Data and information management

The study analyzes the evolution of DRR policies in Brazil and examines how these developments are reflected in disaster outcomes.

This publication explores how generative language models combine social media, news and weather data to detect disasters quickly and accurately, using satellite data to validate results from flood and wildfire case studies.

ML models for hydrological forecasting, even when given perfect inputs, cannot yet match the reliability of experienced human forecasters operating real-world prediction systems.

The authors call for expanded detection and attribution studies to better quantify how specific climate shifts drive health outcomes, and to translate those findings into actionable public health guidance.

This study developed an integrated urban flood risk assessment framework for cities with significant spatial heterogeneity, combining differentiated hydrological-hydrodynamic modeling with complementary grid-based.

Early warning is most effective when governance, data, and protocols align, helping health systems act earlier on climate-sensitive risks.

By improving their weather data, bolstering emergency response, and sharing resources, counties in Southwest Virginia can efficiently prepare for future floods.

Researchers from the ETH have now introduced a new artificial intelligence (AI) model that has learned these interactions and feedback autonomously and, compared to previous AI models, more precisely captures how air, land and water interact on Earth.