Supplementary data of ``U-RNN high-resolution spatiotemporal nowcasting of urban flooding''

The dataset is a spatial and temporal dynamic simulation of rainfall-induced urban flooding. A total of 56 rainfall events, including 20 measured rainfall events and 36 design rainfall events, are used to assess urban flooding induced by various rainfall conditions. The measured rainfall events are the most severe rainfall events recorded in Shenzhen City, China, during the period from 2008 to 2018, with a duration of 6 hours, a rainfall amount of 48-211 mm, and a return period of 0.1-20 years. To analyze the potential extreme rainfall risk due to climate change, we design some extreme rainfall events based on Chicago storm profiles and Shenzhen rainfall intensity-duration-frequency relationships, with a duration of 3 h, a return period of 100-500 years, a time-peak ratio range of 0.1-0.9, and a data recording frequency of 1 min. The flooding spatio-temporal dynamics data are generated by MIKE Plus version 2023 simulation with a recording frequency of 1 min. We also provided geo-information data including DSM, drainage inlet area and imperviousness.本数据集为降雨诱发城市内涝的时空动态模拟数据集。共包含56场降雨事件，其中实测降雨事件20场、设计降雨事件36场，用于评估不同降雨条件下的城市内涝情形。实测降雨事件为中国深圳市2008年至2018年间记录的极端降雨事件，单场降雨历时6小时，降雨量介于48毫米至211毫米之间，重现期为0.1至20年。为分析气候变化背景下潜在的极端降雨风险，研究团队基于芝加哥雨型（Chicago storm profiles）与深圳市降雨强度-历时-频率关系设计了多组极端降雨事件，此类降雨事件单场历时3小时，重现期为100至500年，峰现时间占比范围为0.1至0.9，数据记录频率为1分钟。内涝时空动态数据通过MIKE Plus version 2023模拟生成，记录频率为1分钟。本数据集还附带了地理信息数据，包括数字表面模型（DSM, Digital Surface Model）、排水口汇水区面积以及不透水率。