Scholarlink

<p>The Newmark displacement (ND) method is a reliable tool for assessing earthquake-induced slope deformation, yet a universally accepted regional-scale approach is still lacking. Existing ND equations often rely on regression from diverse strong-motion datasets or landslide statistics that were initially unavailable. To address this, we propose a site-adaptable ND approach (SAND) for various tectonic mountainous environments and scales. This knowledge- and data-driven method models cumulative slope displacement assuming normal pre-earthquake conditions, considering a quadratic relationship with peak ground acceleration (PGA) and a non-linear relationship with critical acceleration (A<sub>c</sub>). We gradually incorporate region- and site-specific conditions, including fault type, focal mechanism, hanging/footwall effects, topographic amplifications, terrain roughness, and wetness coefficient. Case studies include landslides triggered by the Haiti earthquakes in 2010 and 2021, as well as those from Taiwan (1999), Lushan (2013 and 2022), and Luding (2022). SAND outperforms previous regression-based models in predicting landslide locations. The SAND threshold varies between 0 and 10&#xa0;cm, in line with previous studies. The SAND approach quickly predicts shallow ETLs to support rescue efforts, using data on earthquake magnitude, epicenter, and focal mechanism (ideally the fault activated segment).</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

Effective prediction of earthquake-induced slope displacements, considering region-specific seismotectonic and climatic conditions

  • 期刊论文

  • 发表日期：

    2025年3月18日

查看全文
  • Danny Love Wamba Djukem,
  • Xuanmei Fan,
  • Hans-Balder Havenith

摘要

The Newmark displacement (ND) method is a reliable tool for assessing earthquake-induced slope deformation, yet a universally accepted regional-scale approach is still lacking. Existing ND equations often rely on regression from diverse strong-motion datasets or landslide statistics that were initially unavailable. To address this, we propose a site-adaptable ND approach (SAND) for various tectonic mountainous environments and scales. This knowledge- and data-driven method models cumulative slope displacement assuming normal pre-earthquake conditions, considering a quadratic relationship with peak ground acceleration (PGA) and a non-linear relationship with critical acceleration (Ac). We gradually incorporate region- and site-specific conditions, including fault type, focal mechanism, hanging/footwall effects, topographic amplifications, terrain roughness, and wetness coefficient. Case studies include landslides triggered by the Haiti earthquakes in 2010 and 2021, as well as those from Taiwan (1999), Lushan (2013 and 2022), and Luding (2022). SAND outperforms previous regression-based models in predicting landslide locations. The SAND threshold varies between 0 and 10 cm, in line with previous studies. The SAND approach quickly predicts shallow ETLs to support rescue efforts, using data on earthquake magnitude, epicenter, and focal mechanism (ideally the fault activated segment).