Scholarlink

<p>Albeit there have been many works in the field of landslide hazard assessment covering the Ukhimath area in the Mandakini River basin of the Uttarakhand Himalaya, there is a serious need for managing satellite/geospatial data with novel AI/ML techniques for the assessment of landslide hazards. Hence, a comprehensive machine learning-based framework is adopted, employing XGBoost, LightGBM, and Random Forest on Sentinel-2B multispectral and SRTM DEM data for landslide mapping, which helped to achieve exact delineation of high-risk zones with accuracy provided by acceptable Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) scores. This innovative methodology throws light on how the best AI/ML techniques can be utilized for the development of a Landslide Early Warning System (LEWS), thereby helping in Disaster Risk Reduction (DRR).</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

Integrated Usage of Earth Observation and Machine Learning in Detection of Landslide Hotspot: A Case Example from Ukhimath, Uttarakhand

  • 期刊论文

  • 发表日期：

    2026年1月24日

查看全文
  • Sudikin Pramanik,
  • Shovan Lal Chattoraj,
  • Ashutosh Kumar Jha

摘要

Albeit there have been many works in the field of landslide hazard assessment covering the Ukhimath area in the Mandakini River basin of the Uttarakhand Himalaya, there is a serious need for managing satellite/geospatial data with novel AI/ML techniques for the assessment of landslide hazards. Hence, a comprehensive machine learning-based framework is adopted, employing XGBoost, LightGBM, and Random Forest on Sentinel-2B multispectral and SRTM DEM data for landslide mapping, which helped to achieve exact delineation of high-risk zones with accuracy provided by acceptable Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) scores. This innovative methodology throws light on how the best AI/ML techniques can be utilized for the development of a Landslide Early Warning System (LEWS), thereby helping in Disaster Risk Reduction (DRR).