Contenu du sommaire : Système d'information géographique et gestion de l'environnement

Geographical information systems are not software tools but information systems created to respond to precise needs and to establish a definitive modelling of the real world. Among the four components of a G.I.S. (methods, technical tools, information and structure) the text presents the two which are the most undifferentiated. The technical component integrates various and often multiple forms of software. The information component is organised around a data base which is either primary or derived in character, and of internal or external origin ; it is the quality of this data which determines the ability of the system to respond to its objectives. Geographical Information Systems can be classified on a continuum in relation to their intrinsic characteristics according to a sequence of development "inventory/analysis/decision", or according to the decisional level to which they correspond. Environmental geographical information systems have developed in the context of complex models. This complexity relates both to the description of objects and the modelling of their operation and dynamics, as well as to the need in decision making to take account of numerous actors who are often in disagreement. Imperceptibly geographical information systems have become an aid to spatial decision making. This should lead geographers to implicate themselves much more in this field at both a conceptual and operational level.

The experience of the Hautes- Chaumes du Forez Geographical Information System underlines many conceptual and organizational issues relating to the elaboration of an environmental G.I.S. project on a specific territory. Such problems are due to the information diversity, but also to the difficulty to state precisely each responsibility between technical, scientific and political choices that are to be made at every stage of the project development. The first step, the database realization, is a mainly technical work. The second step, based on sensitivity and suitability analysis, can not be carried out without a permanent dialog between political and scientific actors. The third step, modeling and elaboration of scenarios of evolution, requires clear and precise political choices.

Ecological mapping at a scale of 1/50 000 of a rural area of 3 000 sq.km. provides the basis for the five-yearly revision of the development plan. Land use, the base map, the map of infrastructures, and ecological maps have been digitized and incorporated in software for spatial analysis. Numerous thematic maps, with varying degreas of complexity, have been produced. These include constraints on development, the potential for agriculture and forestry, flood risks, danger of aquifer contamination, the erodability of soils and archaeological potential. This analysis, in real conditions (large scale area, large data-bases, and the participation of planners) has enabled a better understanding of the difficulties related to a technique which is still in its infancy. It has also shown the unquestionable advantages of integrating landscape ecology and G.I.S. in the environmental management of an area and its resources.

The steppe-like character of the Causses region of Lozère means that this area contains habitats and species of international interest, although their long-term existence is threatened. Effective conservation of these areas requires an understanding of the dynamics of the agro- pastoral system. Study of work being undertaken in the context of the Life-Nature programme and of research on livestock systems allows the two approaches to be linked and the level of investigation or organisation to be specified. Finally, the different approaches are compared to offer guidelines for further reflection on the management of open landscapes and on methodological improvements which might be made.

The erosion protection division of Cemagref at Grenoble has perfected a method of ranking forests in terms of their role as a protection from avalanches. This is based on risk analysis carried out by the R.T.M. services of the O.N.F. This method takes into account the importance of natural hazards (avalanches, rock falls, etc.), the structure and dendrometric parameters of the forest cover and the vulnerability of the threatened sites. It also provides the criteria by which to determine those afforested areas from which avalanches are likely to start. A statistical model to calculate the stopping distance, developed by the Norvegians and adapted to French topographical conditions by Cemagref at Grenoble, is used to determine the probable stopping zones of these phenomena. All these criteria have enabled the completion of a Map of the Probable Location of Avalanches, based exclusively on observed events. The comparison of such information with forest data shows that forest maintain these phenomena at a level of potential activity. However, understanding of the control of this hazard by the forest requires a better comprehension of the interactions between forests and snow, and forest and the snow mantle. To do this, an experimental study plan of snow in forests has been made.

The project "Land Use Potential Analysis Achenkirch" was to enable the development of solutions for conflicts between various forms of land use in an alpine region of Austria. The spatial information system developed for this project is based on three modules. These are data, methods and G.I.S. -technology. G.I.S. technology is seen as a tool rather than a method. The modular approach was chosen in order to stress the difference between the development of new approaches in land use analysis and land use planning, which are of course inspired by new technological possibilities, and the functionalities, an individual software vendor is providing. Integrated data were used to perform assessments on the relative potentials for timber production, hunting, grazing, nature conservation and recreation in the project area. Potential conflict areas were identified by locating areas, where high land use potentials for competing forms of land use existed. The comparison with actual land use conflicts in the area provided information on the conflict-tendency of the land use intensities in the area. A combination of small to medium area priority land use allocations and large area intensity-adaptation was chosen as the guideline in the suggested new land use concept.