MONITORING IN SUPPORT OF POLICY: AN ADAPTIVE ECOSYSTEM APPROACH

Michelle Boyle

James J. KayEnvironment & Resource Studies

University of WaterlooCanada

Bruce PondScience Development and Transfer Branch

Ontario Ministry of Natural Resources

REFERENCE:Boyle, M., Kay. J., and Pond, B., 2001. Monitoring in Support of Policy: an AdaptiveEcosystem Approach, in Munn, T., (eds), Encylopedia of Global EnvironmentalChange, Volume 4, pp. 116-137, John Wiley and Son.

The development of a monitoring programme is usually regarded assynonymous with generating indicators. This simplification reflects the belief thatthere exists a “correct” set of things to measure. Once these measures arediscovered, there remains only the task of collecting the data and making the resultsavailable. Such a traditional approach permits monitoring programs to be developedthat are disassociated from the context – that is, people and their concerns.Ungrounded initiatives can produce fragmented programmes1 that are not effectiveand do not use resources efficiently. Or, their abstruse utility may cause them to fizzleout before they are even implemented.

These shortcomings occur despite the best intentions and conscientious work.The difficulty lies in the fact that normal science and available problem-solving toolsare insufficient when dealing with complex systems. Issues of sustainability arepersistently untidy and demand an examination of the wider ecological-economicsystem to resolve. Thus, conventional monitoring methods applied to sustainabilitytend to fail. New approaches and tools that incorporate complexity into monitoringprogrammes are required.

A reframing of science, propelled by the need to unravel complexity, is emergingunder the rubric of "Post-normal science" (Kay et al. 1999)(Funtowicz and Ravetz,1993). In this chapter, we present a scheme where the activities of monitoring andassessment are treated as an integral component of an adaptive ecosystemapproach to sustainability and health.

1 For example an inventory of one biosphere reserve we were involved with revealedover 60 separate and uncoordinated monitoring initiatives.

2

© M. Boyle 2001

In this ecosystem approach, governance, management and monitoring, form atriad of activities that are carried out in the context of an issues framework of humanconcerns and an explicit conceptual model of the ecological-economic system (seeFigure 1). Taken together, the issues framework and the conceptual model focus thediscussion of sustainability: the sustained health and integrity of the ecological-economic system. Governance, management and monitoring activities then informdecisions to resolve the necessary tradeoffs and continually chart a course tosustainability.

Traditionally, in most western jurisdictions, the issues framework andconceptual model, which animate an ecosystem approach, take the form of policiesand/or regulations which are enacted as part of a legislative process. In thistraditional mode, monitoring, as described in this chapter, is undertaken in thesupport of policy. However, there are forms of virtual governance emerging whichhave no formal legislative powers and whose governance does not take the form oftraditional policy. For example, United Nations biosphere reserves are often overseenby an informal and ad hoc assembly of concerned individuals and non-governmentorganizations. Even though these individuals may have no legal power, they do act asstewards of the reserves and in effect, manage them. Monitoring, as discussedherein, is meant to be in the service of such initiatives as well as the more traditionalpolicy-based approaches to environmental and sustainability issues.

1. COMPLEX SYSTEMS THINKING, SETTING THE STAGEA central challenge of successfully resolving the issues of sustainability for

ecological-economic systems, is learning to deal with complexity; that is,understanding situations in which dynamics are dominated by self-organizingphenomena (Kay et al. 1999; Kay and Schneider, 1994; Schneider and Kay, 1994b).A theoretical basis for such an understanding is emerging and is referred to ascomplex systems thinking (see Table 1). Its main elements are hierarchy and self-organization theory.

The hierarchical nature of complex systems requires that they be studied fromdifferent types of perspectives and at different scales. There is no one correctperspective. Rather, a diversity of perspectives is required for understanding.

By their nature, such systems are self-organizing. This means that theirdynamics are largely a function of positive and negative feedback loops. Linear,causal mechanical explanations of their dynamics are precluded. In addition,emergence and surprise are normal phenomena in systems dominated by feedbackloops. Inherent uncertainty and limited predictability are inescapable consequencesof these system phenomena.

Complex systems organize about attractors. Even when the environmentalsituation changes, the system's feedback loops tend to maintain its current state.However, when system change does occur, it tends to be very rapid and evencatastrophic. Precisely when the change will occur, and to what state the system willchange, are generally not predictable. In a given situation, there are often severalpossible system states (attractors) that are equivalent. Which state is currentlyoccupied is a function of its history. There is not a "correct" state for the system,although there may be a state which is preferred by humans. This new understanding

3

© M. Boyle 2001

of complexity suggests a different role for science and monitoring in the quest forsustainability. Following is a brief discussion of the key notions of complex systemsthinking with particular emphasis on the ecological side of the ecological-economicequation.

Table 1: Properties of complex systems to bear in mind when thinking aboutecological-economic systems.

•NON-LINEAR: Behave as a whole, a system. Cannot be understood by simplydecomposing into pieces which are added or multiplied together.

•HIERARCHICAL: Are holarchically nested. The system is nested within a system andis made up of systems. Such nestings cannot be understood by focusing on onehierarchical level (holon) alone. Understanding comes from the multipleperspectives of different types and scale.

•SELF-ORGANIZING (INTERNAL CAUSALITY): are Non-Newtonian, not a mechanism,but rather self-organizing. Characterized by: goals, positive and negative feedback,autocatalysis (self-reinforcing processes), emergent properties and surprise.

•WINDOW OF VITALITY: Must have enough complexity but not too much. There is arange within which self-organization can occur. Complex systems strive foroptimum, not minimum or maximum.

•DYNAMICALLY STABLE?: may not have equilibrium points (that is they may not existfor the system).

•MULTIPLE STEADY STATES: Do not necessarily have a unique preferred systemstate in a given situation. Multiple attractors can be possible in a given situationand the current system state may be as much a function of historical accidents asanything else.

•CATASTROPHIC BEHAVIOUR: Experience sudden unpredictable behaviour and statechanges.

Bifurcations: moments of unpredictable behaviour when a system's developmentmay proceed in either of two divergent directions

Flips: sudden discontinuities, rapid changeHolling four-box cycle: The ongoing process of birth-growth-death and renewal,

e.g. The shifting steady state mosaic of a wildfire-driven forestedlandscape. (Holling, 1986)

•CHAOTIC BEHAVIOUR: have limited predictability. Our ability to forecast and predictis always bounded (for example, to between five and ten days for weatherforecasts) regardless of how sophisticated our computers are and how muchinformation we have.

4

© M. Boyle 2001

1.1 HIERARCHYThere is growing comprehension that sustainability issues must always be

examined within their broader context. Every system is a component of anothersystem and is, itself, made up of systems. So, a wetland must be understood in thecontext of the subwatershed it is a part of, and in terms of the processes and specieswhich make it up. The body of thinking which deals with these issues is calledHierarchy Theory. (Allen and Starr, 1982)(Ahl and Allen, 1996) Its central tenent is thatsustainability issues can only be understood in terms of systems embedded insystems which are also embedded in systems or, in the vernacular of hierarchytheorists, as nested holons.

Hierarchy theory requires that a study of complex systems begin by carefulconsideration of the types of perspectives required and the appropriate scales ofinvestigation. For example, the traditional ecological perspective, which dictates anapproach based on studies of population dynamics of individual species, is nothelpful in understanding fire dynamics in ecosystems. Rather, understanding of thefire phenomena comes from studies of communities and landscapes over local andregional scales.

The researcher must take care to identify the human sustainability issues athand and the appropriate perspectives and scales of investigation necessary to dealwith these issues in an ecological-economic context. This identification process canonly occur in the context of human values and requires bringing a diversity of views tobear on the question at hand.

To do otherwise is to court disaster. For example, research on ecosystems andsustainable food production in some areas of the upper Amazon watershed havehistorically centred on cattle production and its effects. A community based,hierarchical system analysis of the situation revealed the main food source to be fishwith cattle playing an incidental role. Thus, community health and developmentinitiatives focusing on beef production were sorely misplaced. Another examplecomes from an early Canadian environmental impact assessment that examined theeffects of development on reproduction of caribou herds. Later, it was noted that themost serious effect of development was on the food source of the caribou, but thiswas not studied as part of the impact assessment. The study focus was on thereproductive habitat of a single species, and ignored the (much more relevant)broader food chain.

The challenge for science is to abandon the normal approach of searching forthe single correct model for dealing with a problem. Instead, the message ofhierarchy theory is that we must develop a manner of investigation that uses adiversity of different perspectives and models, which brings different players to thetable, and which synthesises the different perspectives together into understanding.

1.2 SELF-ORGANIZATION AND ATTRACTORSComplex systems have multiple possible operating states or attractors, and may

shift or diverge suddenly from any one of them (Holling, 1986; Kay, 1991; Kay, 1997;

5

© M. Boyle 2001

Ludwig et al. 1997). For example, a portion of a natural area1 in Southern Canada isa closed soft maple swamp in a wetland community. However, the amount andduration of the flows of water can radically alter this operating state. Drying events,such as an extended drought, could change the operating state to an upland forestcommunity or grassland with associated vegetation structure. If extended periods offlooding do occur, high water levels would shift the operating state to a marshecosystem. The shift happens because red and silver maple are tolerant to floodedconditions within 30% to 40% of the growing season. If flooding events are greaterthan this threshold, the forest trees die and give way to more water tolerantherbaceous marsh vegetation. The feedback mechanism which maintains theswamp state is evapotranspiration (i.e. water pumping) by the trees. Too much wateroverwhelms the pumping capability of the trees and not enough shuts it down. Thepoint of this example is that the current ecosystem state is a function of its physicalenvironment and the accidents of its history. A single dry or wet season can change,for decades, what is on the landscape.

