Public emotions and the built environment in hazards: A case study of the Shenzhen catastrophic landslide

Abstract

Capturing spatiotemporal changes in public emotions is an important component of urban human-oriented resilience. Built environment has the potential to conserve the negative effects of hazards. By relying on massive Weibo tweets, this study captures the spatiotemporal patterns of public negative emotions (sadness, fear, and anger) at both the individual and public levels after a landslide. The influence of the built environment on the change rate of public emotion was explored through geographically weighted random forest (GWRF). This study revealed that individuals with a pre-hazard intensity of negative emotions <25 % are more likely to experience heightened negative emotions. The impact of the landslide on the average intensity of public negative emotions is strongest within 3 days after and within a 10 km radius of the landslide. Moreover, townships to the south of the city, farther from the hazard, experienced more pronounced fluctuations in negative emotions, potentially leading to additional social management challenges. Furthermore, in 57.14 % of the townships, the most significant factors were either the built-up area or the industrial land ratio. This study expands the current understanding of public emotions and the influence mechanism of the built environment.