertical integration is used to describe the links that have been made—or in some cases re-established—between different types of stakeholder groups and participants. These vertical connections may result in specific collaborative research initiatives between different stakeholder groups (Phase II, Table 4) or may manifest through a new cycle of meetings, dialogue, and development of relationships between new stakeholder groupings (Phase II, Table 4). Key vertical connections in the TCCHP are depicted in Figure 3 and include: Ecosystem Health  The Taieri Catchment and Community Health Survey (TCCHS) The TCCHS was one of the first specific initiatives resulting from the dialogue with Community Reference Groups (CRGs, see Phases III and IV; Table 4). The decision to undertake the survey was made after the second of a series of four CRG meetings, and was planned as a new phase of collaborative research and action designed and undertaken by the facilitator with ongoing input, dialogue, feedback, and exchange with the CRGs. The survey, which was primarily developed as a vehicle to mobilize community knowledge regarding the experiences (and concerns) people have about land, water, and community health issues in the Taieri Catchment area, also responded to CRG interest in views and opinions outside of the specific reference groups.28 Wide dissemination of the find- 28 The Survey drew on a much larger sample than the qualitative data collected through the COPAR fieldwork with CRGs. Approximately 500 people throughout the catchment area responded to the survey representing about 40% of those asked to participate. Vol. 7 No. 2 June 2001 101   ings from analysis of the qualitative and quantitative data will provide an important vertical flow of information between community participants, other researchers, and relevant agencies. The findings will also act as a base for future dialogue about these issues between stakeholder groups, including researchers, agencies, and research funders. Establishing an ongoing Community-University partnership: the TAIERI forum. Building from horizontal links between co-researchers at the University of Otago and between Community Reference Groups (including the cross-catchment community meeting, Figure 2), both community and researcher participants expressed interest to the facilitator in the opportunity to meet together. An unprecedented community-university meeting in a location chosen by the Community Reference Groups subsequently highlighted the shared interests in the future safety and sustainability of the Taieri River catchment, as well as the potential for an ongoing catchment forum to address related issues in the long term. These interests were formalized in a Community-University Partnership: The TAIERI29 forum (Figure 3). Consolidating links between the TAIERI forum and external agencies/funders. The choice by representatives of the four Community Reference Groups and coresearchers from the University of Otago to form an ongoing partnership in the catchment created a new dynamic between the PAR participants in TCCHP. With the survey undertaken and a new partnership underway, the type of action of particular interest to the TAIERI forum relates to the establishment of ongoing communitybased catchment initiatives in the Taieri River catchment. This new cycle of research and action necessitates a transition (or feedback-loop/iteration) into a new phase of 'partnership building’(Phase II, Table 4) including consideration of types of participation and type of partici- 29 TAIERI  “Taieri Alliance for Information Exchange and River improvement” (see Figure 3). The TAIERI catchment forum includes involvement from four University of Otago Research themes: Ecology, Conservation and Biodiversity Research Group, New Zealand and the Pacific in Global Context, Public Health, and Spatial Information Processing. pant, negotiating roles and representation, accessing resources, and engaging in dialogue regarding shared plans. The TAIERI forum must also negotiate new and existing relationships with funding bodies, facilitators, and co-researchers as the three-year funding and time-frame of the initial PAR component of the TCCHP comes to a close. Of particular interest to the TAIERI forum is the