Scholarlink

<p>The study investigates the influence of uncertainties and spatial variability in slope stability and landslide runout analysis, with particular emphasis on the Western Ghats of India, a highly landslide-prone region. Traditional models often overlook the inherent variability of soil properties, resulting in inaccurate hazard assessments. The analysis is structured around three cases study. The first two cases considered are a regional-scale landslide susceptibility assessment and a site-specific slope stability analyses by quantifying uncertainties and incorporating spatial variability using probabilistic approaches. For regional-scale analysis, landslide susceptibility is evaluated using the transient rainfall infiltration and grid-based regional slope stability (TRIGRS) model, incorporating the first-order second moment (FOSM) method to account for uncertainties in geotechnical properties like angle of internal friction angle and cohesion. The results demonstrate that probabilistic models significantly improve the accuracy of landslide predictions on regional scale compared to conventional statistical methods. In the site-specific analysis, the random finite element method (RFEM) is employed to account for the spatial variability of shear strength parameters following Gaussian and exponential autocorrelation models. The findings reveal that neglecting spatial variability, as seen in the single random variable (SRV) approach, leads to unconservative slope stability predictions, with SRV overestimating mean factor of safety (FoS) and underestimating the slope failure probability. The effect of parameter uncertainties in modeling of landslide runout is also explored as a third case of interest. For runout modeling, the Voellmy model is employed to simulate the debris flow from a documented landslide event in Peringalam, Kerala, using Monte Carlo Simulations (MCS) to quantify uncertainties in turbulent coefficient and basal friction angle. The results show that incorporating uncertainties leads to more reliable predictions of flow velocity, depth, and deposition extent. Overall, the study underscores the significance of probabilistic methods and spatial variability in improving the accuracy of landslide hazard assessments and mitigation strategies.</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

Uncertainties and Spatial Variability in Slope Stability and Landslide Runout Analysis

  • 期刊论文

  • 发表日期：

    2025年7月9日

查看全文
  • Abhijith Ajith,
  • C. Kavinkumar,
  • Rakesh J. Pillai

摘要

The study investigates the influence of uncertainties and spatial variability in slope stability and landslide runout analysis, with particular emphasis on the Western Ghats of India, a highly landslide-prone region. Traditional models often overlook the inherent variability of soil properties, resulting in inaccurate hazard assessments. The analysis is structured around three cases study. The first two cases considered are a regional-scale landslide susceptibility assessment and a site-specific slope stability analyses by quantifying uncertainties and incorporating spatial variability using probabilistic approaches. For regional-scale analysis, landslide susceptibility is evaluated using the transient rainfall infiltration and grid-based regional slope stability (TRIGRS) model, incorporating the first-order second moment (FOSM) method to account for uncertainties in geotechnical properties like angle of internal friction angle and cohesion. The results demonstrate that probabilistic models significantly improve the accuracy of landslide predictions on regional scale compared to conventional statistical methods. In the site-specific analysis, the random finite element method (RFEM) is employed to account for the spatial variability of shear strength parameters following Gaussian and exponential autocorrelation models. The findings reveal that neglecting spatial variability, as seen in the single random variable (SRV) approach, leads to unconservative slope stability predictions, with SRV overestimating mean factor of safety (FoS) and underestimating the slope failure probability. The effect of parameter uncertainties in modeling of landslide runout is also explored as a third case of interest. For runout modeling, the Voellmy model is employed to simulate the debris flow from a documented landslide event in Peringalam, Kerala, using Monte Carlo Simulations (MCS) to quantify uncertainties in turbulent coefficient and basal friction angle. The results show that incorporating uncertainties leads to more reliable predictions of flow velocity, depth, and deposition extent. Overall, the study underscores the significance of probabilistic methods and spatial variability in improving the accuracy of landslide hazard assessments and mitigation strategies.