Researchers at the Korea Institute of Geoscience and Mineral Resources have developed technology to predict the possibility of landslides following forest fires.
Kim Min-seok, head of the Landslide Research Center at the Korea Institute of Geoscience and Mineral Resources, noted that the research team developed technology to assess risk levels within two and a half hours after extreme rainfall and predict debris flow disasters following landslides, allowing for essential evacuation time.
The roots of trees in the forest serve as stakes and nets while functionally preventing landslides. However, when trees die due to forest fires, the possibility of landslides increases when heavy rain causes groundwater to rise within the soil, saturating it. This is also true in cases where trees die from other factors, like the recent large-scale forest fire.
The research team developed a physically-based landslide prediction model linked from one-dimensional to three-dimensional based on the Korea Meteorological Administration's ultra-short-term forecast data, the localized forecast model (LDAPS). When applied to landslides in the Yecheon area and those near the Bulguksa temple in Gyeongju that occurred in 2023, the prediction accuracy reached over approximately 85%. By incorporating soil characteristics, woody debris, and rock movement in forest areas, the research team increased the accuracy to 90%.
Kim Min-seok stated, "The development of landslide risk prediction technology is crucial for preparing for landslide and debris flow disasters caused by extreme summer rainfall following large forest fires. We will continue to research and develop more accurate and effective landslide response technologies to advance them into world-class disaster response technologies."