"In sustainable development, everyone is a user and provider of information considered in the broad sense. That includes data, information, appropriately packaged experience and knowledge. The need for information arises at all levels, from that of senior decision makers at the national and international levels to the grass-roots and individual levels. The following two programme areas need to be implemented to ensure that decisions are based increasingly on sound information:

(a) Bridging the data gap;
(b) Improving information availability." (Agenda 21, p. 40, UNCED, 1992)

"Sustainable development" has come to summarize the acknowledged importance of nondestructive land use. The idea has come to be widely accepted -- perhaps because of its inherent "constructive ambiguity" -- or perhaps because, like motherhood and apple pie, it is simply a notion that is hard to argue against. But unlike motherhood, it is not something to which an irrevocable commitment can arise from a moment of irrational passion and, unlike pie, it has no simple recipe. The challenge is: how can dynamic communities with changing needs, aspirations and technologies maintain a nondestructive linkage with an environment that is itself dynamic and constantly changing? This clearly requires an adaptive process, and, in the time frame that matters to us now, that adaptive process needs to be based on human intelligence and environmental information. As Agenda 21 notes, there are many forms of information (conventional and non-conventional) and many different "users" of that information all of whom contribute to the net impact of human activity on the environment. Finding ways to optimize the use of available information and ensure that all providers and users of information have effective links to decision-making processes is an essential step towards sustainable development. GIS provides tools to discover, analyze and communicate the spatial relevance of data and information. A critical question -- the theme of this workshop -- is how can high technology tools of information management be brought into the public forum in way that they increase fairness and competence (Webler, 1995) in decision making rather than increasing polarization and marginalization?

This paper describes community-based research intended to bring local spatial information into public consciousness and build local capability to manage and use that information. It focuses on two initiatives in mountain forest villages: one is Invermere, British Colombia, Canada, located between the Rockies and the Purcell Mountains within a few miles of the headwaters of both the Kootenay and the Columbia Rivers in an area that is very diverse ecologically, largely unspoiled, but under competing land-use pressures. The other is Huitzilac, Morelos, Mexico, in an area of spectacular mountain forests less than two hours drive south of Mexico City. The paper explores two particular issues: barriers to information flow (Meredith, 1997a) and the impact of access to information on the dynamics of communities (Meredith, 1997b). In both cases, the projects are managed by groups that involve academic researchers and local citizens. The unifying context for these two case studies, mountain-forest communities experiencing rapid change, is described first. The conclusions of the paper are three: that public participation is essential; that with GIS, the best technology will always, by definition, be ahead of the public’s ability to participate; and with PPGIS, the "process is the product", that is, by the time the public has become involved in generating or understanding a system, the educational and analytic benefits of public participation may already have been achieved.

Biodiversity protection and the role of rural communities.

Anthropologist John Bennett wrote that the requisites for achieving sustainability [a dynamic balance between resources and sustenance] are a nothing short of a "restructuring of human purpose and a total reassessment of cultural, political and moral problems." (Bennett, 1993). If environmental management "decisions [are] about human behaviour rather than physical things" (Grumbine, 1997), public participation in environmental planning is essential (Fisher, 1996, Pepper, 1996).

Rural communities are the custodians of the land. They are often economically dependent on the exploitation of ecological resources and, at the same time, inhabited by people with a great appreciation of the landscape. This sometimes leads to conflicts -- in the worst cases to a "downward spiral" of environmental degradation that leave "habitats half protected, rural economies weakened and personal principles bargained away " (Johnson, 1993). Exploration of participatory decision-making is needed. Community-level environmental decision-making involves accessing data and information, interpreting it and communicating it. Barriers to salient information make this impossible.

Barriers to information flow: the case of Invermere B.C.

Invermere is an excellent place for this type of work. The geologic diversity, elevation range and aspect variability make it very diverse ecologically: from vast permanent wetlands to semi- arid grassland benches, dry Douglas fir forests, montaine spruce-fir forests and alpine tundra all within a few miles. Despite economic strategies ranging from fruit production, mining, forestry, ranching and tourism, the area has remained relatively undeveloped and has attracted residents who are drawn by, and appreciate, the generally unspoiled landscape.