Each of the three ecosystem states in this example is as ecologically healthyand appropriate as the others. There is not a "right" community for this landscape.Each is equally right. Thus, scientists cannot tell decision makers or policy makerswhich of these three states is ecologically better. Scientists can only provideinformation about the different tradeoffs each state represents. A decision regardingwhich of these three states to promote is necessarily a value decision. In fact, sincewater flow is the key influence which determines the organizational state of thisportion of the natural area, decisions about which ecosystem state to promote hingeon what kind of land use and development is allowed on the lands adjacent to thenatural area - clearly a political decision.

A similar example comes from the management of the same natural area (Listerand Kay, 2000). Currently, the aquatic ecosystem consists of a fast-moving, highlyoxygenated cold-water stream in which brook trout thrive. However, another state,which is normally incompatible with the survival of brook trout, is emerging: a slow-moving, low-oxygen, warm-water stream interrupted by ponds and small wetlands. Inthis case, beaver and muskrat are dominant species that shape and maintain thehabitat and its constituent communities. The current ecosystem trajectory is tendingtowards this state due to several factors, including an invading population of beaverand purple loosestrife. The beaver dam the stream, thus providing more habitat forbeaver and the purple loosestrife choke out the other species in the wetland. Bothfactors slow the water flow, increase water temperature and decrease dissolvedoxygen. The emergence of this attractor was a surprise to the local managers of thenatural area. However, managers operating at the much larger scale of the drainagebasin (of which this natural area is part) were not surprised as they have followed thebeaver migration into the area for years.

The group responsible for the natural area would like to maintain both the troutstream and the beaver and their ponds. However, this may not be feasible since the

1 A natural area is a section of countryside which has been set aside to bemaintained in its natural state. An example of a natural area would be a national parkor biosphere reserve.

6

© M. Boyle 2001

two ecosystem states are potentially mutually exclusive. A difficult choice may have tobe made to favour one over the other. The irony is that ecosystem managers mayhave to trap beaver (Canada's national animal) and keep them out of the natural areain order to preserve a "pristine state" which will likely require intensive humanmanagement to maintain.

In making this choice, the values of various stakeholders and communitymembers will play an essential role in the decision making process. The role of thescientists will be to inform decision makers about the ecological options, the tradeoffsand uncertainties involved, and various strategies for influencing what happens onthis landscape. However, scientists cannot inform us about the "correct" way toproceed, nor can they predict with certainty what will happen in the situation. So, therole of science in decision making for sustainability changes from problem solver (inthe sense of providing a solution for the situation) to the role of facilitatingunderstanding about the bio-physical realities of the situation. In this manner, bothexperts and stakeholders contribute to the resolution of the situation.

1.3 NARRATIVESBy their nature, ecological-economic system dynamics cannot be captured in a

single model. Multiple descriptions are required to deal with complexity, and self-organizing phenomena cannot be explained in terms of a linear mechanistic causaldescription. Consider the case of shallow lakes, such as Lake Erie in North America.Two different attractors for shallow lakes have been identified (Carpenter andCottingham, 1997; Kay and Regier, 1999; Regier and Kay, 1996; Scheffer, 1998). Inthe benthic state, the ecosystem is characterized by high water clarity and lake bottomvegetation. As nutrient loading increases the turbidity of the water, the ecosystemcrosses a catastrophe threshold and flips into a hypertrophic, turbid, phytoplanktonecosystem, i.e., a pelagic state. Lakes exist that regularly flip between the twoattractors. At least three quite different descriptions of such a lake will be needed, onefor the pelagic state, one for the benthic, and one for the intermediate stage as thesystem flips between attractors.

The nature of these descriptions is also quite different. The description is interms of the feedback loops that tend to maintain the ecosystem in its current state.Of particular importance are those feedback loops which buffer the system fromchanges in external influences. In the case of Lake Erie, the benthic ecosystem haselaborate feedback schemes, operating at different spatial and temporal scales, tolimit the phosphorous in the water column. The pelagic state has elaborate schemesto accomplish just the opposite. Describing the "flip" from one attractor to the otherinvolves accounting for how environmental influences (acting at different spatial andtemporal scales) disable one feedback system while enabling another. Such adescription takes the form of a multilayer narrative which recounts the ecosystem'soperation from different perspectives and scales (Kay and Regier, 1999; Kay et al. 1999). While individual elements may consist of traditional scientific models anddescriptions, synthesizing these elements together into a narrative transcends themto invoke a more complete picture of the system. Narratives are a cornerstone of thedescription of the dynamics of complex systems. They are also the end product of

7

© M. Boyle 2001

monitoring programmes undertaken as part of an adaptive ecosystem approach tosustainability and health.

1.4 ECOLOGICAL-ECONOMIC SYSTEMS AS SELF-ORGANIZINGHOLARCHIC OPEN (SOHO) SYSTEMS

Our premise is that resolving sustainability issues for ecological-economicsystems entails understanding these systems as “self-organizing holarchic open”(SOHO) systems. Such an understanding comes from thinking through thehierarchical nature of these systems by considering issues of type and scale, and thebounding and nesting of the system. This “hierarchical systems description” of theimportant processes and structures, and their relationships and context, is essential.In addition, the self-organizing behaviours of the system need to be identified,described, and understood in so far as is possible. This involves identifying theattractors accessible to the system, the feedbacks which maintain the system at theattractors, the external influences which define the context for a specific attractor, andconditions under which flips between attractors are likely. The overall understandingof system behaviour that comes from studying aspects of a SOHO system issummarized in the form of a narrative. (A more detailed discussion of these notionscan be found in: (Kay et al. 1999; Kay, 1991; Kay, 1997).)

As mentioned above, a hierarchical system description requires identification ofthe important aspects of the system. What is deemed “important” can only beidentified by examining the context of human values and human sustainability issuesassociated with the given situation. Therefore, in order to undertake a SOHOapproach, the stakeholders and actors involved in the situation must participate. Aswell, their roles, issues, concerns and the aspirations that they bring to the situationmust be understood. This understanding is summarized as an issues framework forthe situation, which includes a vision for the future.

The process of identifying the important ecological and societal aspects of thesystem proceeds on the basis of the human issues framework and vision. Together,the SOHO system description and the issues framework are the starting point for anadaptive ecosystem approach to sustainability and health (Figure 2).

2. MONITORING AS PART OF AN ADAPTIVE ECOSYSTEM APPROACHTO A SUSTAINABLE SOCIETY

The complexity of resolving sustainability issues, necessitates a self-organizingholarchic open (SOHO) system approach. SOHO systems have the ability todecouple their behaviour from environmental change, to generate novel behaviour,and to self-organize in response to change (in a way that we cannot determine).These properties thus limit our capacity to predict how such situations will unfold. Inprinciple, in many situations it will not be possible, or even appropriate, to makeaccurate quantitative models which forecast the future. It may be more useful togenerate qualitative narratives of how the situation might unfold.

The premise of the traditional approach to management (anticipatorymanagement) is that it is possible to predict and anticipate the consequences ofdecisions. Once all the necessary information is gathered to make a scientific

8

© M. Boyle 2001

forecast, the “right” decision can be made. This approach is not valid when dealingwith complex systems, however. Given the limitations imposed by complexity, thefocus of management and decision-making strategies must be on maintaining thecapacity to adapt to changing environmental conditions. Adaptive managementinvolves a very different agenda than anticipatory management (Gunderson et al.,1995; Holling, 1978). In adaptive management, differences between how the futureactually unfolds and how it was envisioned are seen as opportunities for learning.This is in sharp contrast to anticipatory management which sees such deviations as"errors" to be avoided. Much of the agenda of adaptive management is learningthrough experimentation rather than focusing on error avoidance. Monitoring plays acrucial role in adaptive management. It provides feedback necessary for adaptivelearning, information to generate narratives, and advice to determine if alternativecourses of action are required.

At the core of an adaptive ecosystem approach to sustainability and health is thepremise that a sustainable society maintains itself in the context of the largerecological system of which it is part. The formulation of a sustainable society involvesrealizing a vision of how the landscape of human and natural ecosystems should co-evolve as a self-organizing entity. Decision making comes to be understood for what ithas always been, finding our way through partially undiscovered country rather thancharting a scientifically determined course to a known end point. A framework for thisundertaking is presented in Figure 2.

In this post-normal approach to sustainability, science takes on a differentfunction. The scientist's role in decision making shifts from inferring what willhappen, that is, making predictions which are the basis of decisions, to providingdecisions makers and the community with an appreciation, through narrativedescriptions (scenarios), of how the future might unfold. Through these SOHOdescriptions science informs society about known ecological constraints andpossibilities. (See the top left hand box in Figure 2.) People provide an image of howthey would like to see the landscape of human and natural ecosystems co-evolve.(See the top right hand box in Figure 2.) A dialogue must ensue (the diamond box inFigure 2) which explores the desired and the feasible options and reconciles these ina vision of how to proceed. Scientists inform this dialogue by providing futurenarratives that will evolve as scientists participate, as equals, with others in theprocess of resolving the vision. Having resolved a community vision for the future, thenext phase is to design an adaptive programme for the realization of the vision.