challenge of finding ongoing resources to invest in the human, social, and financial capital that will be crucial to ongoing catchmentrelated participatory action research activities. Nonetheless, the time and effort spent on building existing horizontal and vertical connections between participants in the TCCHP, mean that a number of avenues are available to consolidate the commitments expressed by regional agencies to support the TAIERI forum30. CONCLUSIONS We commenced this paper noting that complex inquiries aimed at research and action for healthy communities and healthy ecosystems require approaches that can integrate contrasting biophysical and social knowledges, processes, and concerns. Integrating knowledge of freshwater ecosystems and community health was introduced as one example of such endeavors. We noted relevant and supporting perspectives could be provided from a large range of literature. This included work in rural development, public health, resource management, environmental health, and integrated ecosystems-based approaches. However, 30 In May 2001, the TAIERI forum received preliminary confirmation of three years of funding from the New Zealand Ministry for the Environment’s ‘Sustainable Management Fund’. Core funding to employ a catchment coordinator/project manager and undertake key forum tasks is available, and further details of the project funding will be negotiated between the TAIERI forum (community and university members), the Ministry for the Environment, and key regional agencies in the area. These negotiations will pay particular attention to processes that enable community empowerment and foster change in practice toward improved environmental outcomes. Funding negotiations will also focus on in-kind support and relationship-building with existing agencies to ensure that the project will a) complement, enhance, and avoid duplication in relation to existing initiatives in the region and b) maximize potential for continuity and sustainability through regional (as compared to national) funding after the 2001-2004 period. Parkes and Panelli: Participatory Action Research in a River Catchment 102 FIGURE 3. Integration Fostered by PAR in the Taieri River Catchment: Vertical Connections Across Stakeholder Groups. we also recognized that an integrated/transdisciplinary and practical research program would require an appropriate methodology. Participatory Action Research provides one such alternative. This methodology recognizes research as an active and relational process where knowledge can be continually constructed through reflection on current understandings and the analysis of new and specific actions. After outlining the basic relational (participatory) and active (action) components of this approach—and a communityoriented derivative—we noted that many of the relevant literatures currently employ similar or complementary methods. Different disciplinary fields have focused on elements of research relations, reflection, action, and evaluation, though in more specific and less comprehensive ways than we found with PAR. Taking the example of the Taieri Catchment and Community Health Project, we have indicated how PAR may be applied to investigate links between a specific catchment ecosystem and the community health of local populations. The project has highlighted a number of lessons that can inform the ongoing development of useful and releEcosystem Health vant participatory research to address human and ecosystem health issues. First, the TCCHP study shows that relationship-oriented methodologies such as PAR enable social and scientific capacity to be built while mobilizing scientific and community knowledge. Second, the project indicates the benefits that can be achieved when researchers are willing to engage in “boundary-crossing” activities. Indeed, individual researchers, research teams, and funding agencies have a responsibility to cross boundaries between social and bio-physical sciences. The momentum and integration created by these activities adds weight to the argument for projects having transdisciplinary intentions from the outset (Rapport et al. 1998; Bradshaw & Bekoff 2000). Third, attention to how an integrated approach can be achieved also reveals an oftenoverlooked responsibility. As well as advocating for integrated, participatory, and equitable approaches to inquiry and problem solving, researchers and decision-makers also carry a responsibility to recognize, value, and develop the tacit skills required for the practice of boundary crossing–both in themselves and others. Not least are the skills of listening, critical reflection, humor, and humilVol. 