Scholarlink

This study develops a comprehensive landslide and flood susceptibility map for Himachal Pradesh, India, utilizing advanced Artificial Intelligence (AI) driven predictive analytics integrated with Geographic Information System (GIS) techniques. Covering the region between 30°22′40” to 33°12′20” N latitudes and 75°45′55” to 79°04′20” E longitudes, with altitudes ranging from 271 m to 6,751 m, the study addresses the region’s varied topography and climate. The landslide inventory map records 6,289 landslides, detailing their spatial distribution and movement patterns. Key AI techniques, including Decision Trees and Neural Networks, were employed to model the relationship between environmental factors such as slope, aspect, roughness, Hillshade, Land Use and Land Cover (LULC), and geological structures and landslide occurrences. The models were trained and validated using extensive historical data, leading to the creation of a landslide susceptibility map that classifies the region into Low, Medium, and High-risk zones. This integrative AI-GIS approach enhances the accuracy of disaster prediction and provides critical insights for land use planning and disaster management.

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

AI-Driven Predictive Analytics for Early Disaster Warning Systems

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • M. Madhusudhan Reddy,
  • R. Sandhya Rani,
  • K. T. Padma Priya,
  • U. Praveen Goud,
  • Manojkumar D. Ghongade

摘要

This study develops a comprehensive landslide and flood susceptibility map for Himachal Pradesh, India, utilizing advanced Artificial Intelligence (AI) driven predictive analytics integrated with Geographic Information System (GIS) techniques. Covering the region between 30°22′40” to 33°12′20” N latitudes and 75°45′55” to 79°04′20” E longitudes, with altitudes ranging from 271 m to 6,751 m, the study addresses the region’s varied topography and climate. The landslide inventory map records 6,289 landslides, detailing their spatial distribution and movement patterns. Key AI techniques, including Decision Trees and Neural Networks, were employed to model the relationship between environmental factors such as slope, aspect, roughness, Hillshade, Land Use and Land Cover (LULC), and geological structures and landslide occurrences. The models were trained and validated using extensive historical data, leading to the creation of a landslide susceptibility map that classifies the region into Low, Medium, and High-risk zones. This integrative AI-GIS approach enhances the accuracy of disaster prediction and provides critical insights for land use planning and disaster management.