Researchers reconstruct path and intensity of China's Guangzhou 'April 27' tornado using multi-source data

Existing disaster investigation systems often focus on meteorological cause analysis, while systematic research from the perspective of structural wind engineering on building damage mechanisms remains relatively limited.

In a study published in the journal Advances in Wind Engineering, researchers from South China University of Technology and collaborating institutions described the EF3-level tornado that occurred in Guangzhou, China on April 27, 2024. They achieved this by integrating methods such as UAV remote sensing, detailed ground-based post-disaster surveys, and mesoscale numerical simulation.

"This research combines meteorological simulation with structural damage analysis, systematically examining the tornado disaster from an engineering perspective," shares corresponding author Professor Yi Yang. "The findings can provide empirical data reference for refining the descriptions of engineering structural damage in China's national standard, the 'Tornado Intensity Scale'."

In the study, the research team established six main survey lines and 12 branch lines within an affected area of approximately 15 square kilometers in Guangzhou's Baiyun District, collecting and analyzing disaster data from 471 characteristic sites. By examining nearly 500 sets of UAV images and about 1,200 sets of ground-based photos and videos, the team delineated the tornado's path: moving from west to east-southeast, with a total length of approximately 8.2 kilometers, an average width of 0.7 kilometers, and a maximum width reaching 1 kilometer. Based on the damage levels of indicators such as buildings, trees, and utility poles observed on-site, and referring to the Chinese national standard "Tornado Intensity Scale," the team rated the tornado as EF3 (with wind speeds approximately 61-73 m/s).

In the more severely affected areas (Main Survey Lines 2 and 3), numerous light-steel industrial buildings suffered severe damage to their metal roofing systems. Analysis of 431 building sites revealed that 18.33% experienced severe damage, including collapsed rigid frames and exterior walls. Common metal roof failure modes included localized tearing at self-drilling screw connections and connector pull-out from purlins.

Based on these findings, the team proposed wind-resistant design improvements such as optimizing ridge/eave geometry, increasing connector density, and using standing seam roof systems.

The study also employed the Weather Research and Forecasting (WRF) model to simulate the event. The KF-Eta scheme effectively reproduced the environmental conditions, successfully simulating parent storm characteristics like the "hook echo" signature, storm trajectory, and helicity patterns consistent with observational data. Additionally, by back-calculating the load capacity of two collapsed concrete utility poles, the research estimated peak gust wind speeds during the tornado at 74.59 m/s and 79.77 m/s.

Contact author: 

Yi Yang, State Key Laboratory of Subtropical Building and Urban Science, South China University of Technology, Guangzhou, Guangdong, China, ctyangyi@scut.edu.cn

Funder: 

The authors would like to acknowledge the financial supports from the National Natural Science Foundation of China under Grant No. 52178480, the State Key Laboratory of Subtropical Building and Urban Science Project under Grant No. 2024ZB10 and the Guangdong Basic and Applied Basic Research Foundation under Grant No. 2024A1515011828.

Conflict of interest:

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Jian Wu is currently employed by China Nuclear Power Design Co., Ltd. The authors declare no other competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

See the article: 

Songyan Hu, Zexian Li, Jian Wu, Yi Yang, Muguang Liu, Zhuangning Xie, Post-disaster investigation and WRF model reconstruction of April 27th tornado in Guangzhou based on muti-source data fusion, Advances in Wind Engineering, Volume 3, Issue 1, 2026, 100098, ISSN 2950-6018, https://doi.org/10.1016/j.awe.2025.100098.