Insights into citizens’ experiences of cultural ecosystem services in urban green spaces based on social media analytics

Abstract

Urban green spaces (UGSs) facilitate the interaction of residents with blue and green infrastructure. Various cultural ecosystem services (CESs) generated by UGSs are reflected in social media data, and continuous efforts are needed to consistently characterize citizens’ perceptions of CESs by mining increasingly available social media data. For 50 UGS sites in Shanghai, we established a perception lexicon to cluster CESs, and we analyzed the impacts of landscape elements on citizens’ sentiments via text analytics. Nine types of landscape elements and five types of CESs were identified. Among the five CES types, recreational activities and social interaction were perceived the most frequently, while aesthetic appreciation was perceived the least frequently. Furthermore, the UGS sites were classified into social interaction-oriented, outdoor workout-oriented, history and culture-oriented, or multi-functional spaces. Multivariate regression analysis on the sentiments expressed by reviews revealed the significant impacts of natural landscape elements (plants, animals, and water bodies) and supporting facilities on the perceived sentiments. This study demonstrates the utility of social media data for identifying the CESs experienced by citizens, and the information generated through this approach is potentially useful for urban planners, landscape architects, and policymakers.