MAPEDGER: an innovative mapping tool for forest edges
|
|
©
Inra
|
Forest edges are key components in rural landscapes. They participate in important ecological processes (structural biodiversity, water retention, nesting, food resources for fauna, etc.) and, as a biological continuum, form part of green infrastructure at the junction of territories. In addition to their importance to the management and conservation of biodiversity, they also play social, aesthetic and recreational roles. To ensure the preservation and management of forest edges, land managers require tools to identify and map them so that they can berecorded as tangible elements in their development and management plans. Researchers in the “Dynamics and Ecology of Landscapes”, Joint Research Unit (DYNAFOR) in Toulouse are proposing the development of a novel mapping tool, MAPEDGER, which indeed enables the identification and characterisation of forest edge segments.
|
If the functional roles of forest edges are to be taken into account, it is necessary to characterise their diversity, including their history, physiognomy, orientation and dynamics. Such characterisation supposes the availability of spatial data and geomatics tools. To achieve this, the DYNAFOR team used the European GUIDOS software (Graphical User Interface for the Description of Image Objects and their Shapes) to characterise, quantify and map the diversity of wooded elements in a fragmented forest landscape (patch metrics), and also the concept of "edge segments" which represented edges with lines from which the metrics were calculated. In the context of this research, the more detailed characterisation of edge segments corresponding to spatial units of importance to ecologists was made possible using MAPEDGER, a innovative processing sequence that opens the way to the construction of an integrated geomatics tool to measure landscape metrics.
The GIS, ArcGis 9.3 software program (from ESRI company) enriched with free scripts, and GUIDOS (from the Joint Research Centre), were the tools employed by the team to record vector data to map edges that were often rectilinear (in temperate regions, the morphology of edges is often limited by the linearity of farming operations).
This mapping process takes the form of three, coherent chronological phases. The MAPEDGER method starts by using an extensive land use map generated in raster mode, from which a binary forest/non-forest image is obtained (only the classes of wooded elements are retained) which is then imported under GUIDOS. Its processing by this software is then able to highlight seven classes, which include forest cores, edges, gaps or branches. During the second phase, these data, imported under ArcGis, are converted into vectors: edge segments are thus created and identified (simplification of forest contours and extraction of edge segments). The final phase consists in characterising these segments using variables defined by ecologists as a function of their needs. This takes account of the specific attributes of each edge (localisation, length, orientation, part of a wood) and of data extracted by crossing with other spatial data (Digital Elevation Model, land cover map). Landscape metrics are calculated from three edge variables: orientation (Orientation Cardinale, OC), topography (Orientation par rapport à la Pente, OP) and adjacent landcover (Occupation du Sol voisine, OS).
In order to ensure satisfactory data transmission, a graphic semiology adapted to the representation of forest edges has been defined. The "MAPEDGERSymbol" semiological grid takes account of representation constraints: it manages both the plurality of variables concerning a single segment, whether they are qualitative (OP and OS) or quantitative (OC), and can also differentiate segments using visual signs (simple geometric shape, colour, transect representation).
The GIS project thus developed enables both navigation within the database obtained and queries regarding the characteristics of edge segments in terms of their spatial dimensions and attributes. Synthetic mapping, accompanied by a table of edge segment characteristics, thus reflects these results.
This original work, which formed part of a thesis project, received financial support from the Science for Action and Sustainable Development Division at INRA and the French Ministry for Research and Higher Education. By rendering the concept of forest edges operational, there is now an opportunity to use new metrics to analyse landscapes, while at the same time revealing the interface functions played by edges between different environments. The innovative MAPEDGER mapping technique, whose usefulness has been proven by a large-scale landscape diagnosis, is a tool that could be used by the managers of rural areas in order to better control the future and uses of their territory.
Scientific leaders:
Marc DECONCHAT
Sylvie LADET
Audrey ALIGNIER
Inra - UMR1201 DYNAFOR Dynamiques et écologie des paysages agriforestiers
Chemin de Borde-Rouge
31326 CASTANET-TOLOSAN CEDEX
Information contact:
|
See also