The economic, recreational and aesthetic character of the community are bound up in the environmental quality and so, understandably, the range of perspectives on environmental issues is diverse. Local stakeholders are involved in commercial activities such as forestry, ranching, and nature-based tourism and in personal activities such as hiking, hunting, fishing or simply nature appreciation. Clearly, differences in values will cause disagreements between stakeholder groups. For example, clearcut logging is simply seen in different ways by loggers -- whose livelihood derives from the practice and who can point to healthy second growth forests as proof of the viability of the practice -- and, say, amateur naturalists -- who see nothing but ecological wasteland in the clearcuts and simplified artificial monocultures in the second growth. These value differences may be very difficult to overcome.

But in addition to differences in values, difference in perception of facts can also cause disagreements between stakeholder groups, and these differences can more readily be overcome through information management. This is the intention of the Invermere study. The conditions leading to the study are these: over the past several decades there have been a number of environmental conflicts. For each of these, a process reasonable well described by Downs’ (1972) issue-attention cycle ensues -- the problem is identified, alarm spreads through the community, coalitions are formed to act against perceived threats, the issue is either resolved or the costs of resolving are discovered to be unrealistically high, and people lose interest in the issue. This cycle requires a "reinvention of the wheel" at each crisis and all of the start-up costs and steep learning curves are re-encountered. Our project was intended to reduce these.

There are two GIS-related facets to the project. The first is to create an environmental atlas which will help present information about local environments and thereby help support community-based environmental decision-making (examples to be shown). The procedures, in brief, were to involve members of the local community in identifying (a) priority issues, (b) data needs (c) data sources (d) information "targets" and (e) communication strategies. The second is to produce a dynamic land-cover change map for the region. These exercises have led to the discovery of a number of disempowering realities which can be considered as "barriers to information flow."

Barriers to Information Flow

1. Dispersion of data sources

The most obvious barrier to using information or data is simply lack of awareness of the existence of the information or about even the class of information. ( e.g. a conservation group interested in demonstrating the temporal acceleration in the loss of wetlands was unaware of archived air photos as a class of information). The means of addressing this barrier is not to make more information available but rather to make it easier to determine what relevant information is available.

In the case of the environmental atlas, this proved to be one of the most challenging aspects. The steps from discovering what information is rumored to exist to actually having it in hand are many (some are treated below). Simply amassing an inventory of reliable, current and relevant data meant canvassing agencies of three levels of government as well as crown corporation and private companies (logging and mining firms), international agencies and non-government organizations (bird, wildlife, hiking and hunting groups). Each new data source identified seemed to open the possibility that others might exist. The sorts of investigative effort (time, cost and skill) is not within the grasp of most communities. This suggests that inevitably, even when an effort is made, the data that become available may be random or desultory.

2. Legal barriers

Primary data about forest inventories are won at great expense. It makes no sense for expensive data to be collected as many times as there are potential users. Yet, equally, there is no obvious basis for open access. Forestry inventories that logging companies may require for economic planning may also have considerable significance in environmental conflicts. Private survey data and opinion polls, likewise, create data sets that may not be openly accessible despite their potential importance, and information about corporate plans or activities (past or present) may be closely guarded. Census data cannot legally be disaggregated to the level that, for example, would make herbicide and fertilizer-use data meaningful at the local level. This mean that each stakeholder group works with a maximum data set that is subset of the total. Disparities may be significant. In our own work it was not uncommon for individuals to be willing to share legally restricted information unofficially, but that makes subsequent use of that information complicated.

3. Financial barriers

Financial limits to data access are inevitable, whether the limit is encountered at a very low level where, for example, it may be impossible to hire a project worker to conduct basic background library searches, or whether it is at a high level of, for example, not being able to purchase proprietary information or to hire experts (legal, technical or scientific). An important subcategory is data collected with government support and then sold on some sort of cost recovery basis. Satellite imagery is the most relevant example. The real costs of collecting images are high but the marginal costs of using them once collected are negligible. There is some question as to how those data can best serve the public interest.

4. Technical barriers

The rate of change in electronic data acquisition, storage, transmission, analysis and presentation is such that only trained specialists stay at the cutting edge of progress. Clearly, it is not possible for all potential users to acquire and maintain the requisite technical skills to use them. This technical barrier is inevitable. The question is not whether, but rather where, it exists and what its implications are. The problems of rectification, pixel size differences, band differences, an apparent error in positioning and then haze correction and aspect compensation meant that the community-based work we proposed sank deeper and deeper into the technical space of our GIS lab and further from the understanding of community partners. This limited community control of the process as well as their capacity to verify results.

