Sign in Subscribe
Global warming

Advances in Climate Change Prediction: New Study from Chosun University

Strategic Insights

— 1 min read

A new study on climate change prediction has shed light on the need for accurate models and systematic preparedness to mitigate the effects of extreme weather events. Funded by the National Research Foundation of Korea and the Korea Government, research from Chosun University in Gwangju, South Korea, utilized machine learning algorithms to predict multiple natural hazards, including droughts, floods, and wildfires. The study found that the Extreme Gradient Boosting (XGB) algorithm performed exceptionally well in predicting droughts and floods, while the Random Forest (RF) algorithm excelled in predicting wildfires.

Key Takeaways:

  • The study divided South Korea into a grid system with a 0.01° resolution, using meteorological, climatic, topographical, and remotely sensed data to analyze each grid cell.
  • The research focused on three major natural hazards: drought, flood, and wildfire, using two machine learning algorithms: Random Forest (RF) and Extreme Gradient Boosting (XGB).
  • The analysis showed that XGB achieved ROC scores of 0.9998 and 0.9999 in predicting droughts and floods, respectively, while RF achieved a high ROC score of 0.9583 in predicting wildfires.
  • The study provides foundational data for the development of hazard management and response strategies in the context of climate change.
  • The research offers a basis for future studies exploring the interaction effects of multi-hazards.
  • The study suggests that machine learning algorithms can be used to improve climate change prediction and mitigate the effects of extreme weather events.

Statistics:

  • South Korea was divided into a grid system with a 0.01° resolution.
  • The study used meteorological, climatic, topographical, and remotely sensed data to analyze each grid cell.
  • The ROC scores for XGB in predicting droughts and floods were 0.9998 and 0.9999, respectively.
  • The ROC score for RF in predicting wildfires was 0.9583.
  • The study focused on three major natural hazards: drought, flood, and wildfire.

Sources:

  • Multi-Hazard Susceptibility Mapping Using Machine Learning Approaches: A Case Study of South Korea. Remote Sensing, 2025,17(10):1660.
  • http://www.mdpi.com/journal/remotesensing
  • Remote Sensing, published by MDPI AG.

Read more