Get reliable science

Ecological Site Classification (ESC) is being used to define site quality in order to help British foresters satisfy multiple management objectives while encouraging sustainable woodland design and management. ESC takes climatic and edaphic data to assess site quality. An obvious use of the classification is to predict the suitability of ecologically adapted commercial tree species, amenity tree species and new native woodland communities on any particular site, however ESC can inform other forest management decisions. The analysis can be done manually or with the aid of a computer based decision support system. The Ecosystem Management Decision Support (EMDS) system integrates the NetWeaver knowledge base system with ArcView GIS to provide decision support technology for ecological landscape analysis applications. Whereas the previous implementation of ESC provides decision support for individual landscape units, the EMDS implementation extends ESC to enable assessment of hundreds or thousands of units in large landscapes in a single analysis. The NetWeaver technology that underlies EMDS enables more flexible problem-solving knowledge representations that permit an evaluation of the degree of truth rather than the binary true or false of more traditional rule-based approaches. Additionally, NetWeaver allows an assessment of the effect of missing information, and the incorporation of knowledge bases that can evaluate a range of topics, such as social, economic, aesthetic, and legal issues, which might be related to the ESC biophysical model. This paper describes a prototype ESC-EMDS system which calculates the suitability of tree species or native woodland types at the forest landscape scale. We see this development as the core of a spatial decision support tool which will ultimately link spatial decision support system modules to evaluate the ecological impact of forest design plans. The work brings together two projects teams from Forest Research in Britain and the US Forest Service in Oregon.