5. Paradigms of interpretation

Each site-specific environmental process is unique and so the future, as it arrives, is not fully knowable. Science is systematically trying to collate new data into rational models and the lay public is having to establish new paradigms of understanding that incorporates the new evidence. There are elements of local environmental change that may appear disparate and unconnected, or connected in some incomprehensible way, but which are, in fact, consistent with existing theories or models. These models may be known by scientific specialists, by local resource users, in traditional ecological knowledge or in folk wisdom, but they may not be readily accesible to all stakeholders.The relationships, for example, between forest ecology and stream hydrology or watershed nutrient budgets is vital to planning, but without models that describe the relationships, the available evidence loses power.

6. Non-conventional data

The outcome of negotiations are often predetermined by the definition of the context, the terms of reference and the pivotal issues. Sometimes, at least from a distance or in retrospect, the issues that become pivotal appear to be the wrong ones. The ability to set the agenda of a negotiation process may be the single most important part of the negotiation, and so naturally each stakeholder group will try to set the agenda according to its own perceived rhetorical strengths. Community groups have access to many forms of data and information that are qualities of the community itself and are therefore not available from any source other than from the community. Traditional ecological knowledge, local spirituality, aesthetic and amenity values are relevant examples. These values may be downplayed by stakeholders who have other value sets and priorities or who cannot be shown in quantitative, commensurable, terms concrete evidence of the values. This is a question of competing strategies. If environmental negotiations are couched in the established frameworks of the legal profession or the scientific community, local community groups may be accepting, a priori, a handicap.

7. Barriers of presentation

The soundest argument means nothing if it is not presented; the flimsiest argument can appear compelling if it is coherently and professionally presented. Overcoming each of the previously discussed information limitations presupposes an outlet for the results. The skills of presentation must be seen as an essential part of community-based action.

In a sense, this list of barriers represents reasons to suggest that the hope of effective and equitable access to information can become a reality. And yet it is clear that identifying the barriers does not make the real situation any worse than it already is and that attempting to overcome the barriers in any local situation can only reduce the gap between the real and the ideal. The exercise in Invermere shows the size of the "information mountain" that needs to be climbed, but it also helps move the community part way up the curve.

Changing socio-cybernetics: the case of Huitzilac, Morelos.

The second issue relates to the conceptual framework for spatial decision-making. The concepts of systems theory, and in particular of cybernetics, provide an analytic paradigm for assessing the role of environmental decision-making. An interesting question is in regard to the distinction between first and second order cybernetics. First order cybernetic systems are those external to an observer; second order are those of which the observer is part. In second order systems, the observer’s understanding of the system becomes part of the system. Clearly, community-level information managers are part of the environmental system they are managing.

The concept of "rational expectations" recognizes that the way economic systems function is based not just on externally measurable and/or quantifiable parameters of the economy, but also on what human members of the economic system know about those measures. This radically altered economic research (and arguably, its relevance and impact). So might the recognition of the role of community-level information users alter the theory and practice of environmental modeling and planning. For example,

land cover change modeling based on Markov chains or on logit regression assume that what has happen in the past will happen again. But as Scott Adam (1997) glibly puts it: "any doom that can be predicted won’t happen." If exploring land cover change at the community level changes the perception and awareness of the agents of change, the processes will be different.

Viewing community-ecosystem interactions as cybernetic systems can shape our understanding of environmental problems and solutions. Cybernetic processes can generate stable conditions in environmental systems but they can also support constant adaptive change. It raises the question, as asked by Felix Greyer (1994) : should science support concentrated technical capability and therefore centralized planning or, rather, "strive to improve the competence of actors at the grass roots level so that these actors can steer themselves and their own environment with better results?" Greyer's own evidence strongly supports the latter.

But this alternative view also raises challenges. Greyer asks, "how does one obtain reliable data within ... a framework where nothing is constant and everything is on the move?" He concludes that "the best one can do ... is to accept that there is not any one desirable and sustainable state for society -- only near continuous transition, often coupled with the impossibility to forecast even the near future ..." Greyer describes this situation as one in which "the system -- whether an individual or a social system -- collects information about its own functioning, which in turn can influence that functioning." Greyer goes on to note that "the minimum requirements [for this to occur] are self-observation, self-reflection and some degree of freedom of action." This corresponds to the perceive-interpret-respond sequence of the adaptive model. These questions seemed helpful to understanding the Mexican case study.

