Landslide Risk Prediction Program Using Precipitation Data

Overview

Increased localized torrential rains and heavy downpours due to climate change are heightening the risk of landslides. Existing landslide prediction systems assess risk using precipitation and topography data, but struggle to incorporate real-time data, hindering rapid response. This system aims to predict risk more precisely by monitoring precipitation changes and soil conditions in real time, and prevent damage through early warnings.

Project Description

This program detects precipitation and ground changes and performs functions to assess the risk of landslide occurrence.

• AI-Based Precipitation and Ground Change Analysis
  • Utilizing an AI model trained on Korea Meteorological Administration data, topographical characteristics, and historical landslide records, it analyzes precipitation changes and ground deformation patterns in specific regions. This enables prediction of landslide potential and supports proactive response when necessary.
• IoT-Based Real-Time Soil and Moisture Condition Detection
  • It monitors environmental changes in landslide-prone areas using soil moisture sensors, ground deformation detectors, and barometers. When moisture content exceeds specific thresholds or ground instability occurs, it immediately sends alerts to facilitate early response.
• Risk visualization and automated alert system operation
  • Utilizing GIS (Geographic Information System)-based map data, it visualizes landslide-prone areas, predicts risk levels, and provides alerts to users. In collaboration with government agencies and local authorities, it sends immediate notifications to residents in hazardous areas and guides them through evacuation procedures.

Key technological features

• AI-Based Precipitation and Ground Change Prediction Model: Evaluating Landslide Risk by Analyzing Precipitation Data
• Soil and moisture condition monitoring using IoT sensors: Real-time detection of landslide risk factors using soil moisture sensors, barometers, etc.
• Risk visualization and alert system implementation: Visualize risk areas on a map-based platform and provide immediate alerts when necessary

Expected outcomes

This system enables more precise prediction of landslide risks caused by increased rainfall. The early warning system minimizes loss of life and property damage. Furthermore, continuously accumulating data on landslide-prone areas strengthens long-term prevention and response systems.

Launch year: 2024

Attribution

Case study courtesy of the National Information Society Agency, Republic of Korea, Apolitical's Content Partner. Sourced from Use Cases of Public AI Service, Vol.1,  https://eng.nia.or.kr/site/nia_eng/ex/bbs/View.do?cbIdx=31975&bcIdx=28997&parentSeq=29041