Identification and characterization of forest edge segments for mapping edge diversity in rural landscapes

Abstract : Forest edges are key components of rural landscapes and they influence major ecological processes. There is a variability of forest edges, according to their physiognomy, orientation, history and topography, but few methods are available to identify automatically these different types of edges and to map them at a large scale. We propose to identify edge segments based on morphological subdivision of forest boundaries. These segments can be mapped and characterized by other spatial data. A procedure based on Arcgis tools, applied on the output from GUIDOS tools, is used to identify edge segments. We provide examples of edge segments descriptors obtained from a digital elevation model and from a landcover map. Results showed the feasibility of the automatic mapping of edge variability over a large scale. It opens new perspectives for the analysis of landscape dynamics and their effects on biodiversity.