7 No. 2 June 2001 103 ity that are often essential to genuinely reflexive, innovative, and evaluative processes. Fourth, the TCCHP illustrates how participatory action research can foster horizontal and vertical capacity building that may lead to a sustainable web of connections–in and beyond the specific catchment. In the case of the TCCHP, horizontal links initiated in the early stages of the PAR process (Figure 2) provided a platform for the innovative opportunities of vertical integration outlined in Figure 3. In turn, it could be argued that the motivation provided by vertical links such as the establishment of innovative partnerships like the TAIERI forum, also provide more weight to the horizontal connections between researchers, communities, and agencies. Finally, and perhaps most significantly, the TCCHP has shown that PAR cycles of research and action are not chronologically linear–one commencing as another finishes. Instead, as depicted in Figure 2 and 3, it is important to note that these links and research activities were complexly iterative. Each process of horizontal integration across groups represented its own axis of PAR cycles, many of which were occurring simultaneously during the second year of the project. Furthermore, by building on initial horizontal connections, future iterations focused on vertical connections, such as the TAEIRI forum, have the potential to continue the PAR activities well beyond the initial TCCHP time frame. The combined momentum from multiple axes of PAR activities has created a more favorable environment for a diverse and sustainable web of vertical and horizontal connections between the multiple participants involved within the TCCHP. The potential for varied and ongoing connections indicates the dynamic characteristics of processes with an integrated approach to the overlapping objectives of safe, healthy, sustainable, resilient ecosystems and communities. ACKNOWLEDGMENTS The authors would like to pay particular thanks to the time, enthusiasm, support, and encouragement of individuals and groups throughout the Taieri River catchment who have been involved in this project, and associated discussions with Philip Weinstein, Colin Townsend, and George Benwell. We also gratefully acknowledge the constructive comments received by the journal reviewers. Peer review by Susan Goff has enabled the authors to benefit from Susan’s considerable experience in PAR and resulted in valuable conceptual improvements to the paper. Jane Gregg also gave helpful feedback on an early draft. Thanks also for conceptual and practical support given by Will Allen and Robbie Razack as well as colleagues and friends at the University of Otago Departments of Public Health and Geography. Margot Parkes acknowledges the support of a Training Fellowship in Environmental Health research granted by the Health Research Council of New Zealand. REFERENCES Allen, W.J. (1997) Towards improving the role of evaluation within natural resource management R&D programmes: The case for learning by doing. Canadian Journal of Development Studies: XVIII, Special Issue on Results-Based Evaluation, 629–643. Allen, W.J., Bosch, O., Kilvington, M., Oliver, J. (2001) Benefits of collaborative learning for environmental management: Applying the Integrated Systems for Knowledge Management approach to support animal pest control. Environmental Management 27, 215–223. Baum, F. (1998) Healthy societies and environments. In: Baum, F. (ed) The New Public Health: An Australian Perspective pp. 387–498. Oxford University Press, Melbourne, Australia. Bellamy, J., McDonald, G., Syme, G., Butterworth, J. (1999) Evaluating integrated resource management. Society and Natural Resources 12, 337–353. Berkes, F. & Folke, C. (1998) Linking Social and Ecological Systems: Management Practices and Social Mechanisms for Building Resilience. Cambridge