This adaptive programme consists of a plan and infrastructure for the triad ofactivities, governance, management and monitoring. Governance refers to thecontinuing process of learning, revisioning, resolving tradeoffs, and planning to adaptto the unfolding situation. The ongoing evolution of governance arrangements andstructures is required to accommodate these activities. All around the Great Lakes,for example, this is happening with the emergence of virtual governance, communitybased initiatives which organize to focus on specific elements of the landscape suchas watersheds or bays. Management is the activity of translating the vision into reality.It involves the development and implementation of strategies to promote ordiscourage specific forms of self-organization (i.e. trajectories to particularundesirable attractors or system states)in the context of the communal vision and

9

© M. Boyle 2001

plan. This means maintaining the context for the self-organizing complex (SOHO)systems, rather than intervening in the system in a mechanical way . Maintaining thecontext involves identification of external contextual changes, flows into and from thesystem, and feedback loops to be encouraged and discouraged. Generally,management concentrates on the relationship between humans and naturalecosystems, and guides the human side of the relationship. Monitoring is the activityof observing the human and natural systems and synthesising the observations into anarrative of how the situation has actually unfolded and how it might unfold in thefuture. This narrative is used as the basis for governance and management; that is,for learning, revisioning, and adapting human activities as the human and naturalecosystems co-evolve as a self-organizing entity.

3. THE ENTERPRISE OF MONITORINGAs stated at the beginning of this essay, the traditional view of monitoring tends

to focus solely on indicators. By implication, monitoring is restricted to collecting dataand reporting. Our position is that it is precisely this narrow interpretation ofmonitoring that is responsible for its impotence in the service of decision making andpolicy. Our thesis is that monitoring, done in isolation from a decision makingprocess and community needs, is an academic exercise whose utility is left tohappenstance. This is why we have gone to some lengths to set out the context formonitoring. The details of the activity of monitoring follow.

The input to a monitoring programme is a statement about: which self-organizing holarchic, open ecological-economic system is the focus of attention; whataspects are of importance; how it is hoped that these will change over time (i.e. anarrative of the future); and, probably, a plan or strategy for making these aspirationsreality. This statement is an end result of the process depicted in Figure 2 before thebox labeled "ongoing adaptive management". Such a statement is referred to in thisdiscussion as a sustainability vision.

Traditionally, in most western jurisdictions, the sustainability visions have takenthe form of policies and/or regulations which are enacted as part of a legislativeprocess. In this traditional mode, monitoring, as described herein, is undertaken inthe support of policy. However, there are forms of virtual governance emerging whichhave no formal legislative powers and whose governance does not take the form oftraditional policy. For example, United Nations biosphere reserves are often overseenby an informal and ad hoc assembly of concerned individuals and non-governmentorganizations. Even though these individuals may have no legal power, they do act asstewards of the reserves and in effect manage them. While they have no instrumentsthat resemble policy in its traditional sense, they most certainly will have a vision anda plan for the biosphere reserve. Similarly, coalitions of concerned citizens canemerge around a sustainability issue (for example, air or water quality issues in aregion) and take it upon themselves to make changes in society to alter the currentsituation. Monitoring, as discussed herein, is meant to be in the service of theseinitiatives as well as the more traditional policy-based approaches to environmentaland sustainability issues.

The approach to monitoring that follows is not appropriate for all situations. Forexample, if one is simply trying to enforce a specific regulation, like water quality

10

© M. Boyle 2001

standards, then the monitoring exercise is simply one of appropriate surveillance totrigger an enforcement process. Similarly if one is trying resolve a normal scientificissue, like the effect of a fertilizer on a crop, a different approach to monitoring isappropriate. However by the same token, approaches to monitoring that areappropriate to these situations are not appropriate for dealing with the broader issuesof sustainability.)

An output of a monitoring programme is a narrative of how the organization of theSOHO system has actually changed and how it might unfold in the future. Thisnarrative is compared to the narrative of the sustainability vision. The comparison isused to reflect on all aspects of the sustainability vision and on the appropriateness ofthe monitoring programme itself. This assessment is the basis for learning on thepart of those undertaking management and governance. In this way the monitoringprogramme is a reflexive activity.

3.1 THE MONITORING PROCESSFigure 3 depicts the different activities involved in the process of monitoring.

(The activity of developing a monitoring programme is discussed below.) It should beexpected that, as part of an adaptive approach, the monitoring programme will need tobe revised from time to time. Data collection and calculation of indicators are theactivities traditionally associated with monitoring and there is an abundant literatureon these activities. Synthesis is about generating the narrative about the current andfuture status of the ecological-economic system and comparing it to the goals andaspirations expressed in the sustainability vision or /policy. The next stage isreporting. This is the communication of the findings of the monitoring andassessment to those involved in management and governance. It is a crucialelement of the interface between the triad of activities that make up an adaptiveapproach to sustainability.

The report provides the opportunity to perform two crucial evaluations that are atthe heart of the learning process which makes this approach adaptive. Oneevaluation involves the monitoring programme itself: Is it meeting the needs of thepeople for whom it is undertaken? The other evaluation is an assessment of whetheror not the situation is unfolding as articulated in the sustainability vision or policy.

Given that ecological-economic systems are complex, it is reasonable to expectthat the situation is not unfolding in accordance with the sustainability vision or policy.If this is indeed the case, then it should trigger an investigation into what thediscrepancies teach us about the behaviour of the ecological-economic system as aself-organizing holarchic open system2. Discrepancies should also trigger areevaluation of the policy. Does our description of the system, and what is deemedimportant, need to be revised? Is the vision still feasible, or is a revisioning exercisein order? Are new management strategies or tactics required? In essence, oneshould go through another iteration of the process outlined in Figure 2. Thisreiteration may very likely result in modifications to the monitoring programme itself.

2 Such discrepancies may herald global environmental change.

11

© M. Boyle 2001

3.2 ATTRIBUTES OF A MONITORING PROGRAMMEBased on a review of literature (Boyle, 1998) on monitoring and indicators, and

from consideration of the principles of complex systems theory and post-normalscience, we have compiled a list of attributes which a comprehensive monitoringprogram should possess. This list is quite separate from characteristics of thesystem of interest or from the indicator selection criteria. Our research revealed thatthere is often confusion between attributes, which apply to a monitoring program or tothe set of indicators as a whole, and criteria that serve to evaluate individual potentialindicators. (The latter are discussed separately under the section “IndicatorDevelopment”.) The following list of attributes is not only helpful to keep in mindduring the design of a monitoring program, but it also becomes an important tool inevaluating whether the program actually provides the required information and meetsthe needs of the users.

A comprehensive monitoring program should:

• Be relevant and decision supportive. The monitoring program must be relevant to both the issues of concern, and to theusers of the information. The information provided to decision makers must beunderstandable, timely and as forward looking (anticipatory) as possible so thatdecisions may be made before the system is adversely affected.

• Take into account considerations of scale and type. The monitoring program should provide information at a range of spatial andtemporal scales (hierarchically nested) that are appropriate to the vision yet alsobe sensitive to changes (temporal, spatial and across groups). The programmust examine different types or perspectives of the system and consider abiotic,biotic and cultural factors. A separate issue is that data must be collected andreported at appropriate, but not necessarily the same, scales.

• Be based on a conceptual systems model that explicitly recognizesrelationships between society and environment. The conceptual model of the SOHO system must be scientifically valid and shouldreflect the integration between society and the environment in several ways. Itmust be nested to deal with considerations of scale and type (as above) and bebased on ecosystem and institutional boundaries. Human systems must beacknowledged as a subset of the ecological systems that support them, andlinkages between the economy and the health of ecological systems must bemade. It is also important to monitor the feedbacks between society and theenvironment.

• Allow for an overall integrated evaluation of the system. In order to evaluate the state of the system, the monitoring program should collecta variety of different types of measures. It is important to monitor the sensitivity tomagnitude, direction and duration of known stresses, but also to monitor for

12

© M. Boyle 2001

ecological integrity (which would reveal unanticipated problems). Continuousmeasurement from stressed to non-stressed conditions tracks the pathways ofchange between states. An evaluation of the system should be able to assess: itsorganization (structure and process); states; the quantity and quality of flows ofenergy and materials; and, whether changes are reversible or controllable.Historical and baseline information can be used to identify natural ranges ofvariability and trends over time. The monitoring program should thus reflect ourknowledge of naturally occurring changes as part of normal system behaviour andresponses. Cumulative effects should be measured and compared withthreshold values for the system. Emergent properties of the system should alsobe detected and evaluated. Finally, all system assessments should rely on bothscientific, objective measures as well as experiential, subjective ones.

• Be adaptive and flexible. The program must be adaptive and flexible enough to: deal with environmental(including catastrophic) changes; incorporate new information, technology andscientific research; and adjust to changes in the political context or societal values.It is useful to have a monitoring scheme applicable to diverse operationalsituations at different scales and for different ecological systems, yet the programmust also be tailored for specific applications.

• Be practical. A program design is useless if it is not practical and cost-effective. Employingexisting expertise, data sets and tools (e.g. GIS and modelling methodologies)can save time and money and incorporate the work of others.

We have so far discussed the role of monitoring in an adaptive ecosystem approachto sustainability and health, the activity of monitoring and the attributes of a monitoringprogramme in support of sustainability vision or policy. Explaining the “elements” of amonitoring programme and their development remains.

4. DESIGN OF THE ELEMENTS OF A MONITORING PROGRAMME Many sources in the literature offer steps or stages to follow in designing a monitoringprogram. In reality, of course, the process is not linear. Information and knowledgebecome available piecemeal and many insights are gained along the way. Steps aredone and redone. We have chosen to separate the design “steps” and the process ofmonitoring, which are two quite different activities. A monitoring program, then, ismade up of “elements” (Table 2) that are collectively used in the enterprise ofmonitoring. Developing the elements of the program should not be confused with thedevelopment of the protocol for the activity of monitoring, that is, the monitoringprocess. The latter is about designing a human activity system as depicted in Figure3, the former is about designing tools for the analysis of ecological-economicsystems.

