Scholarlink

<p>Nepal experiences a significant number of natural hazards, with landslides and floods occurring as the predominant events. The Far Western Region exhibits a significant susceptibility to such events. This research assesses the susceptibility to landslides within the Thuligad watershed of Nepal, by using the Frequency Ratio (FR) and Information Value (IV) methods. A detailed landslide inventory was prepared by digitizing mapped landslide events from Google Earth, Sentinel-2 and Sentinel-3 imagery. In total, 1156 landslide locations were identified in the study area. The total landslides were divided into 70% and 30% datasets to develop and validate the models. Thirteen causative factors were analyzed, covering hydrological, geological, topographic, and morphological variables. Landslide susceptibility map was developed in ArcGIS, and their success and prediction rates were accessed. Susceptibility was classified into five classes: very high, high, moderate, low, and very low. Results from the frequency ratio method indicated that 10, 24, 31, 25, and 10% of the area fell into these categories, whereas the IV model showed 5, 16, 22, 31, and 26%, respectively. The ROC analysis revealed that the frequency ratio method achieved success and prediction rates of 66.7% and 60.7%, while the information value method achieved higher rates of 74.1% and 69.8%. Overall, the IV method demonstrated better predictive performance and reliability compared to the FR method. These findings provide valuable input for future studies and can guide policymakers and local authorities in designing effective landslide prevention and disaster risk reduction 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

Application of bivariate frequency ratio and information value models for landslide susceptibility in Thuligad watershed of far-western Nepal

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月1日

查看全文
  • Tekendra Bahadur Saud,
  • Prakash Bahadur Ayer,
  • Mahesh Prasad Awasthi,
  • Bharat Prasad Bhandari,
  • Narayan Raj Joshi

摘要

Nepal experiences a significant number of natural hazards, with landslides and floods occurring as the predominant events. The Far Western Region exhibits a significant susceptibility to such events. This research assesses the susceptibility to landslides within the Thuligad watershed of Nepal, by using the Frequency Ratio (FR) and Information Value (IV) methods. A detailed landslide inventory was prepared by digitizing mapped landslide events from Google Earth, Sentinel-2 and Sentinel-3 imagery. In total, 1156 landslide locations were identified in the study area. The total landslides were divided into 70% and 30% datasets to develop and validate the models. Thirteen causative factors were analyzed, covering hydrological, geological, topographic, and morphological variables. Landslide susceptibility map was developed in ArcGIS, and their success and prediction rates were accessed. Susceptibility was classified into five classes: very high, high, moderate, low, and very low. Results from the frequency ratio method indicated that 10, 24, 31, 25, and 10% of the area fell into these categories, whereas the IV model showed 5, 16, 22, 31, and 26%, respectively. The ROC analysis revealed that the frequency ratio method achieved success and prediction rates of 66.7% and 60.7%, while the information value method achieved higher rates of 74.1% and 69.8%. Overall, the IV method demonstrated better predictive performance and reliability compared to the FR method. These findings provide valuable input for future studies and can guide policymakers and local authorities in designing effective landslide prevention and disaster risk reduction strategies.