University Press, Cambridge, U.K. Booth, M. (2000) Health and wholeness from topology to laughter: Notes toward a theory of connectedness. Ecosystem Health 6, 92–98. Borrini-Feyerabend, G., Farvar, M.T., Nguinguiri, J.C., Ndangang, V. (2000) Co-management of Natural Resources: Organising, Negotiating and Learning by Doing. GTZ and IUCN, Kasparek Verlag, Heidelberg, Germany. Available at: http://www.massey.ac.nz/ nrm/changelinks/cmnr.html. Bosch, O., Allen, W.J., Williams, J., Ensor, A. (1996) An integrated system for maximising community knowledge: Integrating community-based monitoring into the adaptive management process in the New Zealand high country. The Rangeland Journal 18, 23–32. Botes, L. & van Rensburg, D. (2000) Community participation in development: Nine plagues and twelve commandments. Community Development Journal 35, 41–58. Boutilier, M., Mason, R., Rootman, I. (1997) Community action and reflective practice in health promotion research. Health Promotion International 12, 69–78. Parkes and Panelli: Participatory Action Research in a River Catchment 104 Bowden, B. (1999) Integrated catchment management rediscovered: An essential tool for a new millennium. In: Proceedings of Manaaki Whenua Conference: A Three Day Conference on Science for Resource Management. 21-23 April 1999. Manaaki Whenua: Landcare. Te Papa, The Museum of New Zealand, Wellington. Available at: http://www.landcare.cri.nz/ conferences/manaakiwhenua/papers/index.shtml ? bowden. Bowes, A. (1996) Evaluating and empowering research strategy: reflections on action-research with South Asian women. Sociological Research Online 1. Available at: http://www.socresonline.org.uk/1/1/1. html#top. Bradshaw, G. & Bekoff, M. (2000) Integrating humans and nature: Reconciling the boundaries of science and society. Trends in Ecology and Evolution 15, 309–310. Briggs, D., Corvalan, C., Nurminen, M. (1996) Linkage Methods for Environment and Health Analysis. General Guidelines. A report of the Health and Environment Analysis for Decision-making (HEADLAMP) project for UNEP, USEPA and WHO. WHO, Geneva. Bush, R. & Mutch, A. (1999) Capacity building for harm reduction at the district level: Conceptual development and the dimensions of practice. In: Proceedings of Fourth Symposium on Community Action Research and the Prevention of Alcohol and other Drug Problems 8-13 February 1998. Alcohol and Public Health Research Unit: Runanga, Wanaga, Hauora me te Paekaka, Russell, Bay of Islands, New Zealand. Casswell, S. (1999) A decade of community action research. In: Proceedings of Fourth Symposium on Community Action Research and the Prevention of Alcohol and other drug problems. 8-13 February, 1998. Alcohol and Public Health Research Unit: Runanga, Wanaga, Hauora me te Paekaka, Russell, Bay of Islands, New Zealand. Chambers, R. (1994a) The origins and practice of Participatory Rural Appraisal. World Development 22, 953–969. Chambers, R. (1994b) Participatory Rural Appraisal (PRA) - analysis of experience. World Development 22, 1253–1268. Checkland, P. & Scholes, P. (1990) Soft Systems Methodology in Action. John Wiley and Sons, Chichester, U.K. Clean Water Action Plan (2000) Watershed Success Stories. Applying the Principles and Spirit of the Clean Water Action Plan. US Departments of Agriculture, Commerce, Defence, Interior, EPA. Available at: http://www.cleanwater.gov/success/. Cole, D., Pengelly, L.D., Eyles, J., Steib, D., Hustler, R. (1999a) Consulting the community for environmental health indicator development. Health Promotion International 14, 145–154. Cole, D.C., Eyles, J., Gibson, B.L., Ross, N. (1999b) Links between humans and ecosystems: the implications of framing for health promotion strategies. Health Promotion International 14, 65–72. Conway, K., Tunks, M., Henwood, W., Casswell, S. Ecosystem Health (2000) Te Whanau Cadillac - A waka for change. Health Education and Behaviour 27, 339–350. Cornwall, A. (1995) What is participatory research? Social Science and Medicine 41, 1667–1676. Dooris, M. (1999) Health Cities and local Agenda 21: the UK experience - challenges for the new millenium. Health Promotion International 14, 365–375. Duncanson, M., Russell, N., Weinstein, P., Baker, M., Skelly, C. (2000) Rates of notifiable cryptosporidiosis and quality