Scholarlink

The Himalayan region, known for its seismic activity, has witnessed numerous destructive earthquakes and remains at risk of experiencing more in the future. Additionally, the mountainous terrain of the Himalayas is susceptible to cascading hazards such as earthquake-induced landslides (EQILs), exacerbating casualties and causing damage to critical infrastructure, transportation networks, and water supply facilities. Recognizing the vulnerable areas for EQIL becomes crucial to mitigate potential damages. To assess regional-scale EQIL susceptibility, many researchers utilize the well-established Newmark displacement-based slope displacement (SD) prediction models. In this study, five widely recognized SD models [13] (JB07); [23] (SR08); [11] (HL11); [7] (CH14); [21] (NG22) are employed to evaluate their efficacy in predicting landslides triggered by the 1999 Chamoli earthquake (Mw = 6.5). Shear strength parameters, specifically cohesion (c) and friction angle (ϕ) are determined based on geological data. The critical acceleration \(\left( {k_{y} } \right)\) is computed for each pixel in the selected region, combining the shear strength parameters with slope angles derived from the digital elevation map (DEM) of the Chamoli region. The intensity measures (IMs) are estimated from available ground motion models. The slope displacements are calculated for each pixel by integrating IMs and \(k_{y}\) values with the selected SD models. Subsequently, the probability of exceeding 5 cm (P(E)) is obtained. Further, the Chamoli region is divided into 4 different susceptibility classes, i.e., low, moderate, high, and very high based on P(E). The performance of slope displacement models is assessed and validated using the Area Under Curve (AUC). The SR08 model performed better with an AUC of 60.04%, followed by NG22 (60.03%), HL11 (49.46%), and JB07 (46.99%). The study’s findings emphasize the significance of choosing an appropriate SD prediction model for conducting region-specific landslide susceptibility assessments in the Indian Himalayan region.

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

Assessing the Performance of Empirical Slope Displacement Models in Spatial Prediction of Earthquake-Induced Landslides for the Indian Himalayan Region

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • N. Jayasri,
  • Sangeeta,
  • Maheshreddy Gade

摘要

The Himalayan region, known for its seismic activity, has witnessed numerous destructive earthquakes and remains at risk of experiencing more in the future. Additionally, the mountainous terrain of the Himalayas is susceptible to cascading hazards such as earthquake-induced landslides (EQILs), exacerbating casualties and causing damage to critical infrastructure, transportation networks, and water supply facilities. Recognizing the vulnerable areas for EQIL becomes crucial to mitigate potential damages. To assess regional-scale EQIL susceptibility, many researchers utilize the well-established Newmark displacement-based slope displacement (SD) prediction models. In this study, five widely recognized SD models [13] (JB07); [23] (SR08); [11] (HL11); [7] (CH14); [21] (NG22) are employed to evaluate their efficacy in predicting landslides triggered by the 1999 Chamoli earthquake (Mw = 6.5). Shear strength parameters, specifically cohesion (c) and friction angle (ϕ) are determined based on geological data. The critical acceleration \(\left( {k_{y} } \right)\) is computed for each pixel in the selected region, combining the shear strength parameters with slope angles derived from the digital elevation map (DEM) of the Chamoli region. The intensity measures (IMs) are estimated from available ground motion models. The slope displacements are calculated for each pixel by integrating IMs and \(k_{y}\) values with the selected SD models. Subsequently, the probability of exceeding 5 cm (P(E)) is obtained. Further, the Chamoli region is divided into 4 different susceptibility classes, i.e., low, moderate, high, and very high based on P(E). The performance of slope displacement models is assessed and validated using the Area Under Curve (AUC). The SR08 model performed better with an AUC of 60.04%, followed by NG22 (60.03%), HL11 (49.46%), and JB07 (46.99%). The study’s findings emphasize the significance of choosing an appropriate SD prediction model for conducting region-specific landslide susceptibility assessments in the Indian Himalayan region.