Author(s): Zhengzheng Zhou; Nuo Lei; Shuguang Liu
Linked Author(s): Shuguang Liu, Zhengzheng Zhou
Keywords: Short-Duration Extreme Rainfall Spatiotemporal Variability Earth Observation (EO) Regional Storm Design
Abstract: Understanding the spatiotemporal variability of short-duration extreme rainfall events is critical for assessing urban flood risks and designing effective flood management strategies. This study employs Earth Observation (EO) data integrated with a high-resolution sub-hourly rain gauge network to investigate the spatiotemporal patterns of short-duration rainstorms in Shanghai, China. A dataset of 207 independent rainstorm events during 2014-2018 is analyzed using hierarchical clustering to characterize rainfall duration, intensity, and spatial coverage across districts with varying degrees of urbanization. The results reveal pronounced spatial heterogeneity, with heavier rainfall magnitudes concentrated in the northern districts and more frequent, intense short-duration rainstorms observed in the urban core. Rainstorms are classified into three types based on their spatiotemporal processes: Type delayed-slow, Type advanced-fast, and Type advanced-slow. Delayed-peak storms with higher intensity gradients are more prevalent in the western districts, while advanced-peak storms dominate along the coastal areas. Additionally, advanced-peak storms with inverse trends of intensity and coverage cluster during June-August and late afternoon hours. Earth Observation data further highlight the role of atmospheric drivers in shaping storm patterns and enable the identification of localized trends, such as 65.6% of delayed-peak storms being confined to a single district, whereas more than half of advanced-peak storms occur independently without overlap from other storm types. These findings emphasize the importance of EO-based observation and analysis in understanding rainfall dynamics and provide a scientific basis for regional storm design and urban flood prevention strategies. The integration of EO technologies in this study demonstrates their potential for improving numerical modeling, urban flood resilience, and policy development in response to evolving hydrometeorological challenges.
Year: 2025