of drinking water in Aoteroa New Zealand. Water Research 34, 26–34. Ecology Research Group (1997) Research in the Taieri catchment. In: Proceedings of Research in the Taieri Catchment. A symposium supporting multi-disciplinary environmental research in Otago and celebrating the sesquicentennial of Dunedin, Dunedin, New Zealand. Ecology Research Group, University of Otago, Dunedin, New Zealand. Estrella, M., Blauert, Campilan, D., Gaventa, J., Gonsalves, J., Guijt, I., Johnson, D., Ricafort, R. (eds) (2000) Learning From Change: Issues and Experiences in Participatory Monitoring and Evaluation. ITDG Publications, London. Evans, S. & Birchenough, A. (2001) Community-based management of the environment: Lessons from the past and options for the future. Aquatic Conservation: Marine and Freshwater Ecosystems 11, 137–147. Ewert, A. & Kessler, W. (1996) Human health and ecosystems: Impacts and linkages. Ecosystem Health 2, 271–278. Ewing, S.A., Grayson, R.B., Argent, R.M. (1997) Research Integration in ICM (Integrated Catchment Management): Review and Discussion Document. Centre for Environmental and Applied Hydrology Report. 1 ⁄ 97. University of Melbourne. Melbourne, Australia. Ewing, S.A., Grayson, R.B., Argent, R.M. (2000) Science, citizens and catchments: Decision support for catchment planning in Australia. Society and Natural Resources 13, 443–449. Flora, C., Odell, S., Kinsley, M., Luther, V., Wall, M. (2000) Measuring Community Success and Sustainability: Measuring the local and regional impacts of economic and community development processes used to enhance rural community sustainability. Aspen Institute, Rural Economic Policy Program, Aspen. Available at: http://www.ncrcrd.iastate.edu/Community_ Success/entry.html. Funtowicz, S.O. & Ravetz, R. (1994) Uncertainty, complexity and post-normal science. Environmental Toxicology and Chemistry 13, 1181–1885. Gallopin, G. (1995) The potential of agroecosystem health as a guiding concept for agricultural research. Ecosystem Health 1, 128–140. Garvin, T. (1995) “We’re strong women”: Building a community-university research partnership. Geoforum 26, 273–286. Gilbertson, M. (2000) Living with Great Lakes chemicals: Complementary strategies and cross-paradigm reconciliation. Ecosystem Health 6, 24–38. Vol. 7 No. 2 June 2001 105 Gleick, P.H. (2000) The changing water paradigm. A look at twenty-first century water resources development. Water International 25, 127–138. Goff, S. (2000) Draft Facilitators Participatory Action Research Resource Kit. Queensland Health, Brisbane, Australia. Hales, S., Weinstein, P., Woodward, A. (1997) Public health impacts of global climate change. Reviews on Environmental Health 12, 191–199. Hart, E. & Bond, M. (1995) Action Research for Health and Social Care: A Guide to Practice. Open University Press, Buckingham, U.K. Heron, J. (1996) Co-operative Inquiry: Research into the Human Condition. Sage, London. Hiebert, W. & Swan, D. (1999) Positively fit: a case study in community development and the role of participatory action research. Community Development Journal 34, 356–364. Hildén, M. (2000) The role of integrating concepts in watershed rehabilitation. Ecosystem Health 6, 39–50. Hooper, B. (1997) Improving watershed management using an innovative integrated resources management approach in Australia. Journal of Professional Issues in Engineering education and practice April, 57–61. Imperial, M. (1999) Institutional analysis and ecosystem-based management: The Institutional Analsysis and Development framework. Environmental Management 24, 449–465. Israel, B., Schulz, A., Parker, E., Becker, A. (1998) Review of community-based research: Assessing partnership approaches to improve public health. Annual Review of Public Health 19, 173–202. Karabanow, J. (1999) Creating community: a case study of a Montreal street kid agency. Community Development Journal 34, 318–327. Karr, J.R. (1997) Bridging the gap between human and ecological health. Ecosystem Health 3, 197–199. Kay, J., A Regier, M., Boyle, G., Francis, G. (1999) An ecosystem approach to sustainability: addressing the challenge of complexity. Futures 31, 721–742. Kemmis, S. & McTaggart, R. (1988) The Action Research Planner (3rd edn). Deakin University, Victoria, Australia. Liepins, R. (2000a) Exploring