Assessing regional resilience of different land use types during snowstorms using mobile data

Abstract

Understanding the resilience of areas with different land-use types can enhance a city's ability to respond to and recover from disasters. Based on Docomo mobile GPS data and the 2018 Fukui Prefecture snow disaster, this study explores the resilience of areas with different land-use types from the grid level. First, the resilience triangle method was employed to assess the robustness, vulnerability, and survivability of different land-use types at the 500-m grid level. Second, Pearson correlation analysis and causal inference determined that snow depth is a causal factor leading to changes in grid resilience, thereby identifying the optimal time points for each grid to distinguish between disaster end and recovery start. Finally, a resilience map of Fukui Prefecture was created, integrating road data to identify cities with poor resilience and road sections prone to congestion during snow disasters. The results indicate that land-use types such as factories, agricultural land, building land, and mid-high-rise buildings have better resilience, especially mid-high-rise buildings, which exhibit strong resistance and recovery ability when facing disturbances, allowing them to return to normal states quickly. In urban planning, prioritizing such resilient building types could be beneficial. In contrast, sparse low-rise buildings, parks, facility land, and dense low-rise buildings exhibit poorer resilience and take longer to recover. Additionally, areas with poor traffic resilience in Fukui Prefecture are mainly concentrated in the northern part, particularly the vertical sections in Fukui City, Awara City, and Sakai City, as well as the vertical section from Sabae City to Echizen City and the horizontal section from Eiheiji-cho to Katsuyama City.