As climate change increases the frequency and severity of natural disasters, predicting snow avalanches and reducing associated risks have become critically important. A team of researchers led by Enes Can Kayhan and Assoc. Prof. Dr. Ömer Ekmekçioğlu from Istanbul Technical University has developed an innovative hybrid machine learning framework to generate highly accurate snow avalanche susceptibility maps in the French Alps. By combining different meta-heuristic optimization and machine learning techniques, this study presents a groundbreaking approach to identifying high-risk areas, aiming to enhance disaster preparedness and public safety.
Using real satellite and meteorological data, the research mapped 14 massif regions in the French Alps, classifying risk levels into very low, low, moderate, high, and very high. The findings are of great significance for disaster risk management, policy-making, and emergency response planning. The researchers emphasize that this advanced framework can be extended to other avalanche-prone areas, further improving prediction capabilities through real-time data processing integration.
Future studies aim to enhance AI-supported risk models by integrating more environmental variables, increasing algorithmic efficiency, and evaluating industrial-scale applications. As machine learning continues to transform natural disaster prediction, this research paves the way for developing more accurate, efficient, and applicable snow avalanche forecasting tools.
Access the full article here: [DOI: 10.3390/w16223247]