rurality through ‘community’: Discourses, processes and spaces shaping Australian and New Zealand rural communities. Journal of Rural Studies 16, 325–341. Liepins, R. (2000b) New energies for an old idea: Reworking approaches to ‘community’ in contemporary rural studies. Journal of Rural Studies 16, 23–35. Ludwig, D. (2001) Crisis and transformation. Conservation Ecology 5, art 11. Available at: http://www.consecol. org/vol15/iss11/art11. Martin, M. (1996) Issues of power in the participatory research process. In: de Koning, K., Martin, M. (eds) Participatory Research in Health: Issues and Experiences pp. 82–93. Zed Books, London, U.K. McAllister, K. (1999) Understanding Participation : Monitoring and Evaluating Process, Outputs and Outcomes. International Development Research Centre, Ot- tawa, Canada. Available at: http://www.idrc.ca/ reports/index.cfm. McGinnis, V.M., Wooley, J., Gamman, J. (1999) Bioregional conflict resolution: Rebuilding community in watershed planning and organising. Environmental Management 24, 1–12. McMichael, A.J. (1999) From hazard to habitat: Rethinking environmental health. Epidemiology 10, 460–464. McMichael, A.J., Bolin, B., Costanza, R., Daily, G.C., Folke, C., Lindahl-Kiessling, K., Lindgren, E., Niklasson, B. (1999) Globalization and the sustainability of human health. An ecological perspective. Bioscience 49, 205–210. McTaggart, R. (1991) Principles of participatory action research. Adult Education Quarterly 41, 170. Ministry for the Environment (1997) The State of New Zealand’s Environment, 1997. New Zealand Ministry for the Environment, Wellington, New Zealand. Ministry for the Environment (1999) Making Every Drop Count: the National Agenda for Sustainable Water Management. Action plan. Draft for Public Comment. New Zealand Ministry for the Environment, Wellington, New Zealand. Ministry for the Environment (2000) Managing Waterways on Farms: A Guide to Sustainable Water and Riparian Management in Rural New Zealand. New Zealand Ministry for the Environment, Wellington, New Zealand. Moles, A. (1995) Les sciences de l’imprecis. Seuil, Paris, France. Motteux, N., Etienne, N., Rowntree, K., Binns, T. (1999) Exploring community environmental knowledge through participatory methods in the Kat River valley, South Africa. Community Development Journal 34, 227–231. Mullen, M. & Allison, B. (1999) Stakeholder involvement and social capital: keys to watershed management success in Alabama. Journal of the American Water Resources Association 35, 655–662. Murray, T.P., Kay, J.J., Waltner-Toews, D., Raez-Luna, E. (1999) Adaptive Methodology for Ecosystem Sustainability and Health (AMESH): An introduction. In: Proceedings of Conservation Medicine Conference. 29 April-2 May 1999. White Oak Conservation Center, Florida. Neller, A. (2000) Opportunities for bridging the gap between environmental and public health management in Australia. Ecosystem Health 6, 85–91. Otago Regional Council (1999) Taieri River Catchment Monitoring Report. 99 ⁄ 316. Otago Regional Council, Dunedin, New Zealand. Panelli, R. (2001) Community-Oriented Research: Reviewing Ideals and Negotiations with Power Relations. Department of Geography, University of Otago, Dunedin, New Zealand. Parkes, M. (in Progress) Integrating catchment ecosystems and community health: Participatory environmental health research at the catchment scale. Ph.D. Thesis, Department of Public Health and Department of Geography, University of Otago, Dunedin, New Zealand. Parkes and Panelli: Participatory Action Research in a River Catchment 106 Parliamentary Commissioner for the Environment (2000) Aging Pipes and Murky Waters: Urban Water Systems for the 21st Century. Parliamentary Commissioner for the Environment, Wellington, New Zealand. Available at: http://www.pce.govt.nz/ reports.html#Water Supply. Pekkanen, J. & Pearce, N. (2001) Environmental epidemiology: Challenges and opportunities. Environmental Health Perspectives 109, 1–5. Postel, S. (1997) Freshwater ecosystem services. In: Daily, G. (ed) Nature’s Services: Societal Dependance on Natural Ecosystems. pp. 195–214. Island Press, Washington D.C. Pretty, J., Pretty, J.N., Guijt, I., Thompson, J., Scoones, I. (1995) A Trainers’ Guide to Participatory Learning and Action. IIED, London. Pretty, J. (1998) Towards sustainable rural communities, part III. In: Pretty, J. (ed) The Living Land. pp. 195–264. Earthscan Publications Ltd, London. Pretty, J. & Frank, B.R. (2000) Participation and social capital formation in natural resource management: Achievements and lessons. In: Proceedings of International Landcare 2000. 