13

© M. Boyle 2001

Table 2: Elements of a monitoring programme.

• An issues framework. The impetus of a monitoring program is always to assess progresstowards a set of human goals. Hence, a clear articulation of thegoals and users of the information is the foundation of anymonitoring program.

• A conceptual model of the world. The model represents how we look at the world in the context of thegoals. It serves to delineate the system that should be monitored. Itprovides a framework that relates the indicators to each other in thecontext of the overall system being monitored.

• A set of indicators. The indicators characterize the system being examined in ameaningful way for the users.

• A methodology for data collection and storage. Carefully laid out procedures to address the practical and technicalissues involved in data collection must be established to ensureaccuracy, consistency and statistical robustness. Equally important isthe storage of the data, so that it is accessible and usable in thefuture.

• A methodology for calculating indicators. The data collected will have to be manipulated in order to derivevalues for the indicators. Again, a method to do this accurately,consistently and in a statistically appropriate manner is required.

• A process for synthesis. Synthesizing the information that the indicators provide into an overallnarrative of the status of the system is essential to completing thecentral task of the monitoring program, that is, to assess progresstowards the human goals and aspirations which motivated theenterprise.

• A methodology for reporting. The values of the indicators and the results of the systemassessment (the narrative which results from the synthesis) must bereported to the intended audience or users of the information. Amethodology for presenting it in a clear, purposeful, and timelymanner for decision making is crucial to the utility and success of themonitoring program.

14

© M. Boyle 2001

The relationship among these elements is shown in Figure 1. Strictly speakingthe issues framework and conceptual model should be developed as part of anoverall adaptive ecosystem approach to sustainability and health, rather than asspecific elements of a monitoring programme. These elements would be producedas part of the analysis of the situation, as described by the right- and left-hand topboxes of Figure 2. They would provide the context for the monitoring programme.However, it is usually the case that the development of a monitoring programme is notinitiated in the context of a larger approach to sustainability. If this occurs, thedevelopment of a monitoring programme (in particular the development of the first twoand last two elements of the programme in Table 2) can be used as impetus tostimulate the development of the governance and management activities necessaryto complement monitoring.

4.1 ISSUES FRAMEWORK The decisions that guide our choices and actions in relation to the environment areinextricably rooted in human concerns that are based upon our world view andshaped by our culture and values. These considerations determine our generalvision of how we want the world to be. In many cases, where a traditional approach istaken, the perspective brought to bear on a problem remains implicit in the problemsolving procedure. The same can be said of decision making criteria. This fact isoften a source of conflict as sustainability is not primarily a technical, scientific, ormanagerial problem. It is about dealing with people and their diverse cultures,interests, visions, priorities and needs (Cormick et al., 1996). Pursuing sustainabilityentails participation of these myriad stakeholders, each having a legitimateperspective, and arriving at a resolution that all can live with. The purpose of theissues framework is to bring the diverse interests, visions, priorities and needs of thestakeholders into focus in an integrated way.

It is difficult to bring stakeholders with diverse beliefs and opinions together toresolve often highly contested and emotional issues. These problems, thoughsignificant, are not insurmountable. See, for example, Ramírez (Ramírez, 1999) andBurgoyne (Burgoyne, 1994) for methods in stakeholder analysis and collaboration.Related discussions on public participation in environmental planning may be foundin Warriner (Warriner, 1997) and Beckenstein et al. (Beckenstein et al., 1996). Wehave found that a variant of the Soft Systems Methodology (SSM) (Allen et al. 1993;Bunch, 2000; Checkland, 1981; Checkland and Scholes, 1990; Wilson and Morren,1990) is a useful way to proceed to develop an issues framework that recognizes andincorporates the perspectives of all stakeholders.

The issues framework is designed to explicitly: a) identify the actors andstakeholders b) describe and define the situation, c) determine the important issuesand concerns, and d) disclose the values applied in making decisions andjudgments. The issues framework takes the form of a hierarchical mapping ofhuman concerns and preferences onto the SOHO ecosystem description (theconceptual model). Thus, it facilitates comparison of various scenarios that reflect thesocio-economic and ecological possibilities. It defines integrity and sustainability forthe situation in terms of a vision for the future. The issues framework lays thefoundation for developing a clear statement of vision and articulated preferences in

15

© M. Boyle 2001

terms of characteristics or features of the desired ecological-economic system.Therefore, the sustainability vision directs the choice of information which should becollected by the monitoring programme.

4.2 CONCEPTUAL MODEL The conceptual model for a specific situation describes how the importantconsiderations (derived from the issues framework) are related in a dynamic system.Such a description requires complex systems thinking. The practical result is aframework for a narrative of the behaviour of the ecological-economic system ofinterest. A comprehensive conceptual model should naturally lead to indicators byeffectively telling us where to put “meters”.

The explicit development of a conceptual model is crucial for the activities involvedin monitoring, management and governance. Our literature review revealed, however,that (a) many monitoring programmes do not explicitly define the assumed underlyingmodel, and (b) there are many different types of conceptual models in use. (Someexamples are outlined below in the subsection entitled “Traditional conceptualmodels”.) Furthermore, we found that none of the existing conceptual modelsadequately deals with the complexity of ecological-economic systems. Little attentionis paid to the considerations of scale and type, holons, nesting, self-organizingentities, attractors, feedback loops etc. or in short, a systems description to providethe framework for narratives. This means, more often than not, that the monitoringprogramme does not provide sufficient information to generate a system descriptionthat will be able to deal with sustainability issues.

We, therefore, undertook the task of developing a conceptual model thataccommodates the level of complexity involved in monitoring for sustainability. Theresult of our work is the model already introduced in this paper. The theories andinsights behind it are explained in greater depth in the following pages.

4.2.1 Traditional conceptual models

A review of traditional conceptual models used in monitoring programmes can befound in Boyle (Boyle, 1998). Many current monitoring programs are based on astress-response model. When an ecological system is subjected to a disturbancethat it cannot absorb, it will change in some way. The stress-response model(Rapport et al. 1985) attempts to link stresses with responses and therefore provideearly warning or predict how, and when, a system will respond to influences. Thereare many examples of indicator frameworks which utilize this model, for exampleCanada’s State of the Environment Reporting Program (Environment Canada, 1995).An improvement in this monitoring approach is the inclusion of cumulative effects, thatis, compounding effects from more than one source or persistent additions from thesame source. (Cocklin et al. 1992). Monitoring under the stress-response model islimited, however, to addressing questions regarding known cause and effectrelationships in a system. The notions of monitoring for ecosystem health and, morerecently, ecosystem integrity are efforts to monitor and evaluate the status of thesystem as a whole. Ecosystem health monitoring focuses on attributes of the systemthat give an overall indication of any problems in the system, analogous to taking aperson’s temperature in the human health model. (See, for example, Breckenridge et

16

© M. Boyle 2001

al. (Breckenridge et al. 1995) for measuring the health of rangelands; and Svanberg(Svanberg, 1996) on ecosystem health indicators for the Baltic Sea.)

Evaluations of ecosystem health are based on an assumption of homeostasis(as is human health in a medical context). This assumption is generally notappropriate for ecological systems. A system could completely change from a forestto a wetland because of a river diversion, for example. The system may still beperfectly healthy, only it is now a wetland. To incorporate this range of possiblehealthy systems, the idea of ecological integrity has emerged to assess the system’scrucial characteristic – its ability to reorganize itself (Woodley et al., 1993). Thus,integrity includes responses to a changing environment, or even change from within,as long as its ability to self-organize is maintained (Kay, 1991; Kay and Regier,2000). Integrity is a concept that has been used most often in monitoring natural andprotected area e.g., (Noss, 1995; Soulé, 1995; Woodley, 1993).

Straddling the evolution of the ecological conceptual model, the integration ofsocial and economic dimensions has been recognized as important. A shift is alsooccurring in the human goals that motivate monitoring activities. People areincreasingly concerned with living sustainably and wish to assess progress towardthis goal. In the last few years, there has been an explosion of sustainabilitymonitoring initiatives e.g., (Hodge, 1996; Hodge et al. 1995; Maclaren, 1995;Nilsson and Bergström, 1995; Rees, 1996)

An effort to combine both the social and ecological is the "pressure-state-response" model (Adriaanse, 1993; International Joint Commission, 1996). Thismodel suggests three different, but related and integrated, types of indicators for eachsustainability issue. Pressure indicators measure pressures on the environment dueto human activities. State indicators measure the quality of the environment, and thequality and quantity of natural resources. Finally, response indicators measure how,and how well, society is doing in response to environmental changes and issues.The advantage of this framework is that it results in a group of indicators for eachissue that address the sources of the problem, its current status, and what is beingdone about it. The Dobris assessment, used by the European Environmental Agency,uses the DPSIR model of assessment (D: Driving Forces; P: Pressure; S: State; I:Impacts; R: Response). This model takes the "pressure-state-response" model onestep further by adding the human activities which are the "driving forces" behind the"pressure", and the "impacts" on human activities and the environment due to theenvironmental "state" change. (See the European Environmental Agency WWW site:http://themes.eea.eu.int/showpage.php//?pg=37183 )

The conceptual model we propose in this chapter integrates these earlier modelswith the notions of a systems approach to complexity.

