Landscape evaluation of heterogeneous areas using fuzzy sets

Abstract

Landscape evaluation is an interesting field for fuzzy approaches, because it happens on the transition line between natural and social systems. Both are very complex. Therefore, transformation of scientific results to politically significant statements on environmental problems demands intelligent support. Particularly landscape planners need methods to gather natural facts of an area and assess them in consideration of its meaning to society as a whole. Since each land unit is heterogeneous, a special methodology is necessary. Such an evaluation technique was developed within a Geographical Information System (ARC/INFO). The methodology combines several known methods with fuzzy approaches to catch the intrinsic fuzziness of ecological systems as well as the heterogeneity of landscape. Additionally, a way will be discussed to vary the fuzzy inference in order to consider spatial relations of various landscape elements. Fuzzy logic is used to process the data uncertainty, to simulate the vagueness of knowledge about ecological functionality, and to model the spatial structure of landscape. Fuzzy sets describe the attributes of thematically defined land units and their assessment results. In this way, the available information will be preserved in their full diversity. The fuzzy operations are executed by AML-programs (ARC/INFO Macro Language). With such a tight coupling, it is possible to use the geographical functions (neighbourhoods, distances, etc.) of GIS within the fuzzy system directly.

Keywords

• landscape evaluation
• fuzzy sets
• erosion