Le succès d'une gestion des écosystèmes naturels requiert une connaissance approfondie des différents processus qui interviennent et de leurs échelles de temps et d'espace particulières. Pour cette raison, les décideurs ont besoin d'analyser une vaste gamme de données et d'informations géographiques. Les modèles mathématiques, les systèmes d'informations géographi-ques et les systèmes experts sont capables de produire cette analyse, mais seule une minorité de gestionnaires les utilise actuellement. Cet article identifie quelques unes des raisons à l'origine de l'hésitation des gestionnaires à adopter de tels outils d'aide à la décision pour la gestion des ressources naturelles et propose une structure qui pourrait faciliter leur utilisation pour le processus de prise de décision. Cet exercice est réalisé à l'intérieur du contexte de la gestion intégrée par bassin. Une revue des systèmes d'aide à la décision est également présentée.
- Gestion intégrée,
- système d'aide à la décision,
- système d'information géographique,
- modèles mathématiques,
- système expert,
- organisation du travail,
- ressource en eau
Many methods of integrated or watershed management exist which account for the necessary biophysical and socio-economic factors at the watershed level. Some of these approaches are ecosystem oriented while others are socio-economically oriented. Whatever the definition, water management at the watershed level needs to account for a plenitude of variables related to the air, water, soil, biology, and economy. The successful management of natural ecosystems requires a thorough understanding of their characteristic time and spatial scales. Because of this, decision makers need to analyze a wide range of data and geographic information. Mathematical models, geographic information systems and expert systems are capable of performing this analysis, but only a minority of managers are currently using them. This paper identifies some of the reasons why ecosystem managers have been slow to adopt such decision support tools in natural resources management and proposes a framework to facilitate their use in the decision making process. This is done in an integrated watershed management context. A review of related decision support systems is also presented.
Four types of decision-support tools are introduced : mathematical models, expert-systems, geographical information systems (GIS) and decision support systems (DSS). Mathematical models have long been used for simulation, prediction, and forecasting, however, they are often task specific and were rarely developed for management uses. GIS are more and more commonly being used for decision support as they become more affordable and user-friendly and are very well-suited for managing resources at a spatial scale. There exist many kinds of software ranging from a simple viewer used for cartographic purposes to complex GIS oriented toward spatial analysis and modelling. Expert systems are also interesting for decision support when specific goals are being considered. Finally, DSS are perhaps the digital tools most applicable to management purposes, often integrating one or more models, a GIS or expert system functionalities. There are two types of DSS :
1. Environmental Information Systems (EIS), and
2. Integrated Modelling Systems (IMS)
EIS can be very user- friendly, relying heavily upon GIS and statistical functions.
IMS also use GIS capabilities, but integrates several mathematical models as well. The level of integration between models varies considerably and the complexity of IMS are generally high.
Two questions underlie the operational use of digital technologies for decision support. The first is whether or not such technology should be used at all, while the second is why such tools take time to be adopted by government and management agencies. The use of digital technologies is often required when the problem is complex and where there are a wide range of factors involved with different spatial and temporal scales. Three major constraints towards the implementation of decision support tools can be pinpointed :
2. data, and
3. working organization.
Technological constraints include cost, lack of user friendliness, and hardware problems, among other factors. Data constraints are mostly related to availability, cost, heterogeneity and volume. Finally, organization constraints pertain mostly to the manager's perception of the tool and the structural integration of the tool within the decision process.
This paper proposes a 4-step approach to optimize the use of decision-support tools. The first step requires that managers and decision-makers clearly define their project, goals and budget, as well as, decide whether to use an integrated watershed management approach or a more discrete approach. This leads directly to the second step, which consists of choosing the most appropriate digital support tool. This requires communication between managers and scientists, and at this point, data gathering and integration should begin. The third phase consists of the development of a new tool or adaptation of an existing one within the context of the agency's management structure. The final step is the operational use of the decision support tool by the agency, following an initial trial period. The successful use of a decision support tool for management purposes depends on proper planning that accounts for all factors related to management needs, budget, data, ease of use, and organization integration.
- Integrated management,
- decision support system,
- geographic information system (GIS),
- mathematical models,
- expert system,
- structural organisation,
- water resource