· 3 min read
illuminem summarises for you the essential news of the day. Read the full piece on Euronews or enjoy below:
🗞️ Driving the news: Scientists at the University of Cambridge are training artificial intelligence to rapidly detect and eventually predict landslides, a growing threat in climate-vulnerable regions
• Following Taiwan’s 7.4-magnitude earthquake in April 2024, their AI model identified over 7,000 landslides within three hours — a process that would have taken days manually
• This advancement could drastically improve emergency response during cascading disasters
🔭 The context: Landslides are increasingly triggered by extreme rainfall, earthquakes, deforestation, and warming temperatures, especially in mountainous areas
• Traditionally, responders rely on optical satellite imagery, but this can be obstructed by cloud cover or darkness
• The Cambridge team’s model integrates radar data, which can penetrate such obstacles, enabling faster, more accurate assessments in real time
🌍 Why it matters for the planet: As Europe and other regions face intensifying climate impacts — including glacier collapses, flash floods, and unstable terrain — early warning systems are vital
• AI-supported tools offer the potential to save lives, especially in remote or underserved areas, by rapidly identifying danger zones and directing aid more efficiently
• With climate risks growing, scalable AI systems could become essential components of climate resilience and disaster preparedness
⏭️ What's next: The research team, led by Dr. Lorenzo Nava and Prof. Maximillian Van Wyk de Vries, is now piloting an AI-supported early warning system in Butwal, Nepal, in collaboration with local scientists
• They are also working with the European Space Agency and the World Meteorological Organization to refine detection models and improve transparency
• A data science challenge has been launched to ensure AI results are explainable and trustworthy for decision-makers
💬 One quote: “In high-stakes scenarios like disaster response, trust in AI-generated results is crucial.” – Dr. Lorenzo Nava, University of Cambridge
📈 One stat: The Cambridge model detected 7,000+ landslides in under three hours following Taiwan’s 2024 earthquake — a process that previously required several days using manual methods
Explore carbon credit purchases, total emissions, and climate targets of thousands of companies on Data Hub™ — the first platform designed to help sustainability providers generate sales leads!
Click for more news covering the latest on climate change adaptation and green tech
