Mapping urban mobility during extreme weather events using mobile spatial statistics data

Abstract

Human mobility is increasingly vulnerable to disruptions due to the rising frequency of adverse weather events. However, research on the impacts of extreme weather on mobility has primarily focused on individual movement characteristics, with limited attention to changes in network structures. This study utilizes mobile spatial statistics data from Sapporo, observed before, during, and after heavy snow events in 2022, to examine their effects on urban mobility. The network clustering results reveal that mobility tended to cluster into small communities, particularly on weekends and holidays, with the number of main communities ranging from 10 to 15, compared to only 2 to 6 on weekdays. The number of communities was also higher on heavy snow days compared to normal days, and the largest community experienced a nearly 50% reduction in size during these events. Despite these changes, the city center remained integral to the network, with most communities exhibiting high centrality and connections to commercial areas, transportation hubs, and mixed land-use zones. Additionally, the first heavy snowfall event had a more pronounced impact on mobility than the second. The community patterns also suggest that Sapporo follows a monocentric urban form. These findings could provide valuable insights for transportation planning and management while supporting policymakers in developing effective disaster prevention and mitigation strategies.