The site in Mexico was selected because of local concerns about changes in, and current pressures on, forested areas that lie immediately to the south of Mexico City. These areas "which had been until recently little affected by the urban expansion, have now succumbed to rapid development" (Ezcurra, 1990). The urban footprint has expanded from less than thirty km2 in 1910 to almost 1200 by 1990 (Ezcurra and Mazari-Hiriart, 1996). The forests have been protected by inherent properties of the landscape: the mountains rise steeply to almost 3500 m, many areas are quite inaccessible and soils are young, thin and easily eroded. Adjacent cities, divided by the mountain forests, are spreading upwards into the mountains and threatening the mountain villages and forests.

We began our study with the knowledge that there were three information sets that had to be a part of the local "cybernetic process:"

1) knowledge about the local environment

1.1. making sense of the data: It is often difficult to interpret available information. For example, in Morelos there has been concern about the spread of cholera. Cholera is an environmental problem that links waste, land use, hydrology and water use. Geographic information can help show how forest clearance affects stream flow, pollutant transport, and consequently, the spread of the disease. In this case we generated simple images indicating the location of deforested areas, slope steepness and proximity to streams, and that related stream positions to sources of contamination and downstream sites of water extraction. In many instances these connections may be readily apparent but are not articulated or factored into decision making. In other cases ecological connections may not be clear to local resource users. The need is to make them part of the general understanding that shapes land use practices.

1.2. temporal scale - the past: Individuals live in ecological and geological time but we think in human time. The mountains of the Chichinautzin are about 400,000 years old. The Pre-Hispanic civilization may have had some impact on the area for about 1% of that time, recorded history is about 0.1% of that period, and living memory, about 0.01%. The planning horizon is perhaps 0.001%. Ecological changes that take place over a life span may be imperceptible to casual human observers or may be considered insignificant because they are so slow. In this case we were able to locate pre-existing maps of forest cover decline and superimpose areas of forest loss on colour composite satellite images of the region. A series of images that show the present state of forest cover, and losses over twenty years relative to the natural cover conveys a powerful sense of the impact of recent activities. 30 GIS-based time series analyses and simulations help show how ecological processes such as primary succession following mountain building compare in time with human activities such as land clearance.

1.3. temporal scale - the future: Because ecological systems are complex, it is often difficult to predict the effects of any single action. And when there are two or more simultaneous activities, prediction of effects can be almost impossible. For example, how can the effects on ground water in Chichinautzin of a loss of 10% of the forests, 10% more domestic waste, and 10% less rain be predicted? One possibility is the use of GIS to experiment with such possibilities. While some ecological systems models are very complex, may be difficult to use, and unreliable in their output, some projections are possible. For example, it is very simple to demonstrate what the region will look like if residential expansion continues at the same rate for the next 20 years, or if soil extraction along roadways leads to a spread of erosion at an average of two meters per year, or if as much forest is lost in the next 20 years as was lost in the last 20. GIS based tools permit local investigation of alternative scenarios and can make the future clearer and decisions easier to reach.

2) knowledge about the external connections

2.1. spatial scale: Environmental resources are important in different ways. The forests of Chichinautzin are a source of fire wood and medicinal plants for local people; they are an important regional source of building materials; they regulate water supply for people in the south of Morelos; they serve as recreational and residential sites for the population of Mexico City; and nationally and internationally they are both genetic resources and carbon sinks. Spatial data were used to demonstrate two links between local conditions and their contexts: one with water management in the south of Morelos (showing how small local drainages are linked with major rivers that supply other regions) and the other with the increasing extent of urban land adjacent to the forest region. Visualization of this expansion demonstrates the probable consequences of laissez-faire land management, but it also shows the potential for alternative resource uses (such as educational and recreational uses) that are compatible with protection of the environment.

3) knowledge about the decision making process

3.1. communicating priorities: Land use decisions almost always involve compromises. There are competing uses (forest, farm, commercial or residential) and competing users (the rich - the poor, the local - the outsider, the conservationist - the developer). Very often the way decisions are made depends on how well information can be presented in a public forum. Communication skills empower because they allow some control of agendas. Other stakeholders such as governments, developers and corporations rely heavily on geographic information. Communities should have access to the same powers of communication.

