Scholarlink

<p>On 16 June 2024, an extreme rainfall event triggered widespread landslides in Pingyuan County, northeastern Guangdong Province, China, resulting in 27 fatalities and an economic loss of 50.55 million RMB. Using multitemporal Sentinel-2 imagery, we mapped 359 landslides, predominantly shallow slope failures, covering approximately 2.87 km<sup>2</sup>. The maximum landslide density (&gt; 8 landslides) was observed within an ~ 100&#xa0;m radius. While smaller landslides (&lt; 5000 m<sup>2</sup>) accounted for more than half of the failures, a few large (&gt; 50,000 m<sup>2</sup>) landslides severely damaged local infrastructure. Preliminary field checks and remote sensing data indicate that steep slopes (commonly 20–36°), weathered granite lithology, and intense rainfall (up to 360&#xa0;mm/day) were the main triggers. Although local authorities keep historical records of slope failures, the substantial number of newly initiated landslides underscores the need for improved forecasting, early warning systems, and more robust landslide inventory compilation methods under frequent cloud cover. In this study, we employed a hybrid approach that integrates a random forest (RF) statistical model and a physically based model of transient rainfall infiltration and grid-based regional slope-stability (TRIGRS) to enhance future landslide prediction. The model accuracies reached 65.3% for the RF model, 63.5% for TRIGRS, and 76.3% when the integrated hybrid method was used. This report details the event’s background, the rapid mapping process, and initial insights into the factors contributing to the high landslide density, offering a foundation for subsequent hazard mitigation efforts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Severe rainfall-induced landslides in Pingyuan County, Guangdong, China, in June 2024

  • 期刊论文

  • 发表日期：

    2025年5月30日

查看全文
  • Wei Zhang,
  • Muhammad Zeeshan Ali,
  • Wenfeng Cui,
  • Chuangeng Sun,
  • Zhiwen Zheng,
  • Kejie Chen

摘要

On 16 June 2024, an extreme rainfall event triggered widespread landslides in Pingyuan County, northeastern Guangdong Province, China, resulting in 27 fatalities and an economic loss of 50.55 million RMB. Using multitemporal Sentinel-2 imagery, we mapped 359 landslides, predominantly shallow slope failures, covering approximately 2.87 km2. The maximum landslide density (> 8 landslides) was observed within an ~ 100 m radius. While smaller landslides (< 5000 m2) accounted for more than half of the failures, a few large (> 50,000 m2) landslides severely damaged local infrastructure. Preliminary field checks and remote sensing data indicate that steep slopes (commonly 20–36°), weathered granite lithology, and intense rainfall (up to 360 mm/day) were the main triggers. Although local authorities keep historical records of slope failures, the substantial number of newly initiated landslides underscores the need for improved forecasting, early warning systems, and more robust landslide inventory compilation methods under frequent cloud cover. In this study, we employed a hybrid approach that integrates a random forest (RF) statistical model and a physically based model of transient rainfall infiltration and grid-based regional slope-stability (TRIGRS) to enhance future landslide prediction. The model accuracies reached 65.3% for the RF model, 63.5% for TRIGRS, and 76.3% when the integrated hybrid method was used. This report details the event’s background, the rapid mapping process, and initial insights into the factors contributing to the high landslide density, offering a foundation for subsequent hazard mitigation efforts.