Integrating AI and earth observation data for disaster risk reduction

This editorial introduces a special issue of 25 research articles that examine how artificial intelligence (AI) and Earth observation (EO) data are reshaping disaster risk reduction across the full disaster management cycle. The collection spans floods and coastal hazards, landslides and ground instability, drought, water scarcity and land degradation, wildfires, glacial lake outburst floods, and multi hazard applications, with case studies from the Indian subcontinent and from Vietnam, Türkiye, Indonesia, China and southern Africa.

The authors identify a clear methodological convergence toward machine learning and deep learning, with nearly half of the contributions applying ensemble learners or neural architectures, while the most convincing studies couple data-driven models with hydrological, geomorphological or ecological reasoning. Cloud platforms such as Google Earth Engine, multi-sensor fusion of SAR, optical, rainfall and gravimetric data, and a shift from hazard mapping toward risk assessment that incorporates exposure and socioeconomic vulnerability are highlighted as defining trends. Digital twins and near real-time monitoring platforms illustrate growing operational intent.