What has been most interesting about this work is the extent to which initiating, focusing and participating in community-based discussions has influenced the community. Consider the difference in impact if exactly the same steps in data management had been taken by an outside agency. It would have looked at rates of forest conversion, population trends, land use trends, stated policy objectives and other data sets and made predictions about what was likely to happen and, given what they had been told about what was desirable, they would have recommended policy action to convert what is happening on the ground to what they think should be happening. This is first order cybernetics. Instead what has happened (though of course it is in early stages) is that there as been an evolving self-regulation system emerging as people begin to think about factors that affect them directly, think about the data that are linked to those factors, think about the independent variables that drive change and think about avenues of response. For example, the first issue diagnosed as a local concern was waste management. By the time structures were in place to collect reliable data about the nature, scale, causes and consequences of the waste issue, people in the community had become "waste conscious" and had therefore begun to change the circumstances we were proposing to study. The tools of study become an educational -- and therefore an adaptive -- part of the system the study was directed at. It is self-reflective and represents second order cybernetics.

The lesson form this study shows that first order and second order cybernetic approaches to environmental data management are completely different in their impact. From the standpoint of simplicity of design, ease of execution and detached simulation, studies that do not include communities as partners are preferred. But just as rational expectation theory attempted to reconnect economics with the real world -- and in so doing made it "messier" and more susceptible to the vicissitudes of human will -- so must community-based environmental research demonstrate that community participation in information management is essential to sustainability.

Three observations can be derived from these case experiences. The first is that, given the role of human activity in shaping the local environment, and of human value in determining "successful" outcomes, public participation in conservation decision making is essential. The second is that in a rapidly evolving field like geographic information analysis, state-of-the-art is by definition going to be somewhat inaccessible to the general public. If the "best available technology" is desired, part of the design of an effective PPGIS has to be mechanisms for including effective liaison between specialist managers or resource personnel and the general public. The last observation, evoking Marshall McLuhan’s famous dictum "the medium is the message," is that at least in some cases of PPGIS, the "product is the process." The medium of GIS requires new ways of thinking about environments and about spatial data. Programs that bring the public into the new medium change the public. The message of PPGIS is received de facto. Where this message is of new understandings of, or concern about, conservation issues, it may well be that the goal of empowerment through PPGIS has already been achieved by time the a system is up and running. In fact it may not matter if the system never does get up and running. The process of building local capability and motivation -- a process that is part of the set-up of a system -- may be the very product we are hoping to produce.

References

Bennett, John. W., (1993). Human Ecology as Human Behavior. Transaction Publishers. New Brunswick (USA). 357 pp.

Downs, A. (1972). Up and Down with Ecology -- The Issue Attention Cycle, Public Interest 28, 38-50.

Ezcurra, Exequiel, (1990), The Basin of Mexico. In: B.L. Turner II (ed.), The Earth as Transformed by Human Action. Cambridge University Press, Cambridge. pp 577-588.

Ezcurra, Exequiel and Marisa Mazari-Hiriart, (1996) Are Mega Cities Viable? A Cautionary Tale from Mexico City. Environment. 38-1, 6-26.

Fisher, Julie. 1996. Grassroots Organizations and Grassroots Support Organizations. in Moran (ed.) q.v. pp57-101.

Greyer, Felix, (1994), The Challenge of Sociocybernetics. Paper presented at 13th World Congress of Sociology, Bielefield.

Grumbine, Edward, (1997), Reflections on "What is Ecosystem Management?" Conservation Biology 11:1 41-47.

Holling, C.S. 1978. Adaptive Environmental Assessment and Management. John Wiley. New York, USA

Johnson, Kirk, (1993), Reconciling Rural Communities and Resource Conservation. Environment 35:8 16-28.

Meredith, T.C.(1997a). Information Limitations in Participatory Impast Assessment. In : John Sinclair ed. Environmental Impact Assessment in Canada. University of Waterloo Press. pp 125-154.

Meredith, T.C. (1997b), Making Knowledge Powerful: Mexican Village Project Uses Environmental Information Technology to Strengthen Community Voices in Biodiversity Consrvation. Alternatives 23:4. 28-35

Moran, Emilio F., 1996. Transforming Societies, Transforming Anthropology. Univeristy of Michigan Press. Ann Arbor. 361 pp.

Pepper, David, 1996. Modern Environmentalism. Routledge. London. 376 pp.

Webler, Thomas. 1995. "Right" Discourse in Citizen Participation. pp 35-88. In Renn, Ortwin, Thomas Webler and Peter Wiedermann, (1995). Fairness and Competence in Citizen Participation. Evaluating Models for Environmental Discourse. Kluwer Academic Publishers. Boston. 381 pp