Seventeenth International Conference on Environmental, Cultural, Economic & Social Sustainability

Abstract

Cities are becoming key actors in combating climate change impacts. Cities are particularly at threat of devastating climate disasters as growing population and infrastructure are intertwined in urban areas, most of which are located in coastal regions. As a pivotal place to embark on adaptation plans, cities take various adaptation measures from building dykes and dams to expanding green space, managing water resources, and establishing disaster monitoring systems. Nevertheless, little research has been done on urban climate change adaptation on a global scale because of the lack or the inconsistency of urban data. To overcome this limitation, the study employs text mining techniques and examines current practices of global cities, which were yet to be used for data analysis but in fact provide valuable evidence of municipal practices in climate actions. The data for the study are the project descriptions from city governments on the C40 website. The C40 network has selected 100 successful cases every year and published the report ‘Cities100’ since 2015. From this textual dataset (from 2015 to 2019), 47 cases were chosen under the category of adaptation and used for topic modeling. The key method of this study is Latent Dirichlet Allocation (LDA), a machine learning algorithm for finding major topics within the data. Consequently, the study reveals in which ways sustainable cities implement adaptation plans, and by introducing textual data as a new data source, contributes to a global understanding of urban adaptation.