AI-DRIVEN PREDICTIVE ANALYTICS FOR LARGE-SCALE CLIMATE RISK MANAGEMENT

• Dr. Elena Moravik

  Department of Environmental Informatics University of Ljubljana Slovenia

  Author

Climate risk management requires advanced analytical capabilities to predict extreme weather, assess vulnerability, and support climate-resilient decision-making. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), provides scalable tools that outperform traditional statistical models in handling nonlinear climate behavior, large datasets, and spatiotemporal variability. This study examines AI-driven predictive frameworks for large-scale climate risk assessment using integrated models combining remote sensing, climate simulations, socioeconomic indicators, and hazard-specific datasets. It presents a multi-layered architecture for climate risk forecasting, evaluates model performance using multi-decade data, and highlights applications in flood prediction, drought monitoring, cyclone risk, wildfire forecasting, and infrastructure vulnerability. The findings show that AI-based predictive systems significantly enhance lead-time accuracy, reduce false alarms, and support data-driven climate adaptation strategies. Policy implications and future research directions are also discussed.

• pdf

This work is licensed under a Creative Commons Attribution 4.0 International License.