Unveiling Spatio-Temporal Resilience Patterns for Planning: Human Activity Responses to Extreme Rainfall Events

Abstract

Climate resilience, as a strategic approach to mitigate risks and address unpredictable disruptions, has gained widespread international attention.

The measurement and mapping of resilience are pivotal in augmenting spatial planning and regulatory frameworks(Rodríguez-Izquierdo et al., 2022). These efforts aim to characterize the impact of disturbances, facilitating a comprehensive understanding of improvement targets and operational frameworks for enhanced effectiveness. However, current methodologies primarily focus on static indicators, in that the times and locations of people are considered fixed, providing limited insights into the dynamic evolution of urban areas during disasters. In reality, urban mobility under disaster is nonlinear, chaotic, and spans across temporal and spatial scales(Li & Yan, 2024; Yabe et al., 2023).

The utilization of big data presents significant potential for analysing and predicting urban mobility patterns under extreme events, particularly leveraging human mobility data(Haraguchi et al., 2022; Rajput et al., 2023). Our study seeks to establish an empirical research method for characterizing urban resilience through the analysis of mobility data. We aim to assess spatial heterogeneity to identify vulnerable regions within the city which that may experience significant disruptions in traffic flow and mobility pattern, enabling the formulation of targeted planning strategies.

To demonstrate the practical application of this approach, we conducted a case study in Zhengzhou, a city in China that encountered a once-in-a-millennium rainfall event in 2021. Utilizing a comprehensive dataset of smartphone signalling records from July 11 to 31, 2021, we examined how human activity patterns changed before, during, and after this extreme weather event. Then, the changes in mobility served as a key indicator and resilience index with spatio-temporal attributes was calculated at the grid-level for describe the disparity and resilience mechanism. Furthermore, our study involved an examination of the gradient change in resilience characteristics extending from the city centre towards various directions. This analysis aims to elucidate the influence of the urban polycentric development strategy on resilience.

Our analysis sheds light on the spatio-temporal characteristics of resilience patterns, providing valuable insights for future urban planning in regions prone to such disturbances. We observe that resilience levels do not uniformly decrease from the urban centre outward. Communities near urban sub-centres display higher resilience levels. Despite experiencing a decline in activity intensity and flow after the storm, these neighbourhoods show a lower magnitude of change in their temporal patterns, indicative of higher resilience. Conversely, areas situated in urban centres with a single residential function exhibit longer recovery periods following brief emergency population stagnation during storm events, suggesting lower levels of resilience.

With that said, the polycentric development strategy indeed plays a crucial role in bolstering the urban resilience of areas adjacent to the subcentres.

This approach holds the potential for generalization to other cities and disasters, thereby contributing to the shared principle of enhancing urban resilience through spatial strategies.