Ecosystem service evaluation based on local knowledge of residents using spatial text-mining

Abstract

This study aims to evaluate the ecosystem services of Upo wetland, one of the best-known Ramsar sites in Korea, reflecting the characteristics of the ecological assets and local knowledges in the area. Application of spatial text-mining begins with collecting local perceptions and knowledge of residents on the 17 ecological assets of Upo site and surrounding area. Our results identified five important ecosystem services: flood control during heavy rainfall, water purification by aquatic plants, cultural and natural heritages, agricultural products and water provision for crop cultivation. GIS created a map where these ecosystem services were linked to the locations of 17 ecological assets. This map showed which ecosystem service is associated with particular ecological assets and their characteristics from residents’ perspectives. Mapping local knowledge using the spatial text-mining is able to identify multi-functional bases which provide various ecosystem services in the same location simultaneously. Identification of multi-functional bases can provide information for local government to design an effective and comprehensive management plan considering physical-cultural geography of ecosystem services.