Socioeconomic Drivers of Post-Rainstorm Community Resilience: A Comparative Analysis of Two Cases in North China

Abstract

Short-duration and high-intensity extreme rainfall events increasingly disrupt urban systems. Under escalating climate risks, understanding post-disaster recovery dynamics has become critical for resilience assessment and recovery planning. However, research on post-rainstorm recovery remains constrained by limited data availability, insufficient temporal resolution, and inadequate identification of spatiotemporal heterogeneity in recovery trajectories. To address these limitations, this study uses daily NASA VIIRS nighttime light (NTL) time-series data as a proxy for observable functional recovery in two flood-affected areas in North China following the July–August 2023 extreme rainfall event. A pixel-level Recovery Rate Index is constructed to quantify rebound dynamics, and time-series clustering is applied to identify distinct recovery trajectories. Explanatory modeling is further conducted to examine the socioeconomic and environmental drivers associated with different recovery modes. Results show that recovery regimes differ substantially across terrain contexts. The comparison further suggests that mountainous and plain urban environments are characterized by different dominant recovery mechanisms, highlighting the need for geographically differentiated post-disaster recovery assessment and resilience planning.