4.2.2 The basis for the conceptual model: the notion of self-organization

Ilya Prigogine in his Nobel Prize winning work, showed that spontaneous coherentbehaviour and organization (e.g. tornadoes) can occur and are completely consistentwith thermodynamics (Nicolis and Prigogine, 1977; Nicolis and Prigogine, 1989). Thekey to understanding such phenomena is to realize that one is dealing with opensystems with a enduring (not necessarily constant) flow of high quality energy (that is,exergy). In these circumstances, coherent behaviour appears in systems almost

17

© M. Boyle 2001

magically. Prigogine showed that this occurs because the system reaches acatastrophe threshold and flips into a new coherent behavioural state. (This is evidentfor example in the vortex which spontaneously appears when draining water from abathtub.)

In examining the energetics of open systems, (Kay, 1984; Schneider and Kay,1994a; Schneider and Kay, 1994a) have taken Prigogine's work one step further. Anopen system with exergy pumped into it is moved away from equilibrium. But natureresists movement away from equilibrium. (This is the second law of thermodynamicsrestated for non-equilibrium situations.) When the input of exergy and materialpushes the system beyond a critical distance from equilibrium, the open systemresponds with the spontaneous emergence of newly organized behaviour that usesthe exergy to build, organize and maintain its structure. This dissipates the ability ofthe exergy to move the system further away from equilibrium. As more exergy ispumped into a system, more organization emerges, in a step-wise way, to dissipatethe energy. Furthermore these systems tend to get better and better at "grabbing"resources and utilizing them to build more structure, thus enhancing their dissipatingcapability.

The theory of non-equilibrium thermodynamics suggests that the self-organization process in systems proceeds in a way that: a) captures more resources(exergy and material); b) makes more effective use of the resources; c) builds morestructure; d) enhances survivability. These seem to be the kernel of the propensitiesof self-organization in systems. How these propensities manifest themselves aspositive and/or negative feedback loops is a function of the given environment(context) in which the system finds itself embedded, as well as the availablematerials, exergy and information.

This conception of self-organization, as a dissipative system, is presented inFigure 4. Self-organizing dissipative processes emerge whenever sufficient exergy isavailable to support them. The details of the processes depend on the raw materialsavailable to operate them, the information present to catalyze the processes, and thephysical environment. The interplay of these factors defines the context for (i.e.constrains), or the constraints on, the set of processes which may emerge.(Generally speaking, which specific processes emerge from the available set areuncertain) Once a dissipative process emerges and becomes established itmanifests itself as a structure. (In the case of a vortex in the bathtub water, the exergy is the potential energy of thewater, the raw material is the water and there is no historical information, thedissipative process is water draining, the dissipative structure is the vortex. The vortexwill not form until enough height of water is in the bathtub and if too much height ofwater is present the vortex flips into a different form of flow.)

In an ecological setting, examples of the structures are the individuals of species,breeding populations, forests etc. The processes are reproduction, metabolism,evapotranspiration etc. The context is the available set of nutrients and energysources in a physical environment. The information includes the genetic codes andbiodiversity at various scales.

18

© M. Boyle 2001

4.2.3 The Model

We have integrated the traditional conceptual models with the notion of ecological-economic systems as dissipative complex systems to produce a conceptual modelthat serves as a monitoring measurement framework (refer to Figures 5 through 7).In this model, the elements of the landscape (e.g. woodlots, wetlands, farms,neighbourhoods) that make up the societal and ecological systems, are seen as self-organizing entities set in an environmental context. Self-organizing entities areunderstood through consideration of their constituent processes and structures, andthe relationships between these. (For example, in a woodlot: processes would beevapotranspiration and growth of biomass; structure would be the species that makeup the woodlot; and a description of the relationship between these processes andstructure would be Holling’s four-box model of the ongoing process of birth-growth-death and renewal in ecosystems (Holling, 1986; Holling, 1992). The processesinvolve the flows of material, energy and information. The structures are the objects(i.e., trees) we see on the landscape. The processes allow for the emergence, andsupport, of structures which in turn allow for the emergence of new processes and soon. The recognition of this recursive relationship between process and structureseparates this conceptual model from more traditional ones.

Once the relevant processes and structures have been identified, a suite ofvariables may be specified which can be used to monitor the state of the processesand structures. These variables must be sufficient to define the well-being (in thecase of ecosystems, the integrity (Kay, 1991; Kay and Regier, 2000)) of the self-organizing entity.

Ecological communities provide exergy, materials and information required forhuman societies to sustain themselves. This is depicted in Figure 5. The societalsystem depends on the flow of exergy, materials and information from the ecologicalsystem to support its processes and structures. These flows, along with thebiophysical environment provided by the ecological systems, are the context forsocietal systems. The context constrains the possible societal processes andstructures in a specific location. While Figure 5 illustrates a single ecological systemproviding the context for the societal system, the reality is that it is a suite of adjacentecological systems (for example: woodlots, fields and wetlands adjacent to a farm)that provide this context. Alterations in these adjacent systems will alter the contextand thus the possibilities for the system in question.

However, the societal system can also influence the ecological system in twoways. The first influence is through changes in the structure of the ecological system(for example: cutting trees down in a woodlot, filling wetlands, and all the humanactivities which involve removing or dismantling ecological structures on thelandscape). Such actions, of course, alter the flows from the ecological systems tothe societal systems and thus create a feedback structure on the landscape. This isrepresented in Figure 5 by the lower arrow back from the societal system to thestructure in the ecological system. The feedback to the societal system occursbecause changes in the ecological structure change the context for the societalsystem.

The second influence occurs when the context of the ecological system is alteredby the societal system. For example, the runoff into a wetland or stream may be

19

© M. Boyle 2001

altered by human activities on adjacent properties. It is depicted in Figure 5 by theupper arrow from the societal system back up to the context of the ecological system.This influence is qualitatively different than the structural influence just discussed.

The resulting feedback loop has more steps and accordingly is more indirect. Bychanging the context of the ecological system, the societal system affects theecological processes and, in turn, the ecological structure and ultimately the societalsystem’s own context. For example, modifying the runoff into a waterway candramatically alter the character of the waterway, and hence the type of fish found in it,and therefore the sport fishery and associated economic system.

To summarize this discussion, each self-organizing entity resides in anenvironment that provides: (a) the biophysical surroundings in which the entity exists;and, (b) flows of exergy, material and information that the entity depends upon for thecontinuation of the self-organizing processes which maintain its structure. Thebiophysical surroundings, in conjunction with the flows into the system, constitute thecontext for the self-organizing entity. It is necessary to monitor this context, sincechanges in it will effect the processes of the entity.

Referring to Figure 5, the relationship between societal systems and ecologicalsystems is three-fold:

• Ecological systems provide the context for societal systems. That is, theyprovide the biophysical surroundings and flows of exergy, material andinformation that are required by the self-organizing processes of the societalsystems.

• Societal systems can alter the structures in ecological systems. (Forexample, cutting down a woodlot, removing beaver from a watershed.)Changes in the ecological structure can then, of course, alter the context forthe societal systems themselves.

• Societal systems can alter the context for the self-organizing processes ofecological systems. (For example, a change in the drainage patterns into awetland, or a change in the nutrient runoff into a shallow lake from adjacentagricultural fields .) Changes in ecological process can alter ecologicalstructure and consequently the context for societal systems.

Each of these represents a qualitatively different aspect of the relationship betweenecological and societal systems that should be considered when monitoring.

4.2.3.1 The Hierarchical Structure of the Model

Figure 5 applies to one hierarchical level, but as observed earlier, sustainabilityissues can only be understood in terms of “nested holons”. Figure 6 illustrates thisidea of nesting. On the ecological side, “local landscape” can be thought of as asubwatershed, for example. The hydrological cycle is an example of a process in thesubwatershed. The structures which make up the subwatershed are the ecologicalcommunities (e.g. woodlots, wetlands, open fields, etc.). The communities are in turnmade up of species. On the societal side, municipalities are scattered across thelocal landscape. These in turn are made up of neighbourhoods, which are made upof families and businesses.

20

© M. Boyle 2001

In many cases, the local subwatershed defines the context for the localmunicipality. However, the municipality can, and does, directly modify the ecologicalcommunities in the subwatershed and thus its own context. Similarly, the context forlocal neighbourhoods is determined by the adjacent ecological communities.Nonetheless, the local neighbourhood is quite capable of influencing ecologicalcommunities, through direct structural change (such as harvesting wood from awoodlot), or changing the context of an ecological community (for example, changingdrainage patterns into a wetland).

Figure 6 and the examples are meant to be illustrative and not exhaustive,although they do demonstrate that such changes can cascade through the nestedholons to ultimately affect individual families and businesses. What is unique aboutthis conceptual model is that it explicitly identifies these connections andrelationships and provides a framework for deciding what to monitor.

Figure 7 shows the full conceptual model which would be used as a template todevelop a particular conceptual model in which the important levels, processes,structures, environmental context, and influence/feedback considerations arespecifically identified.

Table 3: Aspects of the SOHO system which should be monitored. a) State of the well-being of the societal and ecological entities: measures of self-organizing processes

- examples: evapotranspiration, recycling measures of self-organizing structure

- examples: biomass, species diversity b) The context of the societal entities: measures of biophysical surroundings - example: average humidity measures of flows into the system exergy - example: fuel wood consumed materials - example: aggregate consumption

information - example: flow of hormone-like chemicals from theenvironment into food

c) Influences of the societal systems on ecological systems: structural influences

- examples: wetlands destroyed, woodlots removed contextual influences

measures of biophysical surroundings - example: average temperature measures of flows into the system exergy - example: phosphorous loading materials - example: drainage flows

information- example: hormone-like chemicals into the food chain

21

© M. Boyle 2001

4.2.4 The Conceptual Model and Monitoring

This conceptual model identifies a set of aspects which should be monitored, that is,should have indicators associated with them. These are summarized in Table 3. Ineffect, Figures 5, 6 and 7 define different points where a “monitoring meter” should be“plugged into” the ecological-societal system. A monitoring programme shouldinclude: a) Measures of the state of well-being of each of the self-organizing entities

(ecological and societal). These are represented in the figures by processand structure boxes.

b) Measures of the context for the societal system; that is, the biophysicalsurroundings and the flows into the societal system from the ecologicalsystem. This is depicted in the figures as the societal system in theshadow of the ecological systems.

c) Measures of the contextual and structural influences of the societal systemson the ecological. These are represented in the figures by the arrows goingfrom the societal system, up one level, and into the ecological structure.These arrows show how societal influences on the ecological systems,through changes these influences cause in the process and structure of theecological system, can alter the context for the societal system. So thesocietal influences on ecological systems are part of a feedback loop whichaffects the societal systems.