2-5 March. Melbourne, Australia. Pretty, J. & Ward, H. (2001) Social capital and the environment. World Development 29, 209–227. Public Health Commission (1995) The Local Environment: The Public Health Commission’s Advice to the Minister of Health 1994-1995. Public Health Commission, Wellington, New Zealand Rapport, D., Costanza, R., McMichael, A.J. (1998) Assessing ecosystem health: Challenges at the interface of social, natural and health sciences. Trends in Ecosystem Evolution 13, 397–402. Reason, P. (1991) Power and conflict in multi-disciplinary collaboration. Complementary Medical Research 2, 144–150. Russell, N., Weinstein, P., Woodward, A. (1998) Cryptosporidiosis: an emerging microbial threat in the Pacific. Pacific Health Dialog 5, 137–141. Sindall, C. (1997) Intersectoral collaboration: the best of times, the worst of times. Health Promotion International 12, 5–7. Smit, B., Waltner-Toews, D., Rapport, D., Wall, E., Wichert, G., Gwyn, E., Wandel, J. (eds) (1998) Agroecosystem Health: Analysis and Assessment. University of Guelph, Ontario, Canada. Taieri and Southern Rivers Programme (1994-1999) Taieri and Southern Rivers Programme Annual Reports. University of Otago, Dunedin, New Zealand. Tong, R. & Cox, G. (2000) Clean and Green? The New Zealand Environment. David Bateman Publishers, Auckland, New Zealand. Townsend, C. & Riley, R. (1999) Assessment of river health: accounting for perturbation pathways in Ecosystem Health physical and ecological space. Freshwater Biology 41, 393–405. VanLeeuwen, J., Waltner-Toews, D., Abernathy, T., Smit, B. (1999) Evolving models of human health toward an ecosystem context. Ecosystem Health 5, 204–219. Vitousek, P.M., Mooney, H.A., Lubchenco, J., Melillo, J.M. (1997) Human domination of the earth’s ecosystems. Science 277, 494–499. Von Schirnding, Y. (1997) Addressing health and environment concerns in sustainable development with special reference to participatory planning in initiatives such as healthy cities. Ecosystem Health 3, 220–228. Wadsworth, Y. (1998) What is Participatory Action Research? Action Research International Paper 2. Available at: http://www.scu.edu.au/schools/gcm/ ar/ari/p-ywadsworth98.html. Waltner-Toews, D. (1996) Ecosystem health: A framework for implementing sustainability in agriculture. Bioscience 46, 686–689. Waltner-Toews, D. & Wall, E. (1997) Emergent perplexity: In search of post-normal questions for community and agroecosystem health. Social Science and Medicine 45, 1741–1749. Wates, N. (2000) The Community Planning Handbook. How people can shape their cities, towns and villages in any part of the world. Earthscan, London. Weinstein, P., Russell, N., Woodward, A. (1998) Drinking water, ecology and gastroenteritis in New Zealand. In: Proceedings of 2nd International Symposium on Assessing and Managing Health Risk from Drinking Water Contamination. Santiago. Withington, S. & Chambers, S. (1997) The cost of campylobacteriosis in New Zealand in 1995. New Zealand Medical Journal 110, 222–224. Whyte, W. (1991) Participatory Action Research. Sage, London. Witten, K., Parkes, M., Ramasubramanian, L. (2000) Participatory environmental health research in Aotearoa/New Zealand: Constraints and opportunities. Health Education and Behaviour 27, 371–384. Woodward, A., Hales, S., Weinstein, P. (1998) Climate change and human health in the Asia Pacific Region: who will be the most vulnerable? Climate Research 11, 31–38. World Bank (1996) The World Bank Participation Sourcebook. The World Bank, Washington D.C. World Health Organization (1997) Intersectoral Action for Health: Addressing Health and Environment Concerns in Sustainable Development. WHO/PPE/PAC/ 97.1. World Health Organization, Geneva. Yassi, A., Mas, P., Bonet, M., Tate, R.B., Fernandez, N., Spiegel, J., Perez, M. (1999) Applying an ecosystem approach to the determinants of health in Centro Habana. Ecosystem Health 5, 3–19. Vol. 7 No. 2 June 2001