Bearing these aspects in mind, sustainability is about maintaining the well-beingof the combined ecological - societal system. This means maintaining their self-organizing processes and structures. It will happen naturally if we maintain thecontext for self-organization in ecological systems, which in turn will maintain thecontext for the continued well-being of the societal systems.

Thus, this conceptual model provides a framework for monitoring forsustainability that is more elaborate than the conventional approaches. Traditionally,one would monitor external (usually negative) stresses influencing the system and itsresponse. In this model, we emphasize the importance of the hierarchical monitoringof:

• both process and structure and their inter-relationship in order tounderstand the current well-being of the system;

• the context of the ecological and societal system in terms of the biophysicalsurroundings of each, and the flow of exergy, materials and information intothe system; and,

• the contextual and structural influence of societal systems on ecologicalsystems and hence ultimately on the societal system itself.

Without consideration of each of these aforementioned aspects, we cannot

decide where indicators may be placed to most effectively fulfil their purpose. Forexample, if the intent of a monitoring program is to anticipate future changes in orderto make wise management decisions in the present, indicators must be chosen tomonitor the factors which influence the processes and structure of the system.Measures focusing on the current state of the system will only reveal changes that

22

© M. Boyle 2001

have already occurred and will thus be too late. This is particularly true with regards tomonitoring to detect the effects of global change.

Also, monitoring within the boundaries of the defined system of interest is notsufficient to gain an understanding of its dynamics or to anticipate changes, both ofwhich are extremely important in protecting areas or certain land uses. We mustbroaden our scope to examine the context that constrains or influences the behaviourof the system, and do this from a hierarchical perspective.

When using this model it is important to bear in mind that self-organizationinherently involves internal causality. A self-organizing system has the ability tomaintain itself at an attractor despite changes in its environment. So, the environmentmay change substantially without the system exhibiting major change. Self-organizing systems can respond in a synergistic way to a collection of environmentalchanges. Thus, changes in the system can rarely be tied unequivocally to a specificenvironmental, policy or management change. Furthermore, such systems have thecapability to generate new behaviour that may emerge independently of changes inthe environment. The capacity to organize and maintain itself about an attractor is thehallmark of a self-organizing system. It is important to remember that discussions ofthe behaviour of this model cannot be in terms of one-to-one linkages of contextualelements (i.e., stresses) to the state of processes or structures. Rather, theexplanation must be in terms of morphogenetic mutual causality, that is, narratives(Kay and Regier, 1999).

Having described the issues motivating the monitoring program and theconceptual model which provides a framework for measurement, we now turn ourattention to the development of indicators.

Table 4: Indicator development activities.

• Generate indicator selection criteria. Selection criteria which capture the issues and constraints regardingthe desired set of indicators are used to screen potential indicators.They must be agreed upon by all those participating in thedevelopment of the monitoring program elements.

• Generate potential indicators. A set of potential indicators consists of possible measures whichemerge from consideration of the issues without regard forconstraints. All potential indicators should be recorded and evaluated.

• Select indicators.Not all potential indicators will meet all the criteria equally well andthere will always be trade-offs when choosing some indicators overothers. A method must be in place to assess the feasibility of potentialindicators and to make a subjective judgement on their evaluationagainst the criteria.

23

© M. Boyle 2001

Table 5: Indicator selection criteria (refer to Figure 8).

Categories of Consideration Indicator Selection Criteria

Conceptual Model of theSystem

Indicators should:- clearly relate to specific societal orecological elements of the SOHO description

Issues FrameworkIndicators should be:- clearly relevant to specific articulated goalsand objectives for sustainability

Knowledge BaseIndicators should be:- scientifically valid- statistically and analytically sound- demonstrated by case studies to bepractical

DataIndicators should use data that are:- available and accessible (easy to obtainand maintain at a reasonable cost)- accurate- comparable over time- complete with historical information (toestablish baseline conditions andthresholds)

ReportingIndicators should provide information that is:- understandable to potential users- unambiguous- easy to use- at the appropriate scale for decision making

4.3 INDICATORS AND THEIR DEVELOPMENT Indicators are often confused with a monitoring program. They are an important partof monitoring, but they must be generated, selected and implemented within thecontext of all the elements of a monitoring program. Indicators describe the status ofdifferent attributes of a system. However, the conceptual model is required to providea framework that integrates the indicators, as well as the data collection andcalculation methodologies and a process for synthesizing all the information into anarrative of the system. Otherwise, we cannot describe and assess the overall statusof the ecological-economic system in question.

Indicator development consists of three activities: enumerating indicator selectioncriteria, generating potential indicators and selecting indicators to be implemented forthe monitoring program. These activities are summarized in Table 4.

Figure 8 is schematic for organizing information while generating criteria andpotential indicators. It delineates a set of considerations that directly or indirectlyinfluence indicator selection, as well as the relationship between them. For eachconsideration (box, in Figure 8), criteria for indicator selection and questions to guidethe task of generating indicators are formulated. An explanation of the diagram

24

© M. Boyle 2001

follows, while compiled lists of selection criteria and questions for indicatorgeneration can be found in Tables 5 and 7, respectively.

The development of indicators is motivated by the need to build a narrative ofhow the ecological-economic system is developing so as to guide human activities.The sustainability narrative articulates a vision for the desired development of thesystem and identifies issues or concerns of import to the users of the indicators. Thisconsideration defines what we want to know about. As such, it also serves as theentry point for changes over time, in the issues or concerns of users, to beincorporated into the choice of indicators.

The conceptual model defines the structural and functional components of thebiophysical and societal systems and describes the linkages and feedbacks betweenthem. It provides a reference frame which identifies the appropriate scales at whichmonitoring should occur and for which indicators are needed. It also provides aframework which shows the relationships between the indicators and some sense ofhow useful or timely the resulting information will be. The issues framework extractsspecific management or user concerns and arranges them into a hierarchicalframework to associate them with the appropriate scale and type. Thus, theconceptual model and the issues framework are highly inter-dependent and bothfacilitate the decision of what to measure.

Our knowledge base also determines what we elect to monitor. Science, fromthe perspectives of biology, ecology, geology, hydrology, etc. aids in understanding theprocesses and structure of the system and the effects of specific influences orstressors. Complex systems theory adds valuable insight about how we perceive thesystem and what questions to ask when approaching sustainability issues. Statisticsoffers rules and standards for sampling and measurement as well as techniques fordata analysis and transformation that ensure quality results. Case studies allow usto test chosen indicators and evaluate their utility and feasibility in a practical setting.The amalgamation of these is the body of knowledge that helps us not only to judgewhat things are important to examine, but also how to make measurements andcalculate indicators accurately and consistently.

There are also many considerations attached to the data themselves that affectwhich indicators are chosen. Data availability encompasses considerations relatedto cost, collection, maintenance of databases, accessibility, and who is responsiblefor each task. The amount (i.e. historical extent) and the quality of data available arealso important.

Finally, the development of indicators must be done with an awareness of thereporting requirements and the needs of the users. This category providesparameters for the appropriate level of detail and the relative number of indicatorswhich affects the aggregation and/or integration of data to compose the indicators.

It should be noted that there are three independent considerations of hierarchywhich are relevant when developing indicators for a monitoring program. We must beconcerned with:

i) the scale and type of focus for the situation of interest;ii) the scale of data collection; and,iii) the scale of reporting the results.

25

© M. Boyle 2001

The scale and type of focus may be ascertained from the specific sustainabilityissue and is reflected in the issues framework. The scale of data collection need nothave anything to do with the level of focus for the sustainability issue since informationcan be aggregated, disaggregated, or integrated to be meaningful for decisionmaking. Likewise, the results may be reported at several different scales.

4.3.1 Indicator Selection Criteria

Using the above considerations as a framework and drawing upon the literature(Boyle, 1998), we generated the following list of criteria (Table 5) to be used inevaluating potential indicators. These are properties each indicator should possess.Note that these are relatively generic criteria which apply to many different monitoringinitiatives. For a particular project, however, there would be specific criteria judged tobe important that would be added to this list. Keep in mind that some importantcharacteristics of the information the monitoring program should provide (for instance“anticipatory” or “monitor for specific threats”) need not be a quality of every indicator.For a list of such characteristics refer back to the earlier discussion of "Attributes of aMonitoring Programme".

There are some criteria that, while desirable, are not necessarily essential. Thisrecognizes that trade-offs are inevitable. For example, an otherwise excellentindicator may require data which are expensive to obtain, or for which the necessaryhistorical data are not available. Thus, the list of indicator criteria may be used as achecklist to assist in indicator selection. Judgement, and careful consideration ofeach indicator, is still required. The use of indicator categories is one way toaccomplish this.

Table 6: Indicator categories.

Level I : Indicators available for immediate use.All essential criteria are met.

Level II: Indicators with demonstrated potential for use.Most criteria are met but, for example, data are not currently beingcollected, or there is a lack of historical data.

Level III: Possible indicators.There is evidence that these measures would be worthwhileindicators, but further scientific research and/or case studies arerequired to confirm their utility.

4.3.2 Indicator Categories

The purpose of grouping possible measures into categories is to distinguishbetween indicators that are immediately useful (Level I) and those that couldpotentially be used in the future (Levels II and III). Funding limitations, unavailabledata or untested measures are all considerations which would restrict the immediate

26

© M. Boyle 2001

implementation of an indicator. Those indicators placed in Levels II and III become a"wish list" that may become feasible at a later time. This method ensures that thesepotentially valuable indicators are not completely discarded. For some, the case couldbe made that the benefit derived from the indicator is worth the extra expense orresearch required in order for it become a Level I indicator. The final set of indicatorsis chosen after detailed feasibility investigations for each indicator have beenconducted and trade-offs have been considered. Table 6 defines the three indicatorcategories.

Note that potential indicators are associated with indicator categories on thebasis of an initial assessment. Only those indicators that are known to reasonablymeet all the criteria and could be immediately implemented are placed in Level I. Ifthere is some uncertainty about an indicator meeting the criteria, or it is known to belacking data or scientific validity, it is placed in Level II and III for further investigation.Thus, some potential indicators in Levels II and III may be discovered to be, ordeveloped into, Level I indicators at a future date.

4.3.3 Guiding Questions to Develop Potential Indicators

The considerations described in Figure 8 are synthesized, in Table 7, into a listof questions that assist the generation of potential indicators. We have found such aset of questions to be very useful in guiding the process of indicator development andencourage those involved in developing indicators to generate a set of guidingquestions appropriate for their own circumstances.

Table 7: List of questions for potential indicator generation.

1. What is the sustainability goal or objective?

2. What are the potential indicators?

3. Which aspect of the conceptual model does each indicatormeasure?

4. At what scale(s) will the indicators be reported?

5. What is the rationale for each of these indicators (i.e. what does it tellus about the sustainability goals and issues)?

6. Are there any confounding factors or problems which may limit theuse of the indicator?

7. Data Sets: What measurements are required? At what scale will thedata be collected? Who is responsible for data analysis andreporting?

8. Data Sources: Who has the data set? Who is responsible for itscollection?

9. Which category does the indicator fall into (Level I, II or III)?

27

© M. Boyle 2001

4.4 DESIGNING THE REMAINING ELEMENTSWe briefly discuss the design of the remaining elements as they are either very

case specific or are the subject of much literature in their own right.

4.4.1 A methodology for data collection and storage.

Carefully laid out procedures that address the practical and technical issuesinvolved in data collection must be established to ensure accuracy, consistency andstatistical robustness. This is especially important when different people will becollecting the data over time. An emerging design concern is the technical expertiseof those collecting data. Increasingly community groups and volunteers are engagedin data collection. The reliability of the data then very much depends on themethodology being designed to allow the keen but technically unsophisticated tocollect the data in a way which is replicable. In any case the design of themethodology must reflect the capabilities of the users of the methodology, if theresulting data is to be useful.

Equally important is the storage of the data, so that they are accessible andusable in the future. Much has been written on this subject, but in essence theremust be a way of recording the information that allows someone in the future (typicallya decade) to understand what measurements were taken in the field and to be able toconfidently assess their reliability and to repeat them. Failure to do this has oftenresulted in data being rejected simply because its reliability is unknown, thuseffectively wasting the time, effort, and money spent in collecting the data in the firstplace. (Wiersma B., 1995)

4.4.2 A methodology for calculating indicators.

The data collected will have to be manipulated in order to derive values for theindicators. Again, a method to do this accurately, consistently and in a statisticallyappropriate manner is required. This method must be easily applied by those doingthe calculations.

4.4.3 A process for synthesis.

Synthesizing the information that the indicators provide, into an overall narrativeof the status of the system is essential to completing the central task of the monitoringprogram, that is, to assess progress towards the human goals and aspirations whichmotivated the enterprise. The synthesis should in effect bring the conceptual modelto life. How one does this is an art form akin to story telling. As such, the form of thenarrative is very much dependant on the audience and thus is very case specific.

4.4.4 A methodology for reporting.

The values of the indicators and the results of the system assessment must bereported to the intended audience or users of the information. While it seems trite tosay this, a methodology for presenting it in a clear, purposeful, and timely manner fordecision making is crucial to the utility and success of the monitoring program. Manya monitoring report has collected dust on a shelf because the authors paid noattention to the issue of packaging and marketing it. In fact generating the report

28

© M. Boyle 2001

should be treated as a marketing exercise and it should be recognised that a numberof different "reporting" products may be necessary for different audiences. With theadvent of the World Wide Web and browsers with capabilities to access GeographicInformation Systems and Computer Aided Drafting software in quite transparent ways,some very exciting possibilities for reporting the results of monitoring programmesare emerging.

5. IN CLOSINGWe have argued that monitoring needs to be part of an adaptive ecosystem

approach if it is to contribute to the sustainability of ecological-economic systems.Monitoring provides feedback to the two other elements of an adaptive ecosystemapproach, governance and management. At the core of the approach is a conceptualmodel (describing the situation in terms of a self-organizing holarchic open (SOHO)systems description) that addresses the issues of complexity, and an issuesframework that integrates the hopes, aspirations and concerns of the various actorsinto a working definition of sustainability. A central tenant of the ecosystem approachis that the path to sustainability is one of tradeoffs. Science can illuminate thetradeoffs but a resolution, that is, the choice of path, is a political decision and in thedomain of governance. Thus, an ecosystem approach must bring together scienceand the socio-political economic realities of the situation in a synthetic and systematicway.

In an adaptive ecosystem approach, governance is seen as the ongoingprocess of resolving the tradeoffs and charting a course for sustainability.Management is about operationalizing sustainability. Monitoring provides feedbackon how the situation is actually unfolding and where it appears to be going.

An adaptive ecosystem approach cannot be undertaken without a monitoringprogramme. The reality of irreducible uncertainty associated with complex situationslimits our ability to predict how the situation will unfold, and hence where the path tosustainability lies. Thus, any successful approach to sustainability must not onlyemploy traditional anticipatory methods but also methods for adapting to the way inwhich the situation actually unfolds. The feedback provided by the narrative, whichresults from a monitoring programme, is the source of learning that is required for asuccessful adaptive approach to sustainability.

In this context the monitoring program must be capable of providing a narrativedescription of how the ecological-economic system is developing and theimplications of this for the path to sustainability. To do this requires collectinginformation that is clearly tied to the SOHO conceptual model and the issuesframework, and reporting this information in a form which can be used for governanceand management. Indicator development must be undertaken in this broadersystem’s context, and not as a purely scientific exercise. Otherwise, it is more likelythan not to produce yet another set of measures which satisfy disciplinary curiosity,but provide little input to the process of resolving the ecological-economic challengesconfronting our society.

29

© M. Boyle 2001

Adriaanse, A. (1993) Environmental Policy Performance Indicators - A Study ofIndicators for Environmental Policy in the Netherlands, The Hauge: Sdu UitgeverijKoninginnegracht.

Ahl, V. and Allen, T.F.H. (1996) Hierarchy Theory: A vision, vocabulary andepistemology, New York: Columbia University Press.

Allen, T.F.H., Bandurski, B.L. and King, A.W. (1993) The Ecosystem Approach: Theoryand Ecosystem Integrity, International Joint Commission.

Allen, T.F.H. and Starr, T.B. (1982) Hierarchy: Perspectives for Ecological Complexity,Chicago: University of Chicago Press.

Beckenstein, A., Long, F., Arnold, M. and Gladwin, T. (1996) Stakeholder Negotiations:Exercises in Sustainable Development, Chicago: Mirror Higher Education Group Inc.

Boyle, M. (1998) An Adaptive Ecosystem Approach to Monitoring. Waterloo.Environment and Resource Studies, University of Waterloo, Waterloo, Ontario, N2L3G1. URL: ersserver.uwaterloo.ca/jjkay/grad/mboyle/

Breckenridge, R.P., Kepner, W.G. and Mouat, D.A. (1995) A Process for SelectingIndicators for Monitoring Conditions of Rangeland Health. Environmental Monitoringand Assessment 36, 45-60.

Bunch, M.J. (2000) An adaptive ecosystem approach to the rehabilitation andmanagement of the Cooum River in Chennai, India. Waterloo. Department ofGeography, University of Waterloo, Waterloo, Ontario, N2L 3G1. URL:ersserver.uwaterloo.ca/jjkay/grad/mbunch/

Burgoyne, J.G. (1994) Stakeholder Analysis. In: Cassell, C. and Symon, G., (Eds.)Qualitative Methods in Organizational Research, pp. 187-208. London: SagePublications

Carpenter, S.R. and Cottingham, K.L. (1997) Resilience and restoration of lakes.Conservation Ecology 1, Article 2 URL: www.consecol.org/vol1/iss1/art2

Checkland, P. (1981) Systems Thinking, Systems Practice, Toronto: J. Wiley.

Checkland, P. and Scholes, J. (1990) Soft systems methodology in action, Toronto:Wiley.

Cocklin, C., Parker, S. and Hay, J. (1992) Notes on Cumulative EnvironmentalChange 1:Concepts and Issues. Environmental Management 35, 31-49.

Cormick, G., Dale, N., Edmond, P., Sigurdson, G. and Stuart, B. (1996) BuildingConsensus for a Sustainable Future: Putting Principles into Practice., Ottawa, Ontario,Canada.: National Round Table on Environment and Economy.

Environment Canada (1995) State of the Environment Report, Environment Canada.URL: http://www.ec.gc.ca/soer-ree/English/1996Report/Doc/1-3-1.cfm

Funtowicz, S. and Ravetz, J. (1993) Science for the post-normal age. Futures 25,739-755.

Gunderson, L.H., Holling, C.S. and Light, S.S. (1995) Barriers and Bridges to the

30

© M. Boyle 2001

Renewal of Ecosystems and Institutions, USA: Columbia University Press.

Hodge, R.A. (1996) Indicators and their role in assessing progress towardsustainability. In: Berger, A.R. and Iams, W.J., (Eds.) Geoindicators – AssessingRapid Environmental, pp. 19-26. Rotterdam: A. A. Balkema

Hodge, R.A., Holtz, S., Smith, C. and Hawke Baxter, K. (1995) Pathways toSustainability: Assessing Our Progress, Ottawa: National Round Table on theEnvironment and the Economy.

Holling, C.S. (1978) Adaptive Environmental Assessment and Management, Wiley.

Holling, C.S. (1986) The Resilience of Terrestrial Ecosystems: Local Surprise andGlobal Change. In: Clark, W.M. and Munn, R.E., (Eds.) Sustainable Development inthe Biosphere, pp. 292-320. Cambridge: Cambridge University Press

Holling, C.S. (1992) Cross-scale Morphology, Geometry, and Dynamics ofEcosystems. Ecological Monographs 62, 447-502.

International Joint Commission, I.f.E.T.F. (1996) Indicators to Evaluate ProgressUnder the Great Lakes Water Quality Agreement., International Joint Commission.

Kay, J. and Regier, H. (1999) An Ecosystem Approach to Erie’s Ecology. In: Munawar,M., Edsall, T. and Munawar, I.F., (Eds.) The State of Lake Erie (SOLE) - Past, Presentand Future. A tribute to Drs. Joe Leach & Henry Regier, pp. 511-533. Netherlands:Backhuys Academic Publishers

Kay, J., Regier, H., Boyle, M. and Francis, G. (1999) An Ecosystem Approach forSustainability: Addressing the Challenge of Complexity. Futures 31, 721-742.

Kay, J.J. (1991) A Non-equilibrium Thermodynamic Framework for DiscussingEcosystem Integrity. Environmental Management 15, 483-495.

Kay, J.J. and Regier, H. (2000) Uncertainty, Complexity, And Ecological Integrity:Insights from an Ecosystem Approach. In: Crabbé, P., Holland, A., Ryszkowski, L. andWestra, L., (Eds.) Implementing Ecological Integrity:Restoring Regional and GlobalEnvironmental and Human Health, pp. 121-156. The Netherlands: Kluwer

Kay, J.J. and Schneider, E.D. (1994) Embracing Complexity, The Challenge of theEcosystem Approach. Alternatives 20, 32-38.

Kay, J. (1984) Self-Organization in Living Systems. Systems Design Engineering,University of Waterloo.

Murray, T. and Gallopin, G., (Eds.) Some notes on: The Ecosystem Approach,Ecosystems as Complex Systems. Working Document No. 167, edn. 69-98p. Cali,Colombia: Centro Internacional de Agricultura Tropical. (1997)

Lister, N. and Kay, J.J. (2000) Celebrating Diversity: Adaptive Planning andBiodiversity Conservation. In: Bocking. S., (Ed.) Biodiversity in Canada: AnIntroduction to Environmental Studies, pp. 189-218. Toronto: Broadview Press

Ludwig, D., B. Walker, B. and Holling, C.S. (1997) Sustainability, stability, andresilience. Conservation Ecology 1, Article 7 URL: www.consecol.org/vol1/iss1/art7

31

© M. Boyle 2001

Maclaren, V. (1995) Developing Indicators of Urban Sustainability: A Focus on theCanadian Experience, Toronto: ICURR Press.

Nicolis, G. and Prigogine, I. (1977) Self-Organization in Non-equilibrium Systems,New York: J. Wiley & Sons.

Nicolis, G. and Prigogine, I. (1989) Exploring Complexity, New York: W.H. Freeman.

Nilsson, J. and Bergström, S. (1995) Indicators for the assessment of ecological andeconomic consequences of municipal policies for resource use. EcologicalEconomics 14, 175-184.

Noss, R.F. (1995) Maintaining Ecological Integrity in Representative ReserveNetworks, World Wildlife Fund Canada and World Wildlife Fund - United States.

Ramírez, R. (1999) Stakeholder Analysis and Conflict Management. Buckles, D., (Ed.)Ottawa, Ontario, Canada/Washington, D.C.: International Development ResearchCentre / The World Bank.

Rapport, D.J., Regier, H.A. and Hutchinson, T.C. (1985) Ecosystem Behaviour underStress. Am. Nat. 125, 617-640.

Rees, W.F. (1996) Revisiting Carrying Capacity: Area-based indicators ofsustainability. Population and Environment 17, 195-215.

Regier, H.A. and Kay, J.J. (1996) An Heuristic Model of Transformations of the AquaticEcosystems of the Great Lakes-St. Lawrence River Basin. Journal of AquaticEcosystem Health 5, 3-21.

Scheffer, M. (1998) Ecology of Shallow Lakes, London: Chapman and Hall.

Schneider, E. and Kay, J.J. (1994a) Life as a Manifestation of the Second Law ofThermodynamics. Mathematical and Computer Modelling 19, 25-48.

Schneider, E.D. and Kay, J.J. (1994b) Complexity and Thermodynamics: Towards aNew Ecology. Futures 24, 626-647.

Soulé, M.E. (1995) Biodiversity Indicators in California: taking nature's temperature.California Agriculture 49, 40-44.

Svanberg, O. (1996) Monitoring of biological effects. Resources, Conservation andRecycling 16, 351-360.

Warriner, K. (1997) Public Participation and Environmental Planning. In: Fleming, T.,(Ed.) The Environment and Canadian Society, Toronto: International ThomsonPublishing

Wiersma B. (1995) Looking for the Forest Among the Trees. Washington, D.C.:National Academy of Science.

Wilson, K.K. and Morren, G. (1990) Systems approaches for improvement inagriculture and resource management, New York: Macmillan.

Woodley, S., Kay, J. and Francis, G. (1993) Ecological Integrity and the Managementof Ecosystems, Florida: St. Lucie Press.

32

© M. Boyle 2001

Woodley, S.J. (1993) Monitoring and Measuring Ecosystem Integrity in CanadianNational Parks. In: Woodley, S., Kay, J. and Francis, G., (Eds.) Ecological Integrity andthe Management of Ecosystems, pp. 155-176. St. Lucie Press

33

© M. Boyle 2001

Figure 1: Monitoring in context. Within the adaptive ecosystem approach,monitoring, management and governance are interdependent activities. Acomprehensive monitoring program cannot be established without describing theecological-economic system of interest (conceptual model) and the societal goalsand concerns (issues framework).

35

© M. Boyle 2001

Figure 2: An adaptive ecosystem approach to sustainability and health.

37

© M. Boyle 2001

Figure 3: The monitoring process. The development of monitoring programelements and the results of indicator interpretation are used to develop a narrative ofthe current system status. If this differs from the sustainability vision, we must learnwhat this tells us about the system and modify our perception of the system andactivities accordingly. The monitoring program must also be evaluated to ensure thatit provides the necessary information to make sound decisions and meet user needs.

39

© M. Boyle 2001

Figure 4: A conceptual model for self-organizing systems as dissipative structures.Self-organizing dissipative structures emerge whenever sufficient exergy is availableto support them. Dissipative processes restructure the available raw materials inorder to dissipate the exergy. Through catalysis, the information present enables andpromotes some processes to the disadvantage of others. The physical environmentwill favour certain processes. The interplay of these factors defines the context for(i.e., constrains) the set of processes which may emerge. Once a dissipative processemerges and becomes established it manifests as a structure. These structuresprovide a new context nested within which new processes can emerge, which in turnbeget new structures, nested within which … Thus emerges a SOHO system, anested constellation of self-organizing dissipative process/structures organized abouta particular set of sources of exergy, materials and information embedded in aphysical environment. The canon of the SOHO system is the complex nestedinterplay and relationships of the processes and structures, and their propensities,that give rise to coherent self-perpetuating behaviours that define the attractor.

41

© M. Boyle 2001

Figure 5: A model of a single holarchical level of ecosystems and theirinteractions. Dotted lines represent one system forming the context for another. Thegreen arrow represents direct societal influence on the ecological system (i.e.,changing structure). The red arrow represents indirect societal influence on theecological system (i.e., changing the context for the ecological system whichcascades down to change the societal system).

43

© M. Boyle 2001

Figure 6: An example of components and interactions over three holarchicallevels. The “stacked deck” effect is a reminder that each level is made up of aconglomeration of defined systems. That is, many species together comprise anecological community and many communities together form the local landscape. It isthe aggregation of these local landscapes that makes up the landscape mosaic of aregion (such as a province or state). On the societal side, families and businessescomprise neighbourhoods. Municipalities are made up of neighbourhoods andfinally, the province/state is politically divided into municipalities and counties. Notethat this diagram demonstrates only one possible way of parsing the system.

45

© M. Boyle 2001

Figure 7: The Big Picture - Interrelationships and influences between ecological andsocietal systems in the biosphere. The largest two arrows represent the relativelynew ability for humans to influence the biosphere and the global context.

47

© M. Boyle 2001

Figure 8: Considerations for indicator development. Arrows and accompanying textrepresent influences on the development of indicators. The main issues to bedecided